Loughborough University

Institutional Repository

Alternative, more
sustainable, wall
construction techniques than
brick and block, for new
housing in England and
Wales

This item was submitted to Loughborough University's Institutional Repository

by the/an author.

Additional Information:

• A Doctoral Thesis. Submitted in partial fulfilment of the requirements

for the award of Doctor of Philosophy of Loughborough University.

Metadata Record: https://dspace.lboro.ac.uk/2134/12375

Publisher: c Fiona Hamilton-MacLaren

Please cite the published version.

This item was submitted to Loughborough University as a PhD thesis by the
author and is made available in the Institutional Repository
(https://dspace.lboro.ac.uk/) under the following Creative Commons Licence
conditions.

For the full text of this licence, please go to:

http://creativecommons.org/licenses/by-nc-nd/2.5/
 Alternative, more sustainable, wall construction
techniques than brick and block, for new housing in
England and Wales

by

Fiona Hamilton-MacLaren

Doctoral Thesis

Submitted in partial fulfilment of the requirements

for the award of
Doctor of Philosophy of Loughborough University

May 2013

© by Fiona Hamilton-MacLaren 2013

Acknowledgements

Many thanks go to my family and friends for their support; to my supervisors Professor
Dennis Loveday and Doctor Monjur Mourshed. Many thanks also to all those who
responded to the questionnaires that formed an essential part of this work.

i
Abstract

There is a need to reduce the emissions of the country as a whole, to limit the risk of
climate change due to Global warming and to meet targets set by the Kyoto agreement
and the Climate Change Act. The large number of houses constructed annually in
England and Wales have an important role to play in this. By reducing emissions,
resulting from both the manufacture of construction materials and the energy used by
house occupants, housing can help achieve the necessary emissions reductions.
Alternative construction methods can contribute to this, either by having a lower
embodied energy or by demonstrating good thermal properties to limit heat loss and
hence operational energy. However, it is essential that both the construction industry
and the public accept the alternative construction methods for them to be economically
viable. In addition, there should be no loss of performance as a result of using
alternative construction methods.

Six methods of construction were studied in depth, including generating embodied and
operational energy requirements and identifying their performance in terms of
airtightness, wall thickness, and fire resistance. Public and industry acceptability were
examined by use of questionnaires. A comparison of the data collected showed that
identifying the best, or optimal, option visually is a challenging task as no single
method of construction is best in all areas. A methodology was created to aid the
selection of a wall construction method. The methodology is capable of examining
multiple variables, in this work it is demonstrated with construction method and front
building dimension. To identify the optimal method, optimisation by genetic algorithms
is used. Use of the methodology was demonstrated with a case study based on the
most frequently constructed housing type for England and Wales. The importance of
weighting was demonstrated with the use of weightings based on concerns held by
different parties. It was found that minimising the external wall area gives the optimal
solution as less material is needed and there is less opportunity for heat loss. For the
situation examined in the case study, Structural Insulated Panels (SIPs) were identified
as having the potential to reduce the environmental impact of housing construction in
England and Wales without impacting saleability or performance.

ii
Contents

1 Introduction ................................................................................................. 1
1.1 Background and motivation .............................................................. 1
1.1.1 Climate change ................................................................... 1
1.1.2 House building ..................................................................... 2
1.1.3 Use of alternative construction methods ............................. 3
1.1.4 User views ........................................................................... 4
1.1.5 Selection of construction methods ...................................... 5
1.2 Work carried out ................................................................................ 5
1.3 Aims and objectives .......................................................................... 8
1.4 Limitations ......................................................................................... 8
1.5 Thesis structure ................................................................................ 9

2 Literature review ......................................................................................... 11

2.1 Introduction ....................................................................................... 11
2.2 Housing construction in England and Wales .................................... 11
2.2.1 Construction figures ............................................................ 11
2.2.2 Construction type ................................................................ 12
2.2.3 Sustainability in construction ............................................... 12
2.3 Using alternative methods of construction ........................................ 14
2.3.1 Available methods ............................................................... 14
2.3.2 Innovation ............................................................................ 17
2.3.3 Operational energy .............................................................. 18
2.3.4 Embodied energy ................................................................ 19
2.3.5 Performance ........................................................................ 20
2.4 Decision making in construction ....................................................... 21
2.4.1 Assessment methods .......................................................... 21
2.4.2 Construction method selection ............................................ 23
2.4.3 Optimisation by use of genetic algorithms ........................... 29
2.5 The level of user acceptability in construction decision making ........ 32
2.6 Contribution to knowledge ................................................................ 34
2.7 Summary ........................................................................................... 34
3 Methodology ................................................................................................. 37
3.1 Introduction ....................................................................................... 37
3.2 Construction method selection ............................................................. 37
3.3 Construction method details ................................................................ 37
3.3.1 U-value ................................................................................. 38
3.3.2 Embodied energy ................................................................... 39
3.3.3 Air tightness, wall thickness and fire performance ................ 41
3.4 Acceptability of construction methods .................................................. 41
3.4.1 Data collection method .......................................................... 41
3.4.2 Questionnaire design ............................................................ 42
3.4.3 Population and sample .......................................................... 45
3.4.4 Questionnaire distribution ...................................................... 46
3.4.5 Response analysis ................................................................ 47
3.5 Optimisation based selection methodology ......................................... 48
3.5.1 Problem definition .................................................................. 48

iii
 3.5.2 The use of optimisation ......................................................... 49
3.5.3 Optimisation method ............................................................. 49
3.6 Case study development ................................................................... 50
3.6.1 Case study design ................................................................. 50
3.6.2 Optimisation variables ........................................................... 52
3.6.3 Mutation and crossover probabilities ..................................... 53
3.6.4 Number of generations .......................................................... 54
3.6.5 Population .............................................................................. 55
3.7 Summary ........................................................................................... 56
4 Construction methods ................................................................................... 57
4.1 Introduction ....................................................................................... 57
4.2 Selected methods ............................................................................. 57
4.3 Discarded methods ........................................................................... 58
4.4 Brick and block .................................................................................. 59
4.4.1 Construction details ............................................................. 59
4.4.2 U value calculation .............................................................. 61
4.4.3 Embodied energy ................................................................ 61
4.4.4 Airtightness, wall thickness and fire performance ................ 63
4.5 Structural Insulated Panels (SIPs) .................................................... 64
4.5.1 Construction details ............................................................. 64
4.5.2 U value calculation .............................................................. 66
4.5.3 Embodied energy ................................................................ 67
4.5.4 Airtightness, wall thickness and fire performance ................ 68
4.6 Insulating Concrete Formwork (ICF) ................................................. 69
4.6.1 Construction details ............................................................. 69
4.6.2 U value calculation .............................................................. 71
4.6.3 Embodied energy ................................................................ 72
4.6.4 Airtightness, wall thickness and fire performance ................ 74
4.7 Prefabricated straw bale and lime plaster panels ............................. 75
4.7.1 Construction details ............................................................. 75
4.7.2 U value calculation .............................................................. 76
4.7.3 Embodied energy ................................................................ 76
4.7.4 Airtightness, wall thickness and fire performance ................ 78
4.8 Thin joint block work ......................................................................... 79
4.8.1 Construction details ............................................................. 79
4.8.2 U value calculation .............................................................. 80
4.8.3 Embodied energy ................................................................ 81
4.8.4 Airtightness, wall thickness and fire performance ................ 82
4.9 Timber frame with brick cladding ...................................................... 83
4.9.1 Construction details ............................................................. 83
4.9.2 U value calculation .............................................................. 85
4.9.3 Embodied energy ................................................................ 86
4.9.4 Airtightness, wall thickness and fire performance ................ 88
4.10 Summary ........................................................................................... 89

5 Public acceptability of construction methods ............................................... 91

5.1 Introduction ....................................................................................... 91
5.2 Questionnaire results and discussion ............................................... 91

iv
 5.2.1 Response rate ..................................................................... 91
5.2.2 Geographical representativeness of responses .................. 91
5.2.3 Sampling of data ................................................................. 92
5.2.4 Importance of characteristics in house purchase ................ 94
5.2.5 Awareness of alternative construction methods .................. 103
5.2.6 Acceptability of Brick and block ........................................... 104
5.2.7 Acceptability of Structural Insulated Panels (SIPs) ............. 107
5.2.8 Acceptability of Insulating Concrete Formwork (ICF) .......... 109
5.2.9 Acceptability of Prefabricated straw bale and lime plaster 111
panels
5.2.10 Acceptability of Thin joint block work ................................... 115
5.2.11 Acceptability of Timber frame with brick cladding ............... 117
5.2.12 Method of increasing interest in alternative construction ..... 119
methods
5.3 Summary ........................................................................................... 121

6 House builder acceptability of construction methods .................................. 122

6.1 Introduction ....................................................................................... 122
6.2 Questionnaire results ........................................................................ 122
6.2.1 Response rate ..................................................................... 122
6.2.2 Representativeness of responses ....................................... 122
6.2.3 Importance of characteristics in construction method .......... 122
selection
6.2.4 Awareness of construction techniques ................................ 124
6.2.5 Past experience of construction methods ........................... 125
6.2.6 Acceptability of alternative construction methods ............... 127
6.2.7 Methods of increasing interest in alternative construction 132
methods
6.2.8 Comparison of survey results .............................................. 132
6.3 Summary ........................................................................................... 133

7 Development of the optimisation based methodology .................................... 134

7.1 Design of the fitness function ............................................................ 134
7.1.1 Input variables ..................................................................... 134
7.1.2 Constraints .......................................................................... 135
7.1.3 Wall dimensions .................................................................. 136
7.1.4 Embodied energy ................................................................ 138
7.1.5 Operational energy .............................................................. 139
7.1.6 The single objective calculation ........................................... 140
7.2 Optimisation procedure ..................................................................... 143
7.2.1 Optimisation variables ......................................................... 143
7.3 Use of the decision making methodology ............................................ 144
7.4 Summary ........................................................................................... 145

8 Case study ................................................................................................. 147

8.1 Introduction ....................................................................................... 147
8.2 Case study optimisation results ........................................................ 147
8.2.1 Solution metrics for equal weighting case study .................. 147
8.2.2 Fitness landscapes .............................................................. 148

v
 8.2.3 Phi array visualisation ......................................................... 151
8.2.4 Non-equal weightings .......................................................... 153
8.3 Comparison of results with brick and block construction .................. 156
8.3.1 Equal weighting case study ................................................. 156
8.3.2 Public weighting case study ................................................ 157
8.3.3 Construction industry weighting case study ........................ 158
8.4 Case study recommendations .......................................................... 159
8.5 Summary ........................................................................................... 160

9 Conclusions ................................................................................................. 162

9.1 Findings from the work ...................................................................... 162
9.1.1 Embodied energy of construction methods ......................... 162
9.1.2 Acceptability of alternative methods of construction ........... 163
9.1.3 Areas of concern relating to housing .................................. 163
9.1.4 Increasing acceptability ....................................................... 164
9.1.5 The optimisation based design methodology ......................... 164
9.1.6 Case study solution ............................................................. 165
9.2 Conclusions and recommendations .................................................. 167
9.3 Contribution to knowledge ................................................................ 168
9.4 Further work ...................................................................................... 170

References .................................................................................................. 172

Appendices
A- Sample survey, house holder .................................................................. 185
B- Sample survey, house builder .................................................................. 198
C- Acceptability of construction methods ................................................... 210
D- Degree Day data ..................................................................................... 215
E- The typical house .................................................................................... 217
F- Fitness function ...................................................................................... 219

List of tables

Table 2.1 Number of houses completed in England and Wales from 11

2000 to 2011 (DCLG 2012a)
Table 2.2 Visual representation of using a genetic algorithm to create a 30
second generation
Table 3.1 Surface and airspace R values used in U value calculations 39
Table 3.2 Number of residences constructed in England and Wales 50
during 2006 by type of housing (NHBC 2007)
Table 3.3 Percentage of all housing constructed in England and Wales 51
during 2009/10 by number of bedrooms (DCLG 2012b and
Rees 2011)
Table 3.4 Floor areas by number of bedrooms for the UK, values from 51
Scott Wilson Ltd. (2010)
Table 3.5 Constant values for case study optimisation 52
Table 3.6 Variables and bounds for case study optimisation 52

vi
Table 3.7 Number of generations to complete the optimisation run 55
compared to maximum number of generations.
Table 3.8 Impact of varying population size on the optimisation 56
methodology results.
Table 4.1 Material specification for brick and block construction 60
Table 4.2 L, λ and R values for brick and block components 61
Table 4.3 Material quantities and embodied energy coefficients for1m2 62
brick and block work external wall
Table 4.4 Material quantities and embodied energy coefficients for 1m2 63
brick and block work party wall
Table 4.5 Material specification for SIP construction 65
Table 4.6 L, λ and R values for SIP components 66
Table 4.7 Material quantities and embodied energy coefficients for 1m2 67
SIP external wall
Table 4.8 Material quantities and embodied energy coefficients for 1m2 68
SIP party wall
Table 4.9 Material specification for ICF construction 71
Table 4.10 Values of L, λ and R used to calculate the U value for the ICF 71
wall system
Table 4.11 Material quantities and embodied energy coefficients for 1m2 73
ICF external wall
Table 4.12 Material quantities and embodied energy coefficients for 1m2 73
ICF party wall
Table 4.13 Material specification for prefabricated bale wall construction 76
Table 4.14 Material quantities and embodied energy coefficients for 1m2 77
prefabricated bale wall construction
Table 4.15 Material specification for thin joint block work construction 80
Table 4.16 L, λ and R values for thin joint block work components 80
Table 4.17 Material quantities and embodied energy coefficients for 1m2 81
thin joint block work external wall
Table 4.18 Material quantities and embodied energy coefficients for 1m2 82
thin joint block work party wall
Table 4.19 Material specification for timber frame construction 85
Table 4.20 L, λ and R values for timber frame construction components 86
Table 4.21 Material quantities and embodied energy coefficients for 1m2 87
timber framed external wall
Table 4.22 Material quantities and embodied energy coefficients for 1m2 88
timber framed external wall
Table 4.23 U values, embodied energy and performance values for each of 90
the methods of construction and Building Regulations
Table 5.1 Number of responses taken from each region to form the 93
sample
Table 5.2 Modal and median values of importance for the 95
characteristics considered
Table 5.3 Values for Bartlett’s test of spherecity for each sample 96
Table 5.4 Rotated component matrix for Sample 1 Likert scale principal 98
component analysis
Table 5.5 Rotated component matrix for Sample 2 Likert scale principal 99
component analysis
Table 5.6 Rotated component matrix for Sample 3 Likert scale principal 100
component analysis
Table 5.7 Rotated component matrix for Sample 4 Likert scale principal 101
component analysis

vii
Table 5.8 Rotated component matrix for Sample 5 Likert scale principal 102
component analysis
Table 6.1 Modal and median scores for the importance of each 123
characteristic
Table 7.1 Default values for variables in the single objective equation 135
Table 7.2 Benchmark values for embodied energy, divided by house 141
type.
Table 7.3 Benchmark values for operational energy, divided by region 141
and type of house.
Table 7.4 Benchmark values for performance factors 142
Table 8.1 Solution metrics for the optimisation of the case study 147
Table 8.2 Weighting focusing on the views of the public 153
Table 8.3 Weighting focusing on construction industry views 155
Table 8.4 A comparison of prefabricated straw bale construction with the 157
“typical” method used in England and Wales for the case study
house
Table 8.5 A comparison of the SIP construction method with the 158
“typical” method used in England and Wales

List of figures

Figure 3.1 Crossover fraction against best fitness value for the case 53
study runs.
Figure 4.1 Axonometric view of brick and block construction 60
Figure 4.2 Axonometric view of SIP construction 65
Figure 4.3 Axonometric view of ICF construction 70
Figure 4.4 Axonometric view of prefabricated bale wall construction 75
Figure 4.5 Axonometric view of thin joint block work construction 79
Figure 4.6 Axonometric view of timber framed construction 84
Figure 5.1 Percentage distribution of respondents and population of 92
England and Wales by region
Figure 5.2 The importance of each factor when considering a house 94
purchase, from Sample 1.
Figure 5.3 Percentage of respondents who have heard of each type 103
of construction
Figure 5.4 Acceptability of brick and block construction by region 105
Figure 5.5 Acceptability of SIP construction by region 107
Figure 5.6 Acceptability of ICF construction by region 109
Figure 5.7 Acceptability of prefabricated straw bale construction by 111
region
Figure 5.8 Acceptability of thin joint block work construction by region 115
Figure 5.9 Acceptability of timber frame construction by region 117
Figure 6.1 Percentage scores for the importance of each 123
characteristic
Figure 6.2 Past usage of the construction methods by respondents’ 125
companies
Figure 6.3 Percentage responses to the question “Would you use 127
this method?” for each option
Figure 7.1 Dimension designations for single objective calculation 134
Figure 8.1 Three dimensional representation of the fitness landscape 149

viii
Figure 8.2 Two dimensional graphical display of the fitness landscape for 150
case study optimisation with equal weighting
Figure 8.3 Phi array for single objective function for case study, 152
equal weightings.
Figure 8.4 Two dimensional graphical display of the fitness landscape for 154
case study optimisation with public-focused weightings
Figure 8.5 Two dimensional graphical display of the fitness landscape for 155
case study optimisation with construction industry-focused
weightings

ix
1. Introduction
1.1. Background and motivation

The need to reduce energy use and the associated carbon emissions on a global scale
is widely known. In England and Wales, over 100,000 houses are built every year
(DCLG 2012a).This large number of houses contributes significantly to the energy
emissions of the countries. Energy used to manufacture materials for construction has
an immediate effect on emissions; the performance of the buildings has an impact on
the emissions far into the future across the lifespan of the building. Reducing these
energy requirements has the potential to play a significant role in reducing the carbon
emissions associated with energy generation for England and Wales.

The use of alternative methods of construction, moving away from the traditional
method of brick and block, may offer a solution to this. This may be as a result of
choosing methods which require less energy to manufacture, or ones which
demonstrate better thermal efficiency, resulting in reduced heat loss and hence
reducing the energy required for heating and its associated emissions. However, it is
necessary to maintain other aspects of performance as required by the Building
Regulations (DCLG 2013a). In addition, houses that are constructed must be
acceptable to purchasers and the construction industry. If there is no market for
houses built using alternative methods they may result in a financial loss and hence
would not have the support of the construction industry. The construction industry must
be prepared to use and invest in an alternative method of construction for it to achieve
wide spread use in housing.

Considering all these aspects presents a challenge. A large number of factors must be
evaluated and there is rarely a single solution that presents a clear best option (Tam et
al. 2007). Some level of compromise will always be necessary. In this work
optimisation is used as the basis for a selection method, identifying the best possible
compromise for a given situation.

1.1.1. Climate change

Greenhouse gas emissions are widely acknowledged to have a negative impact on the
environment, in particular, resulting in changes to climates across the World as a result
of “global warming”. The mechanism believed to be responsible for this is the

1
increased quantities of greenhouse gases in the upper atmosphere reflecting heat
back onto the Earth’s surface, rather than allowing its emission out in to space. The
Intergovernmental Panel on Climate Change (IPCC) report on global warming and
climate change found as high as a “90% confidence that human activity was
responsible for global warming” (IPCC 2007). The result of this is a net increase in the
global temperature. Increasing temperatures are predicted to result in widespread
changes to the environment, including changes to regional climates, melting of the
polar ice caps and increasing sea levels. All of these have the potential to negatively
affect populations and ecology on a global scale.

In order to limit the potential for global warming, and the resulting climate change,
emissions of the greenhouse gases responsible need to be reduced. Internationally
this was agreed by many countries at the Kyoto summit in 1998, which resulted in the
Kyoto Protocol (United Nations 1998). The Kyoto Protocol committed the participating
countries to achieve set reductions or stay below a maximum increase in greenhouse
gas emissions by the end of 2012. In the case of the EU the target to be achieved was
an 8% reduction below 1990 levels (United Nations 1998). The emissions reduction
requirements were shared among the member states by the EU. As a result of this the
UK was been set a target of a 12.5% decrease in emissions below 1990 levels
(Department of Energy and Climate Change, 2012). This is an increase on the 8%
required for the UK independently by the Kyoto Protocol. A number of summits have
been held in an attempt to set ongoing targets for emissions reductions once the Kyoto
Protocol deadline has passed. These have taken place in Bali (December 2007),
Poznan (December 2008), Copenhagen (2009), Cancun (2010) and Durban (2011).
However no further agreement has been reached.

To achieve the emission reduction targets set for the UK as a result of the Kyoto
Protocol, the Climate Change Act came into force in 2008. This sets a legally binding
requirement for the UK to reduce carbon emissions to 80% below 1990 levels by 2050
(Climate Change Act 2008). This is an ambitious target, to achieve it all areas of life
will have role to play. In England and Wales, Part L of the Building Regulations
incorporates emissions restrictions to help achieve the required emissions reductions
(DCLG 2013b), see Chapter 1.1.2.

1.1.2. House building

2
House building in England and Wales is controlled by a number of Governing bodies,
including Planning and Building Regulations. Although there has been a dip in the
number of houses constructed annually as a result of the recession, 113,400 dwellings
were completed in England and Wales during 2010-2011 (DCLG 2012a). It can be
seen that even a small saving in the energy and associated emissions on each
property could result in a significant saving each year. This saving would help with
achieving the goals of the Climate Change Act (2008).

Currently the levels of emissions from operational energy, that is, energy used for
heating, lighting, hot water and building services is controlled by the Building
Regulations via the Code for Sustainable Homes. By 2016 all new houses built should
achieve a Level 6 in the Code for Sustainable Homes which requires them to meet the
zero carbon standard (Department for Communities and Local Government 2007a). To
achieve this, the annual net value of heating, lighting, hot water and building services
energy requirements must be zero. The Code for Sustainable Homes also gives
consideration to other aspects of the construction, such as waste management and
materials, however energy use accounts for much of the score achieved.

1.1.3. Use of alternative construction methods

The typical method of housing construction for England and Wales is a double skinned
brick and block construction, with an air cavity between the leaves; it was used for 88%
of housing In England from 1990 to March 2009 (DCLG 2010). There is evidence of
other construction methods being used, in particular, timber frame, which accounted
for 7% of houses constructed in England from 1990 to March 2009 (DCLG 2010).
Examples of alternative construction include those presented by The Insulating
Concrete Formwork Association who provide information on the use of Insulating
Concrete Formwork (2009); Morley (2000) gives details of using Structural Insulating
Panels (SIPs) and Jones (2007) who discusses straw bale construction. The use of a
wide range of materials in construction, their benefits and disadvantages is discussed
by Calkins (2009). The authors who discuss alternative methods of construction
present their advantages, and drawbacks. However, despite the advantages
presented, alternative methods of construction are rarely used, “other” methods
making up only 5% of construction in England from 1990 to March 2009 (DCLG 2010).

Timber framed construction went through a period of greater use in the nineteen sixties
and seventies. However, the popularity fell drastically when a 1983 television report

3
stated that timber framed construction was prone to rotting and fire (Cavill 1999).
Cavill (1999) stated that the use of timber framed construction was increasing in
England and Wales, with prefabricated timber construction of particular interest.

The use of a wide range of construction methods has been demonstrated by the
Building Research Establishment (BRE) at the BRE Innovation Park (BRE 2013).
These methods are predominantly innovative, with methods used which have little
history of use in the UK. This provides an opportunity to examine how these
construction methods perform in the UK climate. Construction methods such as
Structural Insulating Panels and Insulating Concrete formwork can be seen to be
moving from the innovative, research based, area of construction, becoming more
mainstream (The Insulating Concrete Formwork Association 2009).

International experience can also be considered when examining alternative

construction methods. Timber framed construction is widely used in the USA. Although
the variations in climate must be considered, the performance and lessons learned
from overseas examples can be applied to the UK situation.

Typically, innovative methods of construction have been for research purposes and, in
some cases, adopted by self-builders (Jones 2007). Identification of innovative
methods of construction that have the potential for mainstream use is necessary to
achieve the benefits they can offer in terms of enhanced performance.

1.1.4. User views

It was considered that the views of the public are often discounted when making
construction decisions. As they represent the end user and are the ones who will make
an investment in the finished property, their views are important when making
decisions. If a particular option does not have public support, selling the finished house
will be difficult. This makes the option undesirable to the construction industry as they
risk making minimal profit, or a loss. The acceptability of construction methods within
the construction industry must also be considered. Issues with saleability are not the
only factor that can reduce industry interest. Lack of knowledge or perceived difficulties
with the construction method would decrease its appeal. Resistance to change, both
from the public and the construction industry, was expected to be identified in this work
as one of the greatest challenges for alternative methods of construction.

4
1.1.5. Selection of construction methods

It is stated above that selecting the best option from a range of construction methods
can be challenging as desirable characteristics are often in conflict with each other
(Castro-Lacouture et al. 2009, Seo et al. 2004, Diakaki et al. 2010). Making a decision
between the available options, when there is often no one “best” solution can be a
challenge (Emmanuel 2004).

A number of methods exist for assessing a building design, ranging from scoring it
based on the Code for Sustainable homes to a full Lifecycle analysis (LCA) in which all
aspects of the construction throughout its life are considered. The Code for
Sustainable Homes considers options on a limited number of aspects. Those affecting
construction method choice are energy use and materials (DCLG 2007a). Methods
such as LCA can be time consuming and costly (Björklund 2002). A more effective
method that balances the wider range of aspects which need to be considered, with a
reasonable time and cost input is required.

1.2. Work carried out

In this work, the construction method is taken to be the combined materials used to
build the house wall, rather than the actual way in which it is constructed. “Construction
material” was discarded as a term as each method uses multiple materials. Brick and
block construction, with an insulated and air filled cavity is considered to be the typical
method. Alternative methods of construction are any other method, five alternatives
were examined in detail- Structural insulated panels, insulating concrete formwork,
prefabricated straw bale panels, thin joint block work, timber frame with brick cladding.

This work focused on two areas of the issue of using alternative methods of wall
construction as a way to reduce the environmental impacts of housing construction: the
acceptability of a range of alternative construction methods, and the use of
optimisation in a design methodology to assist with choosing the best method. Six
methods of construction were chosen for detailed study. These included brick and
block as a “control” and a range of alternative methods, including modern methods of
construction and a method considered highly unusual. The alternative construction
methods were Structural Insulated Panels (SIPs), Insulating Concrete Formwork (ICF),
prefabricated straw bale panels, thin joint block work and timber frame with brick

5
cladding. Methods were selected to maintain a brick outer appearance where possible
in order to avoid acceptability being affected by visual appearance.

To examine the awareness acceptability of the construction methods two surveys were
carried out. The acceptability was examined from the point of view of the public and the
construction industry. It was considered that both these parties must support an
alternative construction method for it to be considered a viable method for use. It was
found that awareness of alternative methods of construction is fairly high, particularly
within the construction industry. The acceptability of the construction methods
examined varied greatly. Brick and block had the highest acceptability; this was
expected as it the typical method used in the area of study. Good levels of acceptability
were also seen with SIPs, ICF and thin joint block work, indicating these have the
potential to be used in mainstream construction. Poor acceptability was seen for
prefabricated straw bale and timber frame construction, although some respondents
showed interest in their use. These methods are less likely to be suitable for
speculative construction as there would be a limited market and they would present an
economic risk for the builder. The acceptability scores for each construction method
were used in the optimisation based design methodology created later in the work.
Responses to the survey indicated a wide range of concerns about each construction
method. It was noted that many of these concerns have already been disproven by
existing research or anecdotal evidence. If these concerns could be addressed, the
acceptability of all the examined methods may be increased. Education is the key to
achieving this, and accessing the advantages alternative construction methods
present.

Views were also collected on factors which are important when considering a house
purchase or method to use in construction. Principal component analysis, which
combines the factors into groupings which demonstrate key issues, was used to
identify areas of public concern as environmental, financial, and risk. For the
construction industry the most important factor was thermal efficiency. The importance
of factors was used to create weightings for the case study which demonstrates the
use of the optimisation based methodology.

The optimisation based methodology created in this work was intended to demonstrate
the methodology of selecting the construction method and front building dimension
based on the best compromise of energy use, acceptability and performance criteria.
There is great potential for expanding the methodology using the methods
demonstrated to create a fully functional design methodology.

6
Each of the construction methods being examined was studied in depth. Data was
collected and generated regarding the embodied energy and performance of the walls
created, including U-value, wall thickness, airtightness, and fire performance. These
aspects were used as they demonstrate a range of criteria and can all be affected by
changing the materials used in wall construction. Particular consideration was given to
the embodied energy of the construction methods. It was noted that bricks had a major
impact on the total embodied energy of all methods that used them. Consideration was
given to how this impact could be reduced and the impact this may have on the
optimisation. Survey results indicated a strong preference for a brick external finish.
The need to balance the reduced embodied energy with the acceptability based on
appearance was identified as an important consideration. Brick slips were identified as
a possible solution.

An equation was generated which combines all the performance criteria with
weightings and normalisation to create a single value, representing the score for the
combination of variables being considered. Optimisation was then applied to this
equation, running it multiple times with different input values in order to find the best
solution. For this work the variables selected were front building dimension and the
construction method used. Other factors, such as floor area, house type and region
could also be used as variables.

To demonstrate the use of the optimisation based methodology a case study was
created. The values used in this were based on the most frequently constructed house
type and size in England and Wales. An initial run, with equal weightings for all factors
identified brick and block construction as optimal. Values for this optimisation run were
used to demonstrate how examining the fitness landscape by use of visual
representations of the solution space, for example two and three dimensional graphs,
can lead to the selection of optimal and sub-optimal solutions. Equal weightings were
not considered to be representative of a realistic scenario. Two sets of weightings were
generated based on survey responses, one focusing on the public interests and one on
the interests of the construction industry. The results of the public focused weighting
indicated that prefabricated straw bale construction was the best option, with SIP
construction being sub-optimal but scoring well. The industry focused weighting found
SIP construction to be the optimal solution. Based on these two sets of results and
consideration of the two construction methods, SIPs were identified as the best method
of construction for the given scenario and situation.

7
1.3. Aims and objectives

This thesis aims to explore the benefits that alternative methods of construction may
present and the issues with their use. Ways in which the use of alternative methods of
construction can be increased are considered. Increased use would allow the benefits
to be accessed.

The objectives of the work were:

 To examine the potential for alternative methods of construction to reduce the

operational and embodied energy associated with housing whilst maintaining
performance.
 To determine the current acceptability of alternative methods of construction to
house purchasers and the construction industry.
 To identify what might increase the acceptability of alternative methods of
construction.
 To develop an assessment method using optimisation to identify the optimum
construction method to use in a user defined scenario, this may be by
designers, builders, planners.
 To demonstrate the use of this assessment method with a case study.
 To consider the impact the case study solution could have on housing
construction in England and Wales.

1.4. Limitations

In order to complete the work a number of limitations were set.

 The area of study was restricted to England and Wales. At the time of this work
they were covered by a single set of Building Regulations. Scotland and
Northern Ireland were excluded as they are covered by different construction
regulations. However, these countries have many similarities; the design
methodology could be used with only minor alterations.
 The area of study was split into regions based on standard divisions. These
were East Anglia, East Midlands, London, North East, North West, South East,
South West, West Midlands, Yorkshire and Humberside and Wales.

8
  The number of methods of construction was set at six to allow time for the
depth of study required.
 Methods of construction focused on the walls only, other elements such as
roofing, windows and foundations were not included.
 A maximum height of three storeys was considered in this work. Some of the
construction methods considered in this work cannot be used at heights greater
than this, therefore it was set as a limitation.
 Only houses were considered in this work. Flats were not considered as they
typically have different construction methods and tend to be built to a greater
height than the three storeys considered here.

Although these limitations were placed on the work presented here, the optimisation
based methodology was designed so that it could be expanded to allow a wide range
of these aspects to be included. Greater numbers of construction methods, other
countries and regions, other construction elements and different building types could
all be accommodated using the methods presented here. It was intended to present a
methodology, rather than create a fully functional design tool.

1.5. Thesis structure

Literature review- A review of current literature covering this topic was carried out.
The results of this are discussed.

Construction methods- The selection and details of the six construction methods
used in the work are discussed. Consideration is also given to construction methods
not selected and the reasons why. Further details of the selected methods’
performance and embodied energy are given.

Public acceptability- Details of the survey carried out to determine user acceptability
are given. The results and analysis are discussed.

Construction industry acceptability- Details of the survey carried out to determine

the views of the construction industry on alternative methods of construction.
Consideration of the results of the survey and the implication for use of alternative
construction methods.

Creation of the optimisation based methodology- Data identified from the

Construction methods and Acceptability chapters was combined with optimisation

9
techniques to create a method of selecting the best construction method for a given
scenario.

Case study- To demonstrate the use of the optimisation programme a case study was
developed. This was based on the most frequently constructed housing type for
England and Wales. A range of weightings were used. Comparison of the embodied
energy, operational energy and performance of the suggested method with the
traditional method of construction is given.

Conclusions- Conclusions drawn from the work are discussed.

Further work- Potential areas for adjustment and expansion of the design
methodology are given.

References- References used during the work

Appendices- Further details on aspects of the work not included in the main text.

10
2. Literature review
2.1. Introduction

Aspects of existing literature considered to be relevant to this work were housing

construction in England and Wales, the use of alternative methods of construction,
decision making in construction and the role of acceptability in construction decision
making. Existing work on these elements was reviewed in relation to the themes of this
thesis.

2.2. Housing construction in England and Wales

2.2.1. Construction figures

The number of houses constructed in England and Wales has been seen to fall
recently, believed to be as a result of the recession. Table 2.1 shows the number of
houses constructed for the years between 2000 and 2011.

Table 2.1- Number of houses completed in England and Wales from 2000 to 2011
(DCLG 2012a).

Number of houses completed

Year England Wales Total
2000-01 129,870 8,270 138,140
2001-02 137,740 8,310 146,050
2002-03 143,960 8,300 152,260
2003-04 155,890 8,490 164,380
2004-05 163,400 8,260 171,660
2005-06 167,680 9,330 177,010
2006-07 170,610 8,660 179,270
2007-08 140,990 7,120 148,110
2008-09 119,910 6,170 126,080
2009-10 107,890 5,510 113,400
2010-11 118,190 5,580 123,770

It can be seen from Table 2.1 that there was a reduction in the number of houses
constructed between 2007 and 2011. However, the number of completions increased
from 2009-10 to 2010-11. Although the number of completions for 2010-11 was thirty
percent lower than the peak seen in 2005-06, there are still a large number of houses
being constructed. This means that even a small saving in energy requirements and
the associated emissions per housing unit could have a significant impact on the
overall emissions of England and Wales.

11
2.2.2. Construction type

New housing built in England and Wales is typically of the speculative type. Housing is
often built in large numbers on a single site. The design and layout of the buildings is
often replicated many times on a large site, with a small number of layouts being
repeated. The scale of speculative construction is discussed by Barlow (2000) and Roy
and Cochrane (1998). Much of the new housing construction in the UK is carried out by
a relatively small number of building firms, ten firms were responsible for 82.5% of
completions in 2008 (McMeeken 2009).

An important impact of the speculative nature of construction in England and Wales is

the need for the buildings to have widespread public appeal. Houses are typically
planned and built before the end user is identified (Barlow 2000, Roy and Cochrane
1998). In order to be economically viable the selection of housing construction method
must not negatively impact the potential for sale. This work gives greater focus to the
speculative method of construction by including acceptability in the decision
methodology discussed in Chapter 6. The decision methodology could also be used in
a one off design situation to identify the best compromise in terms of energy use and
performance; however acceptability would be based on the client’s preference.

Consideration was given to the typical aspect ratio of housing in England and Wales.
This is affected by the values of front building dimension, which was used as a variable
in the optimisation based methodology. Work which makes use of a typical house as a
case study, such as that presented by Monahan and Powell (2011); show an aspect
ratio in the region of 1:2. In this work the impact of altering the front building dimension,
and therefore the aspect ratio is considered. A dimension of 5.0m is recommended by
Chown (1999) as the minimum front dimension for easy internal layout design,
although it is suggested that dimensions as low as 3.5m are possible. The impact of
this on aspect ratio will vary with the floor area of the house.

2.2.3. Sustainability in construction

The impact of the construction industry on the environment is controlled to some

degree by the need to meet the requirements of the Code for Sustainable Homes. The
code is a set of mandatory and optional standards that are used to assess the
environmental impact of housing. Once assessed, the house is given a level to indicate

12
its performance. From 2013 all new build houses are required to meet Level 4
standards; by 2016 all new build homes must achieve Level 6 (DCLG 2008a).

The element of the Code for Sustainable Homes with the greatest impact is the
requirement for new build housing to meet the zero carbon standard from 2016, i.e. to
achieve Level 6 of the Code for Sustainable Homes. This requires the emissions
associated with heat and lighting, hot water and building services to meet levels which
drop progressively towards zero by 2016 (DCLG 2007a). In addition to representing
the greatest number of points in the Code for Sustainable Homes scoring system,
energy use also has a greater weighting than the other categories, giving it the
greatest significance. Other environmental impacts, such as water use, materials,
waste disposal and impact on ecology are also considered, with points available for
these which can improve the level achieved.

Several elements of the Code for Sustainable Homes potentially have an impact on
this work. Operational energy will be impacted by both the construction method used,
and the value of front dimension for semi-detached and terraced houses. The
construction method will affect the operational energy as a result of varying U-values
across the methods considered; a method with a low U-value will result in a better
score than one with a higher U-value. In the consideration of detached houses the front
dimension does not affect the total external wall area so will have no impact on energy
use. However, for semi-detached and terraced properties, increasing values of front
dimension will result in a greater value of external wall, and hence greater heat loss if
no changes to construction materials are made. This will increase the operational
energy requirements. The score given in the Code for Sustainable Homes for materials
also impacts the construction method selection process. Points are given for
responsible sourcing and the green guide rating. However, all materials considered in
this work achieve a rating of A or A+ so there is little differentiation between them
under the Code for Sustainable Homes (BRE Global Ltd. 2008). Additionally, many of
the aspects considered in this work, such as embodied energy, acceptability and wall
thickness have no impact on the score achievable. As such the Code for Sustainable
homes was discounted as the basis for optimisation work carried out in Chapter 7 of
this work.

The need to minimise the operational energy was taken as an important factor from the
Code for Sustainable Homes. In order to achieve the increasing level requirements as
time progresses it will be necessary to construct ever more efficient buildings.

13
Identifying construction options which contribute towards this will help in the design of
buildings able to meet the required level.

2.3. Using alternative methods of construction

2.3.1. Available methods

A wide range of construction methods and materials that can be used instead of brick
and block exist. Many of these methods have already been used to build houses in
England and Wales, for example, in England timber framed construction was used for
7% of the houses constructed from 1990 to March 2009 (DCLG 2010). “Other”
methods, were used for 5% of houses over the same time period. Brick and block and
Timber frame have a significant history of use in England and Wales for housing. In
addition there is support for their use, such as Accredited Construction Details (DCLG
2011) and Enhanced Construction Details (Energy Saving Trust 2011) which provide
guidance on meeting part L of the Building Regulations. They are currently only
available for brick and block, timber frame and steel frame.

Examples of other methods of construction that have been used in England and Wales
include straw bales, steel frame, prefabricated “pod” construction and rammed earth. It
is worth noting that the Approved Documents, which provide guidance for the Building
Regulations, do not prescribe any particular method, any method can be used,
provided the requirements of the Building Regulations are met (Tricker 2004).
Regulation 7 of the Building Regulations states that materials should be “adequate and
proper”, appropriate for the circumstances and installed properly (HM Government
2006). Any construction method that can satisfy all these requirements can potentially
be used for housing in England and Wales. Details are also given in Regulation 7 of
ways that a material can be shown to satisfy the requirements. These include meeting
British, European or other national standards; having technical approval, for example,
CE marking or approval by an independent certifier; testing and calculation to
demonstrate performance; past performance to the required standard or higher and
sampling and testing of the chosen materials by Building Regulations inspectors (HM
Government 2006).

The environmental impacts, both positive and negative, of a wide range of materials
are discussed by Calkins (2009). It can be seen from the work by Calkins (2009) that
material selection will always require a degree of compromise, all materials require

14
energy to produce and hence have an environmental cost from the outset. There will
be both benefits and drawbacks to any construction methods considered.

A number of authors such as Jones (2007), Morley (2000) and the Insulating Concrete
Formwork Association (2003) identify alternatives to brick and block construction and
discuss their benefits. Choosing between the options available is not usually
considered. Desirable characteristics are often in conflict with each other (Seo et al.
2004, Tam et al. 2007, Diakaki et al. 2010), so compromises must be made. Achieving
the best compromise requires the various properties and their relative importance to be
considered. This challenge of identifying the best compromise is explored in this work
with the use of optimisation.

Timber framed construction shows the greatest level of research and past examples of
the alternative methods of construction considered in this work. It is a method with a
long history of use and is widespread globally. Historically, England and Wales showed
an increase in timber framed housing construction during the nineteen sixties and
seventies. However, a television programme shown in 1983 claimed timber frames
were vulnerable to rotting and fire, this caused a significant loss of consumer
confidence, and resulted in a drop in timber framed housing (Cavill 1999). An increase
to the 7% of new build homes from 1990 to March 2009 (DCLG 2010) has been seen.
Cavill (1999) suggested that timber frame would be popular with builders because of
the time and cost advantages of preassembly, the potential for better thermal
performance and the potential to sell timber as “environmentally friendly”. However, the
demonstration of the impact customer acceptability can have is important to consider.
Mahaptra et al. (2012) give the main advantage of timber frame is given as the ability
to prefabricate.

Examples of alternative methods of construction in England can be seen at the BRE

Watford Innovation park (BRE Ltd. 2013). Buildings at the Innovation park have been
constructed to demonstrate ways in which high levels of sustainability can be achieved
in construction. This includes buildings constructed to meet Level six of the Code for
Sustainable Homes, and buildings constructed using a wide range of methods,
including framing, traditional methods, natural materials, modular construction, off-site
construction, structural insulated panels, (BRE Ltd. 2013). These examples provide
data and examples which can be used to encourage the use of these alternatives on a
wider scale within England and Wales.

Consideration should also be given to the international experience of housing

construction. Framed construction is widely used for housing globally, in particular in

15
the US and Japan (Roy et al. 2002) and Sweden (90% of single family houses
Mahaptra et al. 2012). Work by Mahaptra et al. (2012) examined the acceptability of
timber framed construction to the public and construction industry in the UK, Germany
and Sweden. Negative views of the method were identified in all countries from
construction professionals, and from the public in the UK and Germany. These
negative views were considered to be a barrier to increased use of timber framing.

There is much greater use of timber frame in Scotland; 29% of houses built from 1997
to 2002 were constructed with timber framed walls (Communities Scotland 2002).
Cavill (1999) reports that the fall in acceptability seen during the nineteen eighties in
England and Wales was not experienced in Scotland and suggests this may be a result
of better thermal performance from timber framed construction. This provides
examples of timber frame in a similar climate to that seen in England and Wales and
demonstrates the achievable performance.

Straw bale construction began in Nebraska in the US, as a process it has since been
altered and refined and is now used globally (Jones 2007). An important consideration
when comparing international examples is the impact of the climate experienced by the
England and Wales. The temperate climate experienced, which tends to be warm and
wet, has the potential to negatively impact on construction materials. Long term
evidence of performance is required (Jones 2007). An example of straw bale
construction in a mainstream use exists in the form of social housing constructed by
North Kesteven District Council (North Kesteven District Council 2012), who state it
was a cheaper construction method.

Modern methods of construction, such as Insulating Concrete Formwork (ICF), Thin

joint blockwork and Structural Insulated Panels (SIPs) can be seen as emerging forms
of construction. Much of the work that has been done on these construction methods
comes from international experience, such as work by Morley (2000), Denzer and
Hedges (2007) Johnston and Gibson (2008), Mosey et al. (2009). This existing body of
work based on international experience can be combined with examples in the UK to
provide information and evidence to encourage the use of these methods. Research
has also been carried out on these methods within the UK, for example work by Barista
(2008) and Bregulla and Enjily (2004). Additional UK support for these methods of
construction exists in the form of specialist associations, for example The Insulating
Concrete Formwork Association, who provide guidance on the use of ICF (ICFA 2009)
and the UKSIPS Association who provide guidance on the use of SIPs (UKSIPS 2012).

16
Morel et al. (2001) discuss the potential for local materials to reduce the environmental
impact of housing. This is demonstrated with the use of a French case study. A
complexity which can be seen with the use of local materials is that the available
materials will vary with location.

A key example of international housing construction with an environmental focus is the

development of the Passivhaus standard in Germany. The standard aims to minimise
the energy requirements of housing by the use of passive heating and cooling
methods, for example good thermal properties, airtightness and natural ventilation
(BRE Ltd. 2011). This method of construction is comparatively common in Germany,
with 20,000 units built by July 2012, however it has yet to be widely adopted in the UK,
with just 165 units by July 2012 (NHBC Foundation 2013). This is a set of design
criteria which can be combined with any method of wall construction, provided the
thermal performance and airtightness are sufficiently high.

2.3.2. Innovation

It has been shown that there is little use of alternative methods of housing construction
currently in England and Wales, with the typical methods being brick and block (88%),
timber frame (7%) and “other methods” used for 5% of new build houses (DCLG 2010).

The need for innovation to access alternative methods of construction is considered by

Mahaptra et al. (2012) in relation to timber frames. Methods such as demonstration
houses, education for both the construction industry and house purchasers, financing
of research and legislation are suggested as ways to increase the attractiveness of
alternative methods of construction. Of these methods, the main driver for innovation in
England and Wales may be the Code for Sustainable Homes. When the reluctance of
the construction industry to change is considered, as discussed by Ravetz (2008),
innovation may only occur when it is forced. The role of legislation should therefore be
considered. Support for innovation in relation to alternative methods of construction
could take the form of education, financial incentives and research to prove
performance. The creation of Accredited Construction Details (DCLG 2011) and
Enhanced Construction Details (Energy Saving Trust 2011) for alternative methods of
construction could also support their use. These provide guidance on meeting part L of
the Building Regulations and are currently only available for brick and block, timber
frame and steel frame; having this support for alternative methods of construction may
increase their acceptability and drive innovation.

17
The adoption of innovative methods is discussed by Emmitt and Yeomans (2008), with
the process being awareness of the innovation, gathering of information, making the
decision to adopt or reject the option, if is chosen, then it is implemented and reviewed,
with a decision made as to whether the option should be used again in the future. The
gathering of information is a key step here, which ties with previous comments on the
need for education and information to encourage innovation. Emmitt and Yeomans
(2008) and Mackinder (1980) both note the importance of time in designers adopting
innovative options. Time is required to research and fully understand innovation
solutions, this can discourage their use. Information on products must also be easily
accessible, Mackinder and Marvin (1982) note that designers are reluctant to find
written information from manufacturers regarding products.

Examples of innovation in construction can be seen in work carried out by the Building
Research Establishment (BRE). Research into materials and sustainable construction
is demonstrated at the BRE Watford Innovation Park, where sample houses have been
constructed to high sustainability standards using a range of alternative materials. BRE
support for innovation can also be seen in the promotion of the Passivhaus standard,
with guidance provided for achieving the Passivhaus standard (BRE Ltd. 2011). If
similar guidance were provided for alternative methods of construction, this may
support their adoption by the industry and public.

An example of innovation can be seen in North Kesteven District Council building

social housing using straw bale construction, this has led to Epping Forest District
Council planning a similar development (North Kesteven District Council 2012). This
demonstrates the need for one company, council or group of people to take the first
step, to make the innovation, and demonstrate the benefits which can be accessed.

2.3.3. Operational energy

The operational energy of a building is the energy required to run heating, lighting and
all other power requirements. Legislation requires that from 2016 all new build homes
are constructed to zero carbon standard (DCLG 2007a). In 2011 zero carbon was
defined as meeting all energy requirements for heating, fixed lighting, ventilation, hot
water and building services. Energy requirements associated with other activities, such
as running appliances, is not covered.

18
In order to reduce the heat loss from the building high standards of airtightness and U-
value should be aimed for. The selection of construction methods which perform well in
these areas will reduce the heat loss, and hence the energy required to heat the
building. This places a lower requirement on the energy generation and makes the
required construction standard more achievable. From this it was taken that the U-
values of alternative methods of construction should be considered and their impact on
operational energy requirements considered.

2.3.4. Embodied energy

The embodied energy of a material or system is the energy required for its
manufacture. The energy required for transportation, installation and end of life
disposal can also be considered to form part of the embodied energy. The embodied
energy associated with construction is not currently considered in the assessment of
housing in England and Wales; however, in recommendations made by the Low
Carbon Construction Innovation and Growth Team (2010) it is suggested that a
method should be agreed upon and implemented to allow for the whole life accounting
of carbon in projects. Ravetz (2008) estimates embodied energy is responsible for 27%
of the energy associated with housing. As the value of operational energy falls in line
with the Code for Sustainable Homes requirement, embodied energy will be
responsible for an ever increasing proportion of the lifetime energy cost. An additional
benefit of reducing the embodied energy of construction is that it does not rely on user
behaviour. By making changes to elements which do not rely on a specific type of
behaviour, such as the building fabric, the intended improvements can be maintained
and the effects of unpredictable behaviours limited.

A number of authors have discussed how the use of alternative methods of

construction can contribute to a reduced embodied energy. Buchannan and Honey
(1993) present an example comparing concrete and steel construction with an
equivalent timber building. They demonstrate, through the use of material quantities
combined with embodied energy coefficients, that the use of timber results in a building
with a lower embodied energy and hence reduced carbon emissions. Hammond and
Jones (2008) discuss the generation of embodied energy coefficients and demonstrate
their use in assessing the embodied energy of buildings using a similar method to that
in the work by Buchannan and Honey (1993). Gonzalez and Navarro (2006) carry out
an analysis of a real life building and its equivalent, fictional, counterpart constructed
from low impact materials. They demonstrate that careful selection of materials has the

19
potential to reduce the energy requirements and environmental impacts of
construction. However, it is noted that selecting a material with a higher embodied
energy may be preferable if the lower energy option has other negative environmental
impacts (Gonzalez and Navarro 2006). Issues associated with comparing embodied
energy values for houses and apartments are discussed by Hammond and Jones
(2008), the variations in living space, stairways and external works affect the values for
different housing types.

A higher value of embodied energy may still be a viable construction option if it results
in better performance. The greater embodied energy can be balanced against the
value of operational energy saved. This relates to comments by Gonzalez and Navarro
(2006) about the selection of a material with a higher embodied energy if it has a lower
environmental impact in some other way, such as pollution during manufacture.

2.3.5. Performance

The performance of construction methods can vary greatly, depending on their material
components. The potential exists to greatly improve the performance of housing by
selecting alternative construction methods which demonstrate the desirable
characteristics. For example, the airtightness of brick and block construction is
considered to be 4.5m3/hr.m2 at 50Pa (Miles-Shenton et al. 2007); this can be
compared with an airtightness of 0.5 to 1.0 m3/hr.m2 at 50Pa for Insulating Concrete
Formwork (ICF) construction (Miller 2012). A lower value of airtightness is desirable as
reduced movement of air results in lower heat loss and hence lower operational
energy. The continuous nature of the concrete achieved by ICF contributes greatly to
the low value of airtightness. In comparison, the multi element nature of brick and block
construction makes such levels difficult to achieve.

Aspects to consider could include speed of construction, U-value, embodied energy,

durability, cost, wall thickness, airtightness, fire performance, and any other aspect of
performance that is considered important to those involved with constructing the
building and the end user. Including these aspects in the decision making process
adds to the complexity as it is rare for one construction method to perform best on all
aspects under consideration. The optimisation method discussed later in this work
provides a way to deal with this issue.

20
In this work consideration is given to U-value (assessed through operational energy),
embodied energy, airtightness, fire performance and wall thickness. Although this is a
limited range, it was selected for use in the optimisation process with the intention that
the methods demonstrated could be used for future expansion of the construction
method selection methodology created in Chapter 6.

2.4. Decision making in construction

The decision making process for adopting innovative solutions has been considered in
Chapter 2.3.2. An important consideration from this is that time is required for
designers to become familiar with new products in order to use them (Emmitt and
Yeomans 2008, Mackinder 1980). If time is not available or other difficulties with
decision making arise, such as lack of information or understanding, designers will
typically opt for solutions they are familiar with (Emmitt and Yeomans 2008). In the
case of this work, the familiar decision would be brick and block. This can be seen as a
barrier to the adoption of alternative methods of construction. Emmitt and Yeomans
(2008) also note that construction professionals often have preferred options, which
they will choose over alternatives based on this personal preference.

However, despite the tendency of the industry to resist change, alternative methods of
construction are used in some cases. Methods which can be used in the information
gathering and decision making aspects of adopting innovation are considered here.
Two aspects of assessment are considered- the way in which a particular method
performs can be assessed using a wide range of tools and methodologies; ways in
which a method can be selected from a range of options are also examined.
Optimisation was chosen as the selection method for this work; it is discussed in
greater detail in this chapter. Background information and examples of use are given.

2.4.1. Assessment methods

Performance assessment will typically form part of the method selection. It is usually
necessary to assess the performance of an option in order to compare it with an
alternative. As further options are identified they will need to be assessed to allow
comparison and identification of the best option. As the assessment may need to be
carried out multiple times when selecting from a range of options, it is necessary to
balance time and cost requirements with the level of detail achieved.

21
Options can be assessed in relation to a National, or local, set of regulations or
recommendations. Examples of these include the Code for Sustainable Homes (CfSH)
in England and Wales, and the Leadership in Energy and Environmental Design
(LEED) system in the U.S. An example of this is work carried out by Castro-Lacouture
et al. (2009). Alternatively the options can be assessed against a set of criteria
selected by the method designer, such as in work by Abeysundara et al. (2009),
Emmanuel (2004), Seo et al. (2004), Li and Shen (2002) and Harris (1999). Lifecycle
assessment (LCA) is used as a way of examining the performance and cost of a
building throughout its lifespan. This method is demonstrated by a number of authors,
including Monahan and Powell (2011) and Anastaselos et al. (2009). Issues with LCA
are discussed by Björklund (2002).

Harris (1999) considers the assessment of the environmental impact of building

materials in terms of a number of factors. Embodied energy is considered here, the
calculation method used is the same as the one presented by Hammond and Jones
(2008). Material quantities are combined with the embodied energy per unit to create a
total. Those factors which cannot be assessed using quantitative means were scored
from 0-3 in terms of perceived impact. In work by Harris (1999) the values found for
each impact factor are not combined into a single figure. Harris (1999) considers that
this would risk masking issues in one area with benefits in another, preferring the use
of a profile. As well as assessing the performance of an option, Harris (1999)
demonstrates the use of the profiles to compare two options. Although comparison of a
small number of options can be carried out visually based on a profile, if a larger
number of construction methods are to be compared this would become highly
complex. Visual inspection of a profile may make it difficult to identify the best option as
it is rare for a material to perform best on all factors. In order to compare many different
options in this work a single figure must be generated to allow the use of optimisation.
In addition, the application of weighting would limit issues associated with poor
performance being in a highly important area.

Monahan and Powell (2011) demonstrate the use of a lifecycle analysis in comparing
the construction of a timber framed building with a more traditional equivalent. The
work by Monahan and Powell (2011) examines the embodied energy and associated
carbon emissions of the buildings considered throughout its entire life, from material
production to disposal at end of life. As with other methods presented in existing work,
identification of the best option by Monahan and Powell (2011) is carried out visually.
While this is effective for a small number of options it would be inefficient across the
number of options considered in this work, as an example, the case study presented in

22
Chapter 7 would require the consideration of 5.8x104 options. LCA is widely considered
to give the most accurate values as all inputs and outputs are required. However, to
carry out a full lifecycle assessment on the larger number of options examined in this
work would be highly computer intensive and time consuming, and hence costly and
impractical for design work. In addition to cost and time taken, difficulties associated
with LCA identified by Björklund (2002) include reliability of input data and changes
over time. Therefore, a method of assessment was sought that would consider the
costs and benefits of each option in a way that showed acceptable accuracy but
required a smaller degree of calculation.

The method presented by Anastaselos et al. (2009) makes use of a materials database
to generate a LCA for a range of insulation options. The method relies on the user
inputting their option and examining the results. Identification of the best option is not
carried out by the tool described by Anastaselos et al. (2009); it is an assessment
method rather than a selection tool. Work by Anastaselos et al. (2009) gives
consideration to the embodied energy associated with the materials used. This was
considered to be an important aspect of the environmental cost of construction which is
often ignored in building design. Although social aspects are mentioned in work by
Anastaselos et al. (2009) the acceptability of the options considered is not.
Acceptability, and its impact on economic viability, is considered to be an essential
factor for inclusion in a construction method assessment.

2.4.2. Construction method selection

It has been noted that making a decision which of two or more construction methods to
use can be challenging (Castro-Lacouture et al. 2009, Diakaki et al. 2010, Seo et al.
2004, Li and Shen 2002). Examples of just a few of the criteria which could affect the
decision include environmental impact, cost, buildability, acceptability, safety, past
experience, thermal performance, ability to meet the Building Regulations, and many
more. The large number of factors which apply means that no method is likely to be
easily identified as the best option. For example, a low operational energy may result
from using a large amount of insulation; however this would give a high embodied
energy. Selecting the best option will typically require compromise (Li and Shen 2002).
The relative importance of each factor should also be considered. For example, if the
intention is to meet the requirements of the Code for Sustainable Homes (CfSH),
operational energy would be extremely important, embodied energy is not considered.

23
Therefore, the previous example with low operational, but high embodied energy would
be acceptable. If the importance were reversed it would not be considered acceptable.

The selection of construction materials and methods has been addressed by a number
of authors. For example, Tas et al. (2007) consider the selection of materials based on
a database; Castro-Lacouture et al. (2009) use optimisation to identify the best
material in relation to a modified Leadership in Energy and Environmental Design
(LEED) rating and Seo et al (2004) use fuzzy decision making to select building
materials. Material selection in other industries, for example manufacturing, can also
provide useful insights into methods of making the best compromise when selecting a
material or construction method. Rao (2006) presents the use of a “material suitability
index” for material selection in an engineering environment. Although it is often the
selection of individual materials which is considered in these works, they can be
applied to the selection of a construction system, as examined here. Works which
discuss the selection of materials typically start with the discussion of an assessment
method; the selection is made based on the results of this. Work from international
authors was considered as the methodologies can be incorporated in work with a focus
on England and Wales, although details such as Building Regulations may vary.

A recurring theme in works relating to the issue of building material selection is the
need to balance the many different aspects and impacts of construction. Some of the
issues identified include environment, cost, performance and occupant convenience
(Seo et al. 2004). The fact these requirements are often in direct conflict is
demonstrated by Castro-Lacouture et al. (2009) and noted by Seo et al. (2004) and
Diakaki et al. (2010). The key to resolving this is identified as finding the best
compromise.

The use of an environmental suitability index is discussed by Emmanuel (2004).

Aspects of wall material construction, in this case embodied energy, cost and
reusability were assessed and scored. Normalisation, based on the average value
achieved for each factor between all the options was carried out. The sum of the
normalised values then provides the Environmental Suitability Index, which is used as
the basis for decision making. The work has similarities with the equation used in this
thesis as the basis for optimisation. Different factors are combined after normalisation
to create a score. However, the work by Emmanuel (2004) does not give consideration
to weighting of factors. In addition, the comparison between options is done visually,
which would not be possible over the number of options considered in this work. As
optimisation seeks to identify the lowest value, the equation designed for this work

24
penalises undesirable characteristics by increasing the score, and rewards desirable
ones by reducing it. This is in contrast to the work by Emmanuel (2004) which seeks to
achieve the maximum score for the best option. An important point identified by
Emmanuel (2004) is that the output of a tool cannot be used to identify a definitive
“best” option, it can only identify the best of those considered for the situation
examined. This also reflects the impact of weightings; if they are used to change the
situation, the output may also be altered.

Abeysundara et al. (2009) give consideration to environmental, social and economic

factors. A key element of this work which separates it from others considered was the
assessment of social factors. The use of a survey to identify views on each of the
construction methods from both end users and those involved in construction was
identified as an effective method for determining acceptability in this work. The work
presented by Abeysundara et al. (2009) demonstrates similarities with that by
Emmanuel (2004). Scores are produced for environmental, economic and social
impacts; however, they are presented in a profile format, comparable to that
demonstrated by Harris (1999). As with other methods presented, the final selection is
based on a visual assessment of the scores achieved by each option. A graphical
representation was used by Abeysundara et al. (2009) to show how each option
scored in terms of the performance factors. This was not considered effective when a
large number of options are to be considered, as in this work.

The use of a modified version of the American LEED system for assessing the
environmental performance of buildings, combined with an optimisation approach is
demonstrated by Castro-Lacouture et al. (2009). The work presents a method for
scoring and selection of materials which will result in the highest LEED score. The
focus in the work by Castro-Lacouture et al. (2009) is balancing the requirements of
the LEED rating system with cost. The assessment of each option is evaluated by use
of an equation. The results of this are then compared to identify the optimal solution.
The use of a similar method to examine construction materials and methods in relation
to the Code for Sustainable Homes (CfSH) was considered. However, the CfSH
considers construction method only in terms of the Green Guide to Specification rating
and responsibility of sourcing (DCLG 2007a). This small number of categories makes a
similar process ineffective. In addition, a Green Guide to Specification rating of A or A+
is achievable by all construction methods examined in this work, meaning there is very
little differentiation between them (BRE Global Ltd 2008), therefore this could not be
used as the basis for construction method selection. The use of an optimisation based
approach was considered to be beneficial to this work. However, a greater number of

25
options were to be considered. Therefore, a manual calculation and comparison of
each would be impractical. The use of a genetic algorithm based optimisation
programme was considered to be a more efficient approach. The ability to consider a
range of criteria in the selection process, as demonstrated in the work by Castro-
Lacouture et al. (2009) was desirable. In addition to the benefits of the optimisation
process, a particular aspect of interest presented by Castro-Lacouture et al. (2009)
was the use of constraints, placing a limit on some aspect of the construction. The best
solution must be found, but within this limitation.

Some consideration was given to the topic of material selection from industries other
than construction. Rao (2006) presents to design of a material selection tool for the
manufacturing industry. The importance of material selection is discussed, in particular
with respect to the performance of each of the options in a range of areas. This can be
considered in relation to selecting a construction method for housing, as different
options will result in different performance, for example in terms of durability, fire
resistance, embodied energy. The use of database based selection tools is considered
by Rao (2006) but identified as insufficient, with a mathematical approach being
preferable. The method demonstrated by Rao (2006), based on the use of graph
theory and matrices, combined with fuzzy logic, was not selected for use as
optimisation using genetic algorithms was identified as a more efficient method for the
large number of options to be assessed in this work. However, the creation of an
equation, incorporating normalisation and weighting was used as the basis for the
optimisation process adopted. Fuzzy set theory was used as the basis for a
construction material selection method by Seo et al. (2004) and Li and Shen (2002). As
with work by Rao (2006), weightings and normalisation were incorporated, the use of
these was taken forward in this work although the fuzzy set theory method was not
used. The normalisation method discussed by Rao (2006) was considered for use,
however using an average value as a benchmark, as discussed by Emmanuel (2004)
was selected as preferable to using the highest performing option (Rao 2006) or the
method based on the best and worst scores presented by Seo et al. (2004).

Optimisation is essentially the act of identifying the optimal, or best, solution for the
design (Goldberg 1989). It can be said that all previously discussed methods of
material selection are forms of optimisation. However, they focus on assessing the
options, leaving the final identification of the best option to the designer. While this
method can be applied to a limited number of choices, if the aim is to select from many
options this becomes inefficient. Numerical optimisation, for example, by use of genetic
algorithms as discussed by Mourshed et al. (2011), is a more effective method.

26
Optimisation as considered in this work is a numerical method which uses a search
technique to solve an equation and identify the variables which produce the optimal
solution. The process is iterative; the search method will be repeated until the optimal
solution is considered to have been found. Goldberg (1989) discusses a number of
search techniques which can be used in optimisation; these include calculation based
methods, hill climbing, enumerative methods, random search and genetic algorithms.
In calculation based methods the equation is made to equal zero, the gradient which
will occur at the maximum or minimal point. The solution of the equation will give the
variable value that results in the maxima (or minima). This method becomes
complicated when there are multiple peaks to consider, focusing on a local peak may
cause a greater one to be missed. Hill climbing consists of evaluating the equation,
then following the gradient to find the minimum or maximum value. As with calculation
based methods there is the risk of missing a greater peak by focusing on a smaller
one. Goldberg (1989) identifies these methods as ineffective in real world situations
where multiple peaks commonly occur. Enumerative methods require all the options to
be evaluated. The optimal value is identified from the results. This method is possible
where the number of variables is small, but rapidly becomes inefficient with greater
numbers of variables. The use of random searches evaluates randomly selected points
to identify the optimal value. Goldberg (1989) describes these as little better than
enumerative methods as many points will need to be calculated to determine the
optimal solution has been identified. Genetic algorithms make use of the principle of
natural selection, with the new variables being based on the most successful values
from the previous iteration. Genetic algorithms use multiple values for each iteration,
instead of a single value as used by other methods. This provides a more robust
examination of the solution space. Genetic algorithms were chosen as the search
method to be used in this work as they are more efficient and provide a more robust
search than the other available techniques. The risk of missing the optimal solution by
focusing on local minima is also reduced (Goldberg 1989). Genetic algorithms are
discussed in greater detail in Chapter 2.4.3.

The use of optimisation in construction based decision making is considered by a

number of authors. Examples include; Mourshed et al (2011) present optimisation as a
method of finding the best location for luminares in a patient room; Diakaki et al (2010)
use the method in building design for selection of the building envelope and services.

Diakaki et al. (2010) make use of multi objective optimisation in identifying the most
energy efficient options when designing a building in terms of the envelope and
building services. The model presented by Diakaki et al. (2010) uses a set of three

27
equations, combined with optimisation procedures to identify the building combination
which has the best compromise in terms of cost, energy use and annual carbon
dioxide emissions. The use of optimisation was identified as a method in which many
options could be assessed; this was taken forward for use in this work. However, a
single objective problem was used, following work presented by Mourshed et al.
(2011). No consideration is given by Diakaki et al. (2010) to other aspects which may
be affected by a change in building envelope. Of particular interest in this work is the
impact on acceptability. It will be seen later in this work that a material can perform
very well in terms of operational energy, but have a low acceptability, impacting its
public appeal. The optimisation by Diakaki et al. (2010) may identify this as the optimal
solution; however the impact on economic viability is such that it is not the best solution
to use. This relates to the comments made by Emmanuel (2004) regarding the
subjectivity of what is considered sustainable in construction, and that no “best” can be
identified, only the best for the situation examined.

A significant benefit of optimisation based on genetic algorithms as presented by

Mourshed al. (2011) and Goldberg (1989) is the way in which it is able to cope with a
large number of options. Considering large numbers of options visually would be a
difficult and time consuming task. By use of a genetic algorithms and optimisation
programming to assess the options this is simplified. More details are given on genetic
algorithms in Chapter 2.4.3. The ability to consider multiple inputs, in the case of this
work, construction method and front building dimension, is another area in which
optimisation by use of genetic algorithms can be seen as advantageous over the other
methods of construction method selection considered. The methods discussed above
which do not make use of numerical optimisation would require a two stage process to
deal with two variables, which would be a long and time consuming process. In
addition, the work presented here can be adapted to allow a greater number of
variables with minimal work, whereas with a visual selection method this would cause
far greater complexity.

The possibility of selecting a sub-optimal solution was identified by Rao (2006), Seo et
al. (2004) and Li and Shen (2002); who consider that additional factors may affect the
final decision such that the best scoring option may not be selected for use. Ultimately,
the final decision is made by those responsible for the design. This is noted in terms of
optimisation by Mourshed et al (2011) and Goldberg (1989) who comment that the
optimal solution may ultimately not be the best when other criteria are considered.

28
A benefit of using decision making tools such as those discussed is identified by Li and
Shen (2002) as the increased transparency of the process. It becomes easier to justify
and see why a particular option was selected. This is beneficial when explaining the
decision to interested parties such as designers, financial stakeholders and customers.
In the case of optimisation, the use of Phi arrays, as discussed by Mourshed et al
(2011), provide a clear visual representation of the decision making process.

2.4.3. Optimisation by use of genetic algorithms

The background and details of optimisation using Genetic Algorithms are considered in
greater detail as this method was selected for use in this work. In this work the use of
genetic algorithms as the method used to search the solution space for the optimal
value is considered. In optimisation using genetic algorithms the equation, or fitness
function, is evaluated using a set of values at each evaluation, the population, rather
than a single value as used with other methods. The advantage of this is that it allows
a greater search of the solution space, reducing the risk of missing the optimal solution
by focusing on local minima as can be the case with other search techniques.

The initial values are then adjusted following principles of natural selection based on
those that occur in the natural world. The fitness function is evaluated again using the
new values. This process is repeated either a set number of times or until the optimal
solution is identified. This is more efficient than attempting to evaluate all possible
values or using random search as the use of natural selection causes the values to
move towards the optimum as the process proceeds.

Genetic algorithms are discussed in detail by many authors, including Gen and Cheng
(1997), Mitchell (1996) and Goldberg (1989). The key to genetic algorithms is the way
in which they mimic the process of natural selection. The initial run of the fitness
function uses a random set of values, collectively referred to as Generation 0. These
values are coded; binary is often a suitable method. This will convert the value into a
string. For example, 01000 represents the value 8. Each member of Generation 0 will
have a string and an output value of the fitness function.

The genetic algorithm is then used to create the next generation, Generation 1. This
will be the second set of values that will be used in the fitness function, with the hope
that these values will move closer to finding the optimal value. To create the next
generation a combination of reproduction, crossover and mutation are used. The

29
members of Generation 0 used as the basis (or parents) for Generation 1 are selected
based on their score from the fitness function. Better scoring members of Generation 0
are more likely to be selected as “parents”. This mimics the “survival of the fittest”
concept in the natural world, where the best members of the population are more likely
to survive and pass their genetic code onto the next generation.

Once the “parents” have been selected from Generation 0 it is necessary to carry out
crossover to mix the code and create new strings for Generation 1. A pair of “parent”
strings is randomly selected. At a random point in the string a cut is made and the
following data exchanged with the second parent string.

To avoid the issue of data being lost and points missed due to the genetic algorithm
closing too rapidly on the solution, an occasional mutation is allowed. One of the
characters in a string will be randomly selected and altered, or mutated. This changes
its value. The impact of the mutation will be seen in the calculation of the fitness
function. If the mutation has had a beneficial effect the change will be preserved in the
next generation by the bias towards successful strings. Table 2.2 demonstrates the
process of genetic algorithms visually using binary coded strings as the members of
the population. Coloured text has been used to indicate crossover points and mutation.

Table 2.2: Visual representation of using a genetic algorithm to create a second

generation

Generation 0 Generation 1
String Fitness Chosen as After After Fitness
values function score parents crossover mutation function score
01001 10 01001 01100 01101 16
10001 4 00100 00001 00001 1
00100 6 01001 01000 01000 9
10000 3 10000 10001 10001 8

The process of creating a new generation is repeated until either a set number of
generations have been worked or the improvement in fitness function score is below a
set level, i.e. the optimal solution has been identified.

Using this method eliminates the need to calculate the score for every possible option
and find the best score from this. The “survival of the fittest” theory allows the results to
converge on the optimal solution without need for all values to be calculated. This
results in significant savings in terms of computational time and associated cost.

30
The history of optimisation and the use of genetic algorithms to find the best solution is
discussed by Goldberg (1989); from the beginnings of the method as a way of
analysing biological science to the use of genetic algorithms in decision making from
engineering, medical applications and human behaviour. More recently, the potential
for using optimisation in design decisions is becoming more widely known. A number
of authors have discussed the potential for applying optimisation to a range of
problems in a range of disciplines. For example: Mourshed et al. (2011) discuss use of
the technique to solve the placement of luminaries in a patient room; Diakaki et al.
(2010) make use of multi objective optimisation in selecting energy efficiency
improvements for buildings; Lagaros et al. (2005) consider the use of multi-objective
optimisation to design space frames that will be subject to seismic loading.

Methods used by Mourshed et al. (2011) for identifying a suitable crossover probability,
by varying the value and running the optimisation for a limited number of generations
can be applied to this work.

The importance of sub optimal solutions is expressed by Mourshed et al. (2011); these
values can give valuable information about the problem and may be a preferable
solution when additional criteria, not in the algorithm, are considered, for example
aesthetics. This echoes comments made by Goldberg (1989) that in human situations
the optimal, or perfect, solution is rarely achieved. The target is to identify solutions
which approach the optimal and perform better than others.

The use of Phi-arrays, a graph on which each point displays additional information
about the fit of the point by its size and colour, is also discussed by Mourshed et al.
(2011). Phi-arrays allow a visual representation of the solution space and can be used
to identify locations with values which score well. Identifying these sub optimal, but
good scoring, points can be useful in making decisions. Phi-arrays also demonstrate
regions which have very poor scores. Identifying these gives information about
unsatisfactory solutions to the problem. Unsatisfactory values can be removed from
following optimisation runs, reducing the calculation time and associated cost. Phi-
arrays can be generated from data calculated during the optimisation process; they are
a by-product of the process, that provide useful information about the solution space.
The use of a three dimensional graph to represent the solution space can also be used
to give a visual impression of high and low scoring areas. This is an extension of the
contour plots demonstrated by McKeown et al. (1990) that takes advantage of
improved computer capabilities. However, the nature of three dimensional plots is that

31
some data will be obscured, in particular the low values that are being aimed for. Using
a Phi-array gives a clearer view of the solution landscape.

2.5. The level of user acceptability in construction decision making

The opinion of the housing user is given only limited consideration when construction
method decisions are made. Roy and Cochrane (1998) consider this to be a result of
the speculative nature of house building in England and Wales, with the end user not
being identified until the house is complete. Price, rather than customer focus, is
identified by Roy and Cochrane (1998) as the main motivator for the housing
construction industry. Barlow (2000) compares the house building industry to other
mass production industries, but notes that the housing industry has not developed a
customer focus the way manufacturing has, restricting its efforts to minor details
relating to fixtures and fittings. The need to consider the demands of the purchaser, by
use of a customer focused, rather than cost focused business strategy is suggested by
Roy and Cochrane (1998). Socio-economic impacts are also identified by Ravetz
(2008) as being one of the factors that can affect the building stock.

A demonstration of the importance of end user acceptability is given by Cavill (1999)

who discusses the impact of a 1984 television programme on timber framed
construction. The programme suggested timber frame was at risk from rotting and fire,
the loss of consumer confidence greatly reduced the number of houses built using this
method in the following years (Cavill 1999). Therefore, it can be seen that householder
support is essential for alternative methods of construction to be used on a large scale.

In addition to the primary purchaser, who buys the new build house, it is important to
consider future users. Barlow (2000) comments that buildings which are undesirable,
due to the type of land they are constructed on, may present future issues with value.
This can be considered in relation to the desirability of construction method; if future
purchasers are likely to consider a method undesirable there are implications for
saleability and mortgage valuations.

Roy and Cochrane (1998) give some consideration to the use of alternative methods of
construction. It is suggested that alternative methods may allow a more flexible interior
layout, enabling the customer to determine how they want the building to look
internally. Offering greater opportunities for customisation is also suggested by Barlow
(2000) as a way to increase the appeal of houses.

32
The construction industry is widely viewed as resistant to change (Ravetz 2008).
Methods that have been proven over time are preferred, as factors such as
performance, cost, time to construct and durability are all known. In reality, this means
the industry has a deep rooted preference for brick and block construction; the “typical”
method used in England and Wales. Mackinder and Marvin (1982) note that the
construction industry prefers experience over written information. It is also noted by
Emmitt and Yeomans (2008) that when difficulties are experienced with construction
decisions industry professionals will often opt for a familiar solution. Evidence of this
resistance to change exists when it is considered that although a wide range of
housing construction methods are available, brick and block construction was still used
for 88% of dwellings in England between 1990 and March 2009 (DCLG, 2010). Timber
framed construction was used for 7% of houses built in England during this time period
with “other” methods of construction making up 5% of housing construction (DCLG,
2010). Combating this resistance to change is a key challenge for alternative
construction methods.

The views of both the construction industry and the end user, the public who will
purchase the houses, are an important consideration when choosing a construction
method. If a particular method is not supported by the industry it is unlikely to be
adopted. If a method is not supported by the public, houses built from it are unlikely to
sell. If houses are not saleable they are not economically viable (Barlow 1999), this
was demonstrated by the fall in timber framed construction after a negative television
report in 1983 (Cavill 1999). Although the energy use and performance criteria of an
alternative method of construction may be desirable, ultimately it must be acceptable to
both of these groups to be economically viable, and therefore to be considered as a
potential material for housing construction. In this work the views of both the end user
(the public) and the construction industry have been considered. It was felt that neither
of these could be ignored as they have a significant bearing on the viability of
alternative construction methods in house building.

The need for consideration of social data is considered in a number of the papers
previously discussed in relation to construction method selection. Li and Shen (2002)
consider that housing decisions are too complicated to only consider in terms of
quantitative data, that consideration must also be given to social aspects. For example,
Li and Shen (2002) consider how the quality of life may be affected by construction
decisions and state that these can be difficult to quantify as they may be imprecise. A
number of social factors, including preference and convenience for occupants were
considered in work by Seo et al. (2004). Acceptability was taken forwards in this work

33
as the most significant social factor; however, the methodology demonstrated for
incorporating acceptability could be adapted to include other qualitative factors. The
importance of considering the views of all stakeholders is also considered by Li and
Shen (2002), in this work the views of both the public, who will become the
homeowners, and the construction industry, who will build the houses were
considered.

2.6. Contribution to knowledge

It is the aim of this work to make the following contribution to knowledge based on
gaps identified in the existing work.

 A review of the performance and energy associated with the construction

methods studied; including identifying embodied energy issues with
construction materials.
 Determining the acceptability of alternative methods of construction to the
public and the construction industry.
 Identification of ways to increase the acceptability of alternative methods of
construction allowing their benefits to be accessed
 An optimisation based methodology for the selection of construction methods
that considers energy use, acceptability and performance to identify the best
compromise for a given situation.

2.7. Summary

Elements identified during the literature review which are of particular relevance in
achieving the intended aims of this work are summarised below.

 The potential for making savings in housing energy use and the associated
emissions has been explored with consideration of the scale of new build
housing. Even a small energy saving on each unit would make a significant
contribution.
 Housing construction in England and Wales tends to be speculative; this
increases the importance of acceptability.
 The typical aspect ratio for housing is in the region of 1:2

34
 Brick and block is the main construction method used for housing in England
and Wales, timber frame has some use. Other methods are used very little.
 International use of alternative methods of construction is much higher.
 Ideally the minimum front dimension of housing should be 4.25m, it can be as
low as 3.5m in some cases.
 Any method of construction can be used provided it meets the required Building
Regulations.
 Typically no option will be the best on all counts, compromises must be made.
 The main impact of the Code for Sustainable Homes on this work is the need to
minimise operational energy.
 Construction methods with good airtightness and U values are preferable as
these will contribute to lower operational energy.
 Embodied energy should be considered when selecting a construction method.
 The use of different materials in construction has the potential to reduce
embodied energy.
 Reducing embodied energy is beneficial as the saving is made at construction,
unpredictable user behaviour will not affect the saving.
 Higher embodied energy construction methods may be acceptable if the
method has improved performance in other areas.
 The impact on various aspects of performance should be considered when
selecting a construction method.
 Options must be assessed based on some performance related criteria before
the best option can be selected.
 A wide range of material, or construction method, assessments exist.
 Desirable criteria may be in direct conflict with each other, for example cost of
construction and operational energy.
 Normalisation of the data is required to compare criteria of different sizes.
 Weightings should be included in the assessment method to allow different
areas of concern to be highlighted as appropriate.
 Many existing works use visual inspection of a score for identifying the best
option; this is considered insufficient for this work.
 Optimisation by use of genetic algorithms is able to cope with large number of
options and multiple variables.
 Optimisation based on genetic algorithms was identified as a way to choose
between methods for this work.
 Constraints can be incorporated using a penalty based method.

35
 A sub-optimal solution may be preferable to the identified optimal solution if
other criteria are considered.
 Genetic algorithms make use of survival of the fittest concepts to determine the
best solution.
 Optimisation by genetic algorithms eliminates the need to evaluate every
option, saving time and reducing costs.
 Phi arrays and three dimension graphs can be used as a visual representation
of the solution space. Sub-optimal solutions can be identified from these.
 It is not possible to identify a definitive best option, only the best option for the
situation, out of those considered.
 Innovation is required for the use of alternative methods of construction to be
adopted into the mainstream.
 Innovation requires one company to take the first step.
 Legislation to force innovation may be required.
 Education can encourage the use of alternative methods of construction.
 Decision making tools can increase the transparency of the process.
 Minimal consideration is currently given to customer acceptability when
construction decisions are made.
 The construction industry is viewed as resistant to change.
 Public and construction industry acceptability should both be considered when
selecting a construction method.
 Acceptability of a construction method must be high with both the public and
construction industry for it to be economically viable.
 Combating the resistance to change is key to increasing acceptability of
alternative method of construction.

36
3. Methodology
3.1. Introduction

The work carried out for this thesis builds on initial data collected through literature
study, combined with primary data collected through the use of questionnaires and
processed using optimisation techniques to create a design methodology with the
potential to aid decision making. Each stage of this process requires a separate
methodology.

For this work a limited number of construction methods were studied, the intention was
to demonstrate techniques, which could be used for future expansion of the design
methodology to create a functioning design tool.

3.2. Construction method selection

To find information on a range of construction options, a literature review was carried

out. This allowed a wide range of construction methods to be identified and assessed
for suitability. The construction methods were selected based on the following criteria:

 Prior use for housing in England or Wales.

 Availability of detailed information from academic literature and materials
suppliers.
 Inclusion of at least one method perceived as unconventional and one or more
modern methods of construction.
 Method suitable for use across England and Wales.
 Method to have an estimated lifespan in excess of 60 years.
 Suitable for large scale, speculative construction as carried out in England and
Wales.

The methods used in this work were selected from the identified options based on the
above criteria..

3.3. Construction method details

Methods were developed to assess the construction options for U value, embodied
energy and three aspects of performance which it was considered could be affected by
the changes in materials used across the methods. These were: air tightness, wall

37
thickness and fire performance. These methods can be used to generate values for
other methods of construction as required and for additional selection criteria such as
cost, durability and construction speed.

3.3.1. U-value

U-values were calculated based on the specified set of materials for each method of
construction. Published U-values, such as those produced by the BRE, were only used
where the specified details exactly matched the ones used to generate the published
values. This only occurred in the case of prefabricated straw bale construction.

The calculation of U values achieved by the wall systems were carried out using a
combination of material thermal resistance values as shown in Equation 3.1 (British
Standards Institution 2007, Anderson 2006, Chudley and Greeno, 2005).

(Eqn. 3.1)

1
U

Where

U = U value in W/m2K

Rso = External surface resistance in m2K/W

Rn = Thermal resistance of building component in m2K/W

Ra = Thermal resistance of cavity air space in m2K/W (if present)

Rsi = Internal surface resistance in m2K/W

A correction of 0.020 W/m2K was applied to those walls using wall ties (Chudley and
Greeno, 2005).

R values for building components are based on a combination of the thickness of

material used and its thermal conductivity. They are calculated using Equation 3.2.

38
(Eqn. 3.2)

Where

R = Thermal resistance in m2K/W

L = thickness of material in m

λ thermal conductivity of material in W/mK

Where a material system such as a SIP unit was specified, the R value provided by the
manufacturer was used for the calculation. Values of thermal conductivity for other
materials were sourced from a combination of British Standards Institution (2007),
Anderson (2006) and Chudley and Greeno (2005). If the U value for the entire system
was provided by the manufacturer, the given value was used in preference to
calculating one. Where applicable the external leaf of mortar and brickwork were
treated as a combined material due to the similarity in their thermal conductivity
(Anderson 2006, Chudley and Greeno, 2005).

Values for external surface resistance (Rso), air space resistance (Ra) and internal
surface resistance ( Rsi) were taken from Chudley and Greeno (2005); these are shown
in Table 3.1. Rso is based on a normal level of exposure and a high surface emissivity.
Rsi takes the typical value for walls. Ra is the typical value for a cavity wall void.

Table 3.1- Surface and airspace R values used in U value calculations

Component R value
Rso 0.055
Ra 0.180
Rsi 0.123

3.3.2. Embodied energy

Embodied energy is the energy required to produce the materials and components of a
construction system. There are a number of methods for the calculation of embodied
energy. Input out-put analysis, process analysis, hybrid analysis and coefficients can
all be used to evaluate the embodied energy of a component or system. A comparison
of these methods by Hamilton-MacLaren et al (2009) showed that the most suitable
method to use for this work was the one used by Hammond and Jones (2008) and

39
Buchannan and Honey (1993), where material quantities are combined with embodied
energy coefficients to calculate the total embodied energy for the system. This method
was chosen to give a balance between calculation requirements and accuracy.

Material quantities for 1m2 of each construction type were calculated to allow
comparison between the methods. Where a wall system used units greater than 1m2
the amounts were calculated for a unit then divided by the unit area. Material quantities
were based on the specification given for each construction method combined with
manufacturer data. Values for material quantities were then combined with embodied
energy coefficients for the materials used. Embodied energy coefficients were obtained
from the Inventory of Carbon and Energy V 2.0 database produced by Hammond and
Jones (2011). This was considered to be the most suitable source of data as the
values included have undergone a careful screening process to ensure they are as
suitable as possible for the UK situation and hence are most suitable for England and
Wales. The generation of these values is discussed by Hammond and Jones (2008).
Where a specific material value was unavailable the closest value for a similar material
was substituted. Summing the individual material values provided the total embodied
energy associated with the materials for 1m2 of wall type. This process was carried out
for both external and party walls for each construction method.

Morel et al (2001) demonstrate the effect that transportation of materials can have on
the embodied energy of a project. However, as all materials can be sourced from
within the UK and a precise location is not being examined in this work, the impact of
transportation on energy use is considered to be the same for each construction
method and hence is not included in the analysis. If the method were used in the
assessment of a project with known location the transportation distances of materials
could be calculated and these added to the embodied energy value used in the
optimisation.

For this work, the impact of site activities and worker transportation has not been
assessed. Records from actual case studies would be required to accurately calculate
this. It was considered that the values would be similar in all cases, although it was
noted that faster construction time may lead to lower requirements and hence lower
associated emissions. This was identified as an area for further work.

40
3.3.3. Airtightness, wall thickness and fire performance

Three additional areas of performance were included in the optimisation procedure.

These were air tightness, wall thickness and fire performance. All were considered to
be affected by the use of alternative methods of construction when compared to typical
construction methods.

Details of the performance achieved by each method were obtained from technical
literature. A combination of manufacturers’ product information and published works
was used to provide the most applicable data for the products being considered.

3.4. Acceptability of construction methods

Acceptability of alternative methods of construction is a key theme of this work, in

particular the way this affects their potential for use. Acceptability of the construction
methods considered in this work was examined from two points of view, that of the
construction industry, who would build the houses and of the public, who would buy the
houses.

3.4.1. Data collection method

The aim of the questionnaire was to determine the awareness and acceptability of the
construction methods under consideration to the public. This is an area which is not
typically considered during construction planning, therefore the collection of primary
data was required for this part of the work. To determine the views of the public on the
methods being studied, a questionnaire was used. Kothari (2004) gives benefits of
questionnaires as more cost effective; avoiding interviewer bias; ability to achieve
larger samples; respondents can complete the questionnaire when it is convenient for
them and have time to consider their answers. These factors were considered
beneficial for this work.

Interviews were discarded as a method of collecting the required information as the

number of sources required to give a representative view made interviews impractical.
The main benefit of interviews in this situation is the potential to ask additional
questions and obtain a greater depth of response. As discussed later in this chapter,
the design of the questionnaire allowed for optional open-ended responses to some
questions. This allowed a greater depth of response, achieving results similar to those
which would have been achieved from a structured interview.

41
In addition, it was considered desirable to collect the views from different regions and
determine whether the location of the respondents appeared to impact the acceptability
of each construction method. The use of a questionnaire, as opposed to interviews,
made it possible to collect sufficient responses from a diverse range of locations.
Examples of work using questionnaires to obtain this type of data include Mahaptra et
al. (2012), Osmani and O’Reilly (2009), Goodier and Gibb (2007), Pan et al. (2007),
Lorenzoni et al. (2007), Stubbs (2002) and Chinyio et al. (1998).

Disadvantages associated with the use of questionnaires given by Miller and Salkind
(2002) and Kothari (2004), such as ambiguous responses, missed questions, slow
returns and inflexible questions were addressed with the use of careful questionnaire
design and the use of an internet based distribution and response collection. This
method allowed for compulsory questions, optional open ended responses and faster
collection than would have been achievable by a postal questionnaire. The possibility
of interviewer bias was avoided by careful wording of the questions, and where
possible, the use of repeated questions styles, for example, the acceptability question
for each construction method followed the same format, with a brief description and
image.

3.4.2. Questionnaire design

The theory of questionnaire design is dealt with by a number of authors such as Fink
(2006), Kothari (2004), Gillham (2000) and Oppenheim (1992). From these works it
can be seen that the development of a questionnaire requires the following stages:

 Examination of research aims

 Interviews or background research to find suitable questions
 Deciding on the population and, if applicable, sample of participants
 Development of questions
 Testing of questions
 Layout of questionnaire
 Testing of questionnaire
 Pilot run
 Full run
 Follow up contact as required, e.g. to improve responses or debrief participants
 Analysis of data
 Reporting of results

42
It is considered by Gillham (2000) that all these stages are required to produce a high
quality questionnaire, resulting in the best achievable response rates and providing
meaningful data. In addition, the need for consideration of ethics and approval from an
appropriate ethical committee is considered by Fink (2003).

The research aims of the house holder questionnaire were identified as:

 What factors are important to the public when considering a house to buy?
 Are the public aware of the construction techniques being examined?
 Would the public buy a house built using the construction techniques being
examined?
 What would increase the public’s interest in houses built using alternative
methods?

The research aims of the house builder questionnaire were identified as:

 What factors are important to house builders when selecting a construction

method?
 Are house builders aware of the options being examined?
 Do house builders have experience of using the methods being examined?
 Would house builders use the construction methods being examined?
 What would increase house builders’ interest in alternative wall construction
methods?

Questions were designed based on these aims and findings from the literature review
discussed earlier in this work. The aim was to keep the questions simple and succinct
in order to collect usable data and increase response rates by keeping the
questionnaire short.

The importance of a range of factors was tested using a Likert scale. Factors were
selected based on the perceived interests of the respondents. Each factor was scored
from 1 (not important) to 5 (very important).

Awareness was determined with the use of simple yes/no responses. This was based
only on the name of the construction method, with no additional information being
supplied in order to identify the current situation. House builders were also asked if

43
their company had any past experience with the method, with options of “Yes-
housing”, “Yes- other”, “No” and “Not sure”.

Acceptability was determined by providing a brief description; along with a labelled,

full-colour, image of each method in turn. Respondents could indicate acceptability with
a choice of “yes”, “no” or “maybe” to the question “Would you buy a house built using
this method?”. The use of labelled, colour images, as opposed to technical drawings
was intended to increase the accessibility of the questionnaire; technical knowledge
was not required to understand the images. An optional, open ended, question allowed
for further comments or explanation of the choice. This enabled an interview style
response on a larger scale than could be achieved in person and limited the over-
simplification of responding with only yes, no or maybe.

For the house builder survey, technical terms were used in the descriptions for each
construction method, for example, use of the term leaf to describe the layers of the
wall. House builders were asked “Would you build a house built using this method?”

To determine factors that may increase respondents’ interest in alternative methods of

construction a number of possible options were presented with a “yes/no” response
option. An optional open-ended question was also included, allowing respondents to
add any other factors they felt would increase their interest, as with the questions
regarding acceptability, this allowed for an in depth response and avoided over-
simplification.

The questionnaire was built and distributed using an online survey maker (Survey
monkey 2011). This simplified the design process and allowed automatic collection of
responses. This resulted in greater accuracy as the potential for transcription errors
was removed. Additionally, it was possible to make certain questions compulsory,
encouraging more complete surveys, with fewer missed questions.

A small pilot run was carried out. This allowed testing of the questions, layout and
questionnaire as a whole and provided feedback which made it possible to develop the
questionnaire further. As a result of comments received, the images of each method
were altered from two dimensional cross-sections to labelled, axonometric images as
shown in Figure 4.1. Some respondents felt they would have liked more information on
each method of construction, in particular pros and cons. The decision was made not
to alter this as the intention was to determine acceptability based on a combination of
current knowledge and the given method description.

44
Ethical approval was sought for the questionnaire from Loughborough University’s
ethical approval committee. This was received provided the questionnaire respondents
remained anonymous and any participant recruitment advertising was cleared with the
committee. Conducting an anonymous survey was also considered to be beneficial in
improving response rates.

Samples of the two questionnaires used can be seen in Appendix A and B.

3.4.3. Population and sample

The population for the public questionnaire included residents of England and Wales
with the potential to purchase a house in the future. This was restricted to those aged
18 or over at the time of responding to the survey to avoid the ethical considerations of
working with minors. No upper age limit was set. Based on estimated figures for 2010
from the Office of National statistics (2011), this gives a population of approximately 48
million individuals. The use of online survey distribution limited the sample to those UK
residents with computer access. This is estimated to be 82.9% of the population
(Internetworldstats, 2004), which was considered to be an acceptable limitation.

Calculation of the required number of respondents to achieve a 95% confidence level

with a confidence interval of 5 following the method described by Bartlett et al. (2001)
indicated 384 responses were required. This means that values from the survey will
contain the true mean value 95% of the time if this number of respondents was
achieved. A small confidence interval, such as 5, indicates the sample values will be
close to the true mean values. 384 was taken to be the minimum acceptable number of
responses, with a greater number being considered to improve the quality of findings
by reducing the confidence interval and allowing for partial responses.

The population for the industry survey was considered to be all house construction
firms working in England and Wales, and therefore their employees who would answer
the survey based on a combination of their own perceptions and their company’s views
and strategies. The varying size of construction firms meant that a target population
was difficult to set. A minimum of twenty five responses was considered to be
acceptable as it would allow a range of opinions and provide sufficient responses for
some statistical analysis.

45
3.4.4. Questionnaire distribution

In order to distribute the survey, a website was designed for the project. This provided
a brief explanation of the work, as well as a link to the survey. A list of frequently asked
questions and a contact sheet was provided to give respondents further information if
required. Wording of all sections was crafted carefully to avoid introducing bias into the
results.

Willingness to participate places a limitation on survey responses, therefore a range of

participant recruitment methods were adopted, this included:

 Use of personal contacts via email

 Professional networking websites www.linkedin.com and
www.graduatejunction.com
 The social networking website www.facebook.com
 Staff and student contacts through the University mailing lists.

To increase the distribution, respondents were asked to complete the questionnaire

and to pass details of the website onto their own contacts. This makes use of the
“small world theory”, more commonly known as the “six degrees of separation”
whereby any two individuals can be connected by six or fewer other individuals
(Schnettler, 2009). By use of this theory it would be possible to contact all those
residents of England and Wales who have computer access if all individuals were
willing to participate.

The potential exists for this type of distribution to result in bias due to self selection.
Those who are interested in the subject or hold a strong opinion are most likely to
respond, whereas those who have little opinion may not complete the questionnaire.
To reduce the impact of this bias, the target sample for the public survey was
increased to a minimum of 500. The range of distribution methods was also aimed at
reducing bias in the results by accessing a range of respondents.

The larger sample size also allowed for sampling from within the responses if
necessary. To enable sampling, questions such as region and age band were asked.
Income level was considered as a question but was discarded as it was felt it would
reduce the response rate.

46
Distribution of the industry survey followed the same method used for the public survey
(See 4.2.3). Two additional methods of distributing the survey were used to increase
the industry response rate, these were:

 The forum section of the online construction magazine “Building”

 Emails to the top twenty house building firms in the UK for 2009 as given by
McMeeken (2009)

3.4.5. Response analysis

To analyse the public survey the number of respondents for each region was
compared with the regional population distribution for England and Wales. The
difference in distribution between respondents and the actual population was assessed
using a chi squared test. This indicated that the distributions were not comparable so
sampling was carried out within the responses to create sub-samples which mirrored
the actual population distribution of England and Wales. Five sub-samples were
created and used in the statistical analysis of importance factors.

To determine the importance of factors to respondents, percentage scores, modal and

median scores for the Likert scales were calculated. From visual inspection of the
resulting graph and table it was possible to create a ranking for the importance of the
factors. Principal component analysis was carried out on the householder survey
results to group the factors into a smaller number of categories about which it is felt
householders are concerned. It was not possible to carry out principal component
analysis on the house builder survey due to the lower number of responses.

Awareness of the construction methods being considered and ways to increase

interest were not divided by region, they were considered for the area of study as a
whole. Graphical representation of the results was used to analyse responses.

Acceptability was considered by region. Visual examination of tables and graphs were
used to analyse these responses. Views expressed in the optional open-ended
question were coded, based on content and whether it was negative or positive. Where
a query was made about performance, it was coded as a negative view as it was
considered to indicate a concern about that aspect of the method; e.g. “How does this
perform in fire” was felt to indicate a concern about the fire performance. This allowed
the perceived benefits and disadvantages of each method to be identified.

47
3.5. Optimisation based selection methodology

The final element of this work was the creation of a design methodology to aid with the
selection of construction methods. This combines data collected in the previous
sections and uses optimisation to identify the best option for use.

Sub-optimal solutions can also be identified as they may provide valuable information
as discussed by Mourshed et al. (2011) and Goldberg (1989). The optimisation
process and the use of genetic algorithms have been discussed in detail in the
literature review that forms part of this work (Chapter 2.4). The optimisation based
design methodology in this work requires an equation which produces a single value
as its output, the single objective equation. The single objective equation forms part of
the fitness function. This is a programme which evaluates the input variables and
produces a single value to indicate their performance in terms of the factors
considered.

Use of the fitness function allows the investigation of a range of factors, including
energy use, acceptability and performance, combined with weightings. Entering values
for each variable in the fitness function allows a single situation to be assessed by
creating a single value that incorporates these factors. This can be compared with a
second situation by repeating the process and comparing the output values from the
fitness function.

Optimisation uses this process on a large scale to assess a range of situations and
identify the optimal combination of variables. Variables are selected to be altered
during the process, in this work two have been used, all others are fixed. The design
methodology will run the fitness function multiple times, altering the input values each
time based on a genetic algorithm, until the optimal solution is found.

3.5.1. Problem definition

The design of the optimisation based methodology focused on the data collected
throughout this work. Factors selected for the basis of the decision were embodied
energy, operational energy, public acceptability, industry acceptability, air tightness,
fire performance and wall thickness. It was the intention of this work to demonstrate the
potential of this methodology, rather than to create a fully functional design tool.
Expansion to include other factors which would affect construction method selection is
intended as further work.

48
As part of the fitness function it was necessary to generate values for energy use, both
operational and embodied. Data collected and generated in Chapter 3 was combined
with user inputs to calculate these values. To allow the combination of factors with
different scales all values were normalised by use of benchmark figures. A weighting
was applied to each factor and the values combined to generate the single objective
equation. The output of this is a single value that combines all of the inputs plus
weighting to show their importance- the single objective value. Details of the fitness
function development are given in Chapter 7.

3.5.2. The use of optimisation

The background and details of optimisation using genetic algorithms have been
discussed in Chapter 2.4.3. Optimisation allows the calculation of values which
approach the optimal, eventually identifying the optimal value.

The initial run of the optimisation uses a number of randomly selected variable values.
Subsequent runs use variable values that are based on the most successful values
from the previous run. These values are selected based on a genetic algorithm that
uses the principles of natural selection to identify the best values in the first run and
carry them through to the second run or “generation”. Crossover and mutation are
applied to the values, altering them before the second run of the equation. After the
single objective equation has been evaluated using the second generation of values,
the process will be repeated. This continues until the optimal value is identified by no
improvement in single objective value between the generations or until a set number of
generations have been run. Examples of previous work using this methodology include
Mourshed et al. (2011), Diakaki et al. (2010) and Lagaros et al. (2005).

3.5.3. Optimisation method

To carry out the optimisation a Matlab based optimisation programme was used. This
method uses the Matlab computer programme (Mathworks Inc. 2012). The
optimisation programme used in this work was provided by M. Mourshed, following the
methods used in Mourshed et al. (2011) and Mourshed et al. (2003).

To run the programme for it is necessary to set values for crossover and mutation
probabilities, population size and number of generations. Calculation of mutation and
crossover probabilities was carried out using the method described by Mourshed et al.

49
(2011) using the case study values presented in Chapter 3.6.1, values for mutation and
cross-over can be seen in Chapter 3.6.3. The maximum number of generations and
the population size was investigated using the case study values, the result of this can
be seen in Chapters 3.6.4 and 3.6.5 respectively.

3.6. Case study development

3.6.1. Case study design

Values were chosen for the case study based on housing data for England and Wales.
Data from 2006 was used for numbers of houses constructed, as pre-recession data
was considered to be more representative of the construction industry in a “healthy”
state, which is likely to return as the economy improves. Table 3.2 shows the number
of residences constructed by type for England and Wales.

Table 3.2: Number of residences constructed in England and Wales during 2006 by
type of housing (NHBC 2007)

Semi Flats and

Country Detached detached Terraced maisonettes
England 23728 20762 29660 72667
Wales 2508 1320 1122 1716
Total 26236 22082 30782 74383
Percentage of total 17.09 14.38 20.05 48.46

Only houses were considered for this work, flats and other forms of multi-residence
construction were not included due to the different styles of their construction, in
particular, the greater number of storeys used in multi-house residences. The
maximum number of storeys considered in this work was three. Therefore, the next
most frequently constructed type of housing was terraced.

Table 3.3 shows the number of houses constructed by number of bedrooms for
England and Wales during 2009/10. Figures are shown as a percentage of all housing
constructed.

50
Table 3.3: Percentage of all housing constructed in England and Wales during 2009/10
by number of bedrooms (DCLG 2012b and Rees 2011)

Percentage of all housing

Number of
bedrooms England Wales
1 1 1
2 10 13
3 25 35
4 or more 19 20

From this it can be seen that the most frequently constructed number of bedrooms is
three. Table 3.4 shows the average floor areas for housing by number of bedrooms.
Values for England and Wales were not available, so this figure was based on data for
the UK as a whole.

Table 3.4: Floor areas by number of bedrooms for the UK, values from Scott Wilson
Ltd. (2010)

Number of Mean floor area

2
bedrooms (m )
1 64.30
2 71.20
3 95.60
4 120.60
5 163.50

Using the data from Tables 73.2, 3.3 and 3.4, a case study was designed based on a
terraced, three bedroom house with a floor area of 95.6m2. The East Midlands was
selected as the location for the case study as this region had the most robust results
for acceptability as a result of having the greatest response rate for the questionnaire.
All other values adopted the optimisation programme default values. The values
entered into the optimisation programme are shown in Table 3.5.

51
Table 3.5: Constant values for case study optimisation

Factor Value
2
Total building area 95.6m
Number of storeys 2
Height of each storey 3m
Number of external doors 2
Region 2 (East Midlands)
Building type 3 (Terraced)
Maximum window percentage 25
2
Door size 2.42m

3.6.2. Optimisation variables

The optimisation process was run using the values given in Table 3.5 as constant
values. The variables for the optimisation process were wall construction method and
value of dimension (a). These values were altered by the optimisation process to find
the optimal combination, which produces the lowest value of the fitness function. This
represents the best compromise in terms of energy use, acceptability and
performance. To maintain integer values in the optimisation, dimension (a) was given
in millimetres, this was converted to metres in the fitness function calculation. It was
necessary to specify a minimum and maximum value for each variable, the
optimisation programme used values between these bounds. The bounds for the case
study are given in Table 3.6.

Table 3.6: Variables and bounds for case study optimisation

Variable Lower bound Upper bound

Construction method 1 6
Dimension a 3600 13278

Material values are bounded by the number of options. The minimum value for
dimension a is based on the smallest length that will allow a 1.1m wide door and a
1.0m window plus wall area to support these. This is taken to be 3600mm. In addition,
Chown (1999) suggests that the minimum front dimension that can be practically used
for housing is 3.5m, with 4.25m being preferable. The maximum value of a is based on
Equation 3.3.

52
(Eqn. 3.3)

3.6.3. Mutation and crossover probabilities

Mutation and crossover probabilities affect the way in which the optimisation process
alters the input variables with each successive run. The crossover probability is input
into the optimisation by use of the crossover function, a value between 0 and 1. To
determine the most effective value of crossover function (Pc), the value was increased
from 0.0 to 1.0 in steps of 0.05. The maximum number of generations was set to fifty to
reduce the running time of the experiment. Five optimisation runs were carried out for
each value. The value of best fitness was plotted against crossover function. This
graph can be seen in Figure 3.1.

2.2

2.1

2
Best fitness

1.9

1.8

1.7

1.6
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Value of crossover fraction

Figure 3.1- Crossover fraction against best fitness value for the case study runs.

A quadratic curve was fitted to the results produced. The best value of crossover
function was the one that resulted in the lowest value of best fitness. This is because
the optimal combination of construction method and dimension (a), is the one which
produces the lowest value from the fitness function. It can be seen from Figure 3.1 that
this occurs at a value of Pc=0.4. This was used in the case study run. Lower values of

53
Pc result in a slightly higher best fitness so are less suitable. As Pc increases from 0.4
the value of best fitness achieved increases noticeably, indicating that higher values of
Pc are less suitable. It can be seen that at values of Pc greater than 0.5 the spread of
best fitness results increases, including the outlier seen at a crossover fraction value of
1.0. From this it was determined that a high degree of mutation is necessary for this
programme to achieve accurate values of best fitness.

The mutation probability (PM) is set as a result of the crossover fraction. The mutation
rate comes from subtracting the crossover fraction from unity; therefore, it was set to
PM= 0.6.

3.6.4. Number of generations

The number of generations determines the maximum number of times the optimisation
programme will alter the input variables and run the fitness function. If the improvement
in fitness function value calculated is less than the specified value at any point, the
optimisation programme will cease to run, as the optimal solution is considered to have
been achieved.

It was noted when running the optimisation programme to determine the best value of
crossover fraction that none of the runs were completed in fewer than the fifty
generation maximum. Therefore, the number of generations needed to be higher than
fifty to allow for a complete optimisation run.

To determine the number of generations, the value was initially set at 300. The
optimisation process was run five times, with a note of when the process completed if it
stopped prior to 300 generations. The number of generations was decreased to 200
and then to 150 and a further five optimisation runs carried out for each value.
Population size was set at 50 for the trial runs, crossover fraction to 0.4. The results of
the trial runs are shown in Table 3.7.

54
Table 3.7- Number of generations to complete the optimisation run compared to
maximum number of generations.

Number of generations
Maximum number to complete Best
of generations optimisation fitness
300 89 1.69067
300 82 1.68103
300 96 1.68440
300 143 1.68729
300 160 1.68199
200 85 1.69646
200 Max. number exceeded 1.68055
200 100 1.68296
200 135 1.68922
200 194 1.68103
150 113 1.68199
150 Max. number exceeded 1.68296
150 95 1.68922
150 109 1.68199
150 101 1.68103

Table 3.7 shows that 93% of the optimisation trials completed in fewer than 200
generations. When the number of generations was decreased to 150, only 73%
completed within the maximum number of generations. As a result, 200 was chosen as
the number of generations for the case study.

3.6.5. Population

The population is the number of individual values in each generation. The size of the
population can have a significant effect on the time taken for the optimisation
programme to complete its search for the optimal value, which can result in high costs
(Mourshed et al. 2011). A small population may result in a rapid convergence on a
solution, which risks missing important values.

The optimisation considered in this work has a relatively small runtime; therefore, a
number of trial runs were carried out using varied population sizes (25, 50, 100, 150).
For these runs the number of generations was set at 200, the crossover probability
used was 0.4. The results of these trial runs can be seen in Table 3.8.

55
Table 3.8- Impact of varying population size on the optimisation methodology results.

Number of generations
Population to complete Best
size optimisation fitness
25 83 1.70032
25 140 1.68922
25 95 1.71675
25 133 1.68922
25 171 1.6926
50 89 1.68006
50 144 1.68826
50 143 1.67958
50 160 1.68826
50 88 1.68440
100 93 1.68006
100 106 1.68392
100 175 1.68199
100 125 1.68006
100 108 1.67958
150 90 1.67958
150 148 1.68055
150 131 1.68537
150 132 1.68199
150 149 1.68055

It was noted that a slight improvement (decrease) in the value of best fitness occurs as
the population increases. The time taken for the optimisation to complete at higher
values of population was noticeably longer, in particular once the population reached
150. A population size of 100 individuals was selected as a compromise between
runtime, speed of convergence and value of best fitness achieved.

Examination of the number of generations taken to complete the optimisation in Table

3.7 further supports the use of 200 generations, as all runs completed within this time.

3.7. Summary

The nature of this work means that each section required a very different approach,
including the use of both primary and secondary data, with the final element being the
combination of all data collected and generated along with optimisation techniques to
create the construction method selection methodology. The use of this methodology
was then demonstrated with the case study described in Chapter 3.6. The results for
each section can be found in Chapters 4-8.

56
4. Construction methods
4.1. Introduction

Throughout this work the phrase “construction method” is used to refer to the system
used to construct the house wall, including all materials contained therein. The
materials used will have an impact on the way in which the wall is built, including the
processes involved.

To allow a comparison and identify possible alternatives to brick and block construction
for the walls of new build housing, six construction methods were selected to be
studied in depth. The selection justification and the construction method details are
discussed below. For the work, each construction method used was evaluated using a
set of specified materials, details of these are given in 4.4-4.9. Altering the material
specifications has the potential to affect the comparison between construction
methods. This provides an area for further work.

4.2. Selected methods

Construction methods studied were:

 Brick and block construction (“typical” method)

 Structural Insulated Panels (SIPs)
 Insulating Concrete Formwork (ICF)
 Prefabricated straw bale and lime plaster panels
 Thin joint block work
 Timber framed construction with brick cladding

Brick and block was included as the baseline for comparison as this is the most
frequently used method of construction in the area of study. Structural Insulated
Panels, Insulating concrete formwork and Thin joint block work were included as
examples of Modern Methods of Construction, Prefabricated straw bale panels were
included as an unconventional material, timber frame was included as an alternative
method of construction which some people may have experience with.

It was considered desirable, although not essential, to maintain a brick external

appearance across as many of the methods as possible. This follows comments made
by Roy and Cochrane (1998) that more than half their survey respondents preferred
traditional styling when considering a house purchase. This would allow housing built

57
using alternative methods to blend in visually with existing properties, a fact considered
important for planning purposes as well as to achieve high public acceptability. A brick
finish was achieved for five of the options; prefabricated straw bale panel construction
has a plastered external appearance.

4.3. Discarded methods

As a result of the criteria in Chapter 3.2 some of the options initially considered were
eliminated as they were unsuitable for the work. These were:

 Off-site modular or “pod” construction- It was considered that this method would
not be comparable with the other options considered in this work as it deals
with components other than wall materials.
 Rammed earth construction- This construction method is very site specific,
relying on the availability of suitable soil types for use in construction. As a
result of this it could not be used in all areas of England and Wales so was not
included in the work.
 Local materials- It was considered the definition of what is a local material and
the complexities of creating a methodology including these applicable to the
entire area of study was too great for inclusion at this stage.
 Steel frame and cladding- Expected to have similar levels of acceptability to
timber frame, although with variation in performance values.
 Recycled materials- The use of recycled materials is often considered to be a
particularly sustainable option as it makes use of materials which already exist
(Calkins, 2009). The most common use of recycled construction materials is in
steelwork, as steel framed housing was not included in the work recycled steel
was also discarded.
 Reused materials- differ from recycled materials as it involves use of the
materials in the original form, rather than after reprocessing (Calkins, 2009).
Reused materials can potentially be incorporated into most types of
construction provided the materials can be recovered. The variable availability
of materials for reuse across the area of study meant they were not considered
as a universal option.
 Load bearing straw bale construction- it was considered that this method is not
sufficiently standardised for large scale speculative construction.

58
  Natural materials- Typically based on one of the considered methods, such as
brick and block or timber frame with natural alternatives for elements such as
insulation. Not included at this stage due to the limited difference between this
option and the included ones.

The optimisation based design methodology developed in this work (Chapter 6) has
been designed to allow for further expansion at which point some or all of the
discarded options could be incorporated.

4.4. Brick and block

Brick and block was included as a reference construction method. It is the most
commonly used method in the area of study. As an example it was used for 88% of all
dwellings constructed in England from 1990 to March 2009 (DCLG, 2010). If
alternatives are to replace this “typical” construction method they must show significant
benefits over it.

4.4.1. Construction details

Brick and block walls are constructed using a double leaf system. An external leaf of
brick work is separated from the internal concrete block work leaf by a cavity. Both
leaves are constructed simultaneously, with components held together using mortar.
The cavity contains insulation against the concrete leaf and an air space against the
brick leaf. Wall ties set into the mortar during construction link the leaves. Internally the
wall is finished with plasterboard, a plaster skim and paintwork. This construction
method is shown in Figure 4.1.

59
Figure 4.1- Axonometric view of brick and block construction

To allow the consideration of terraced and semi-detached properties a party wall

specification was also created. This consists of a fully filled and sealed construction to
minimise thermal loss between properties. The wall was taken to be constructed from a
double layer of lightweight concrete blocks, with insulation between the layers. On both
sides the walls are finished with plasterboard, a plaster skim and paint. Material
specifications for both external and party walls are given in Table 4.1.

Table 4.1- Material specification for brick and block construction

Component Specification
Brick Fired clay red brick, dimensions- 102.5 x 65 x 215mm
Mortar 1:3 cement sand mix, 10mm joints
Wall ties Stainless steel wire, plastic insulation retainer, 2.5 ties per
2
1m wall
Air space 25mm width
Insulation Mineral wool insulation, 100mm
Block 8MPa compressive strength, 100 x 215 x 440
Plasterboard 12.5mm thickness
Plaster skim 6mm skim
Paint Double coat of water based paint
Party wall blocks 100 x 215 x 440 lightweight blocks
Party wall ties Stainless steel tie, no insulation retainer
Party wall insulation 75mm mineral wool

60
4.4.2. U value calculation

Current Building Regulation requirements place a maximum value of 0.30W/m 2K on

external walls (HM Government, 2010). Calculation of the U value for this system was
based on values for thermal conductivity given by British Standards Institution (2007),
Anderson (2006) and Chudley and Greeno (2005). These are given in Table 4.2.

Table 4.2- L, λ and R values for brick and block components

L λ R
2
Component (m) (W/mK) (m K/W)
Mortared brickwork 0.1025 0.84 0.12
Air space 0.025 0.18 0.14
Mineral wool 0.1 0.038 2.63
Block 0.1 0.18 0.56
Plasterboard 0.0125 0.16 0.08
Plaster skim 0.006 0.57 0.01

These values, combined with the values for Rso, Ra and Rsi in Table 3.1 give
U=0.257W/m2K.

Corrected for wall tie placement in accordance with Chudley and Greeno (2004) this
results in a value of U=0.277W/m2K for use in this work.

The fully filled design of the party wall, with no cavity as described above, results in a U
value of U=0W/m2K in accordance with Part L of the Building Regulations (HM
Government, 2010). Therefore, heat loss through party walls is not considered in this
work.

4.4.3. Embodied energy

The embodied energy (EE) value for brick and block work was expected to be high due
to the processes involved in making the materials, for example firing of clay bricks and
the creation of cement.

Material quantities were based on the specification given in Table 4.1. The values used
in the calculation of embodied energy for 1m2 of brick and block wall are shown in
Table 4.3 and Table 4.3 for external and party wall respectively. The values for
volumes of material are based on the specification given in Table 4.1, values for the

61
embodied energy coefficients were taken from the Inventory of Carbon and Energy
V2.0 database developed by Hammond and Jones (2011).

Table 4.3: Material quantities and embodied energy coefficients for1m2 brick and block
work external wall

2
Volume Density Mass EE EE per 1m
wall area
3 3
Component (m ) (kg/m ) (kg) (MJ/kg) (MJ)
Brick 0.085 1700 144.330 3.00 432.99
Mortar 0.018 1750 30.800 1.33 40.96
Wall tie N/A N/A 0.513 56.70 29.09
Insulation retainer N/A N/A 0.061 76.70 4.67
Insulation 0.100 25 2.500 16.60 41.50
Block 0.093 1400 130.760 0.59 77.15
Mortar 0.007 1750 2.067 1.33 2.75
Plasterboard 0.013 950 11.875 6.75 80.16
Plaster skim 0.006 1300 7.800 1.80 14.04
Paint N/A N/A N/A N/A 21.00
744.30

Summing the values from Table 4.4 gives a value of 744.30MJ for the embodied
energy associated with 1m2 of brick and block work external wall. It can be seen that
the energy intensive process of brick manufacture has a significant impact on the
embodied energy, contributing over half the total value. The embodied energy of
concrete blocks is lower than expected, although the mass of blocks is high the low
value of embodied energy per kg results in a low overall value.

62
Table 4.4: Material quantities and embodied energy coefficients for 1m2 brick and block
work party wall

2
Volume Density Mass EE EE per 1m
wall area
3 3
Component (m ) (kg/m ) (kg) (MJ/kg) (MJ)
Paint N/A N/A N/A N/A 21.00
Plaster skim 0.006 1300 7.800 1.80 14.04
Plasterboard 0.013 950 11.875 6.75 80.16
Lightweight Block 0.093 600 56.040 3.50 196.14
Mortar 0.007 1750 11.494 1.33 15.29
Wall tie N/A N/A 0.513 56.70 29.09
Insulation 0.075 25 1.875 16.60 31.13
Lightweight Block 0.093 600 56.040 3.50 196.14
Mortar 0.007 1750 11.494 1.33 15.29
Plasterboard 0.013 950 11.875 6.75 80.16
Plaster skim 0.006 1300 7.800 1.80 14.04
Paint N/A N/A N/A N/A 21.00
713.46

The party wall specification for brick and block construction has an embodied energy
value of 713.46MJ for 1m2 of wall. It can be seen that the high embodied energy of the
lightweight blocks results in a higher total embodied energy, even though their mass is
low.

4.4.4. Airtightness, wall thickness and fire performance

The airtightness of brick and block wall relies on good quality construction, in particular
good pointing and plastering of the internal wall face (Lecompte 1987). The
airtightness value used was based on the testing of a number of buildings constructed
using the same method given in the specification for this work. The average value of
airtightness found during this testing over a range of house types was 4.5m 3/hr.m2 at
50Pa (Miles-Shenton et al 2007). This value was used for brick and block in the
optimisation procedure. The Building Regulations maximum value for airtightness is
10m3/hr.m2 at 50Pa (HM Government 2010).

The wall thickness for this construction method was calculated from the values of L in
Table 4.3. This method was considered to have a wall thickness of 0.346m.

The fire performance of brick and block walls has been demonstrated many times over
the years. Testing has shown it will satisfy the Building Regulations requirement of 60
minutes burn time resistance for external walls over 5m in height (HM Government

63
2011), this is supported by anecdotal evidence. A value of 240 minutes burn time was
used in the optimisation procedure based on values given by British Standards
Institution (2005) and The Brick Industry Association (2008).

4.5. Structural Insulated Panels (SIPS)

4.5.1. Construction details

Structural Insulated Panels are composed of two layers of engineered timber with a
foam layer sandwiched between them. Commonly used foam types are expanded
polystyrene (EPS) and extruded polystyrene (XPS), fixed to the boards with adhesive,
and polyurethane foams which do not require adhesives (Morley, 2000). Oriented
Strand Board (OSB) is typically used for the timber layers. Foam thickness can vary,
allowing a range of performance values.

Erection of the building is often done by hand, however, mechanical lifting devices are
required if large sized panels are used. Panels are placed in position before external
cladding or brickwork is constructed.

It is necessary to provide impact and weather protection to the panels. This is achieved
by a breathable membrane applied to the external face of the SIP and either external
cladding or, as in this work, by an external leaf of brickwork. Wall ties link the SIP
panels to the external brickwork leaf. To avoid damage to the panels a small air filled
cavity is maintained between the breathable membrane and the brickwork. Internally
the wall is finished with two layers of plasterboard, a plaster skim and paint. This
construction method is shown in Figure 4.2.

64
Figure 4.2- Axonometric view of SIP construction

Party wall construction for this type of dwelling can be either timber stud work or SIP
panel. SIP panels were chosen for this work to limit the crossover between
construction types. The method consists of a layer of SIP panel, finished on both sides
with a vapour control membrane, two layers of plasterboard, a plaster skim and paint.
The material specification used in this work is given in Table 4.5.

Table 4.5: Material specification for SIP construction

Component Specification
Brick Fired clay red brick, dimensions- 102.5 x 65 x 215mm
Mortar 1:3 cement sand mix, 10mm joints
2
Wall ties Stainless steel, 2.5 ties per 1m wall
Air space 50mm width
Breather membrane Kingspan Nilvent breathable membrane
SIP panel Kingspan Tek SIP, 142mm wide (2 x 15mm OSB sheets,
112mm urethane foam)
Plasterboard 2 x 12.5mm thickness
Plaster skim 6mm skim
Paint Double coat of water based paint

65
4.5.2. U value calculation

SIPs are considered to have very good thermal performance as the thickness and type
of insulation is selected with this in mind. Dye and McEvoy (2008) state that U values
of 0.2W/m2K are easily achieved and White (2007) gives a value of 0.17-0.18W/m2K.
In addition, a report by the Department for Communities and local Government (2009)
notes that additional insulation can be added to the SIP, with an extra 50mm of
insulation resulting in a U value of 0.14W/m2K. Values used to calculate the U value of
the SIP wall system are given in Table 4.6. These values are taken from information
supplied by Kingspan (2009), British Standards Institution (2007), Anderson (2006) and
Chudley and Greeno (2005). The impact of the breathable membrane was discounted
due to the very low value of L for this component.

Table 4.6: L, λ and R values for SIP components

L λ R
2
Component (m) (W/mK) (m K/W)
Mortared brickwork 0.1025 0.84 0.12
Air space 0.050 0.18 0.28
OSB 0.015 0.03 0.50
Urethane foam 0.112 0.023 4.87
OSB 0.015 0.03 0.50
Plasterboard 0.025 0.16 0.16
Plaster skim 0.006 0.57 0.01

From the values given in Table 4.6, the values for Rso, Ra and Rsi in Table 4.1 and the
correction for wall ties given by Chudley and Greeno (2005) a U value of 0.167W/m2K
was calculated for SIP construction using the materials specified in Table 4.5.

The solid construction of the party wall allows it to have a U value of 0 based on Part
L1A of the Building regulations (HM Government, 2010). Therefore, heat loss through
party walls is not considered in this work.

66
4.5.3. Embodied energy

Material quantities and embodied energy coefficients for 1m2 of SIP construction as
detailed in Table 4.5 are shown below in Table 4.7 for external walls and Table 4.8 for
party walls.

Volumes of material are based on the specification given in Table 4.5, values for the
embodied energy coefficients were taken from the Inventory of Carbon and Energy
V2.0 database developed by Hammond and Jones (2011).

Table 4.7: Material quantities and embodied energy coefficients for 1m2 SIP external
wall

2
Volume Density Mass EE per EE per 1m
kg wall area
3 3
Component (m ) (kg/m ) (kg) (MJ) (MJ)
Brick 0.085 1700 144.330 3.00 432.99
Mortar 0.018 1750 30.800 1.33 40.96
Wall tie N/A N/A 0.150 56.70 8.51
Breather membrane N/A N/A 0.140 95.89 13.43
Oriented Strand Board 0.015 680 10.200 15.00 153.00
Urethane foam insulation 0.112 41.07 4.600 101.50 466.88
Oriented Strand Board 0.015 680 10.200 15.00 153.00
Plasterboard 0.025 950 23.750 6.75 160.31
Plaster skim 0.006 1300 7.800 1.80 14.04
Paint N/A N/A N/A N/A 21.00
1464.12

Summing the values from Table 4.7 gives an embodied energy value of 1464.12MJ for
1m2 of SIP wall with a brick outer leaf as detailed above. Elements which have a large
impact on the value of embodied energy include bricks, the OSB sheets and the
urethane foam core. OSB sheets were expected to have a low value of embodied
energy due to their organic nature. However, they are highly processed, resulting in a
high value of embodied energy per kilogram.

67
Table 4.8: Material quantities and embodied energy coefficients for 1m2 SIP party wall

2
Volume Density Mass EE per EE per 1m
kg wall area
3 3
Component (m ) (kg/m ) (kg) (MJ) (MJ)
Paint N/A N/A N/A N/A 21.00
Plasterboard 0.025 950 23.750 6.75 160.31
Breather membrane N/A N/A 0.140 95.89 13.42
Oriented Strand Board 0.015 680 10.200 15.00 153.00
Urethane foam
insulation 0.112 41.07 4.600 101.50 466.88
Oriented Strand Board 0.015 680 10.200 15.00 153.00
Breather membrane N/A N/A 0.140 95.89 13.42
Plaster skim 0.006 1300 7.800 1.80 14.04
Paint N/A N/A N/A N/A 21.00
834.77

The party wall as specified for the SIP construction has an embodied energy of
834.77MJ for 1m2 of wall. The high energy requirements of the OSB and the foam core
raise the value. The effect of removing the brick outer skin on the total embodied
energy can be seen. This suggests systems which use cladding other than brickwork
may perform better in terms of embodied energy.

4.5.4. Airtightness, wall thickness and fire performance

A number of authors such as Barista (2008), Morley (2000) and Greeley (1997)
consider the airtightness of SIPs to be very good as a result of the large panel sizes
with a small number of joints and therefore low potential for air leakage. This is
quantified by Dye and McEvoy (2008) as 1.6 air changes per hour at 50Pa and by
Gaze (2008) as 1.27m3/h.m2 at 50Pa, both an improvement on the building regulations
requirement of 10m3/hr.m2 at 50Pa (HM Government 2010). Rudd and Chandra (1994)
discuss testing carried out to compare the airtightness of a SIP house with an identical
timber framed house and found the SIP construction to be significantly more airtight.
However, it is noted by Morley (2000) that good workmanship on joints is of particular
importance with this construction method to avoid air leakage. For this work a value of
1m3/hr.m2 at 50Pa was used based on data supplied by the SIP manufacturer
(Kingspan 2009).

The wall thickness for the SIP system was based on the material specification given in
Table 4.5. The SIP method of construction was considered to have a wall thickness of
0.326m.

68
It is noted that the components of SIPs are susceptible to fire by Calkins (2009),
Barista (2008) and Griffen et al. (2006). Particular issues associated with the burning of
SIPs are noted by Griffen et al. (2006); these include the possibility of the OSB skin
separating from the core, the ability of fire to spread throughout a building via the panel
cores, the production of dense smoke during burning and the low melting point of EPS.

In order to minimise fire risk the industry standard is the application of plasterboard to
the internal surface of the SIP, to separate the SIP from sources of ignition and provide
sufficient time for occupants to exit the house before the structure is compromised.
Depending on use the requirement is given as one or two 12.5mm sheets of
plasterboard by Hairstans and Kermani (2007). Bregulla and Enjily (2004) state that
the use of a plasterboard lining enables the Part B of the Building Regulations for
England and Wales to be met for SIP structures. The performance of SIPs from each
manufacturer is tested by the British Board of Agrément (BBA) and must achieve set
levels before they can be sold for use in the UK. Based on the use of two 12.5mm
sheets of plasterboard to provide fire protection the fire resistance of the SIP system
used in the work is 73 minutes, given by the manufacturer (Kingspan 2009).

4.6. Insulating Concrete Formwork (ICF)

4.6.1. Construction details

ICF can take the form of large panels, or smaller blocks. This work focuses on the
block system. Blocks consist of two layers of expanded polystyrene (EPS) or extruded
polystyrene (XPS) foam, held apart at fixed dimensions by plastic or metal ties. The
foam thickness varies with the choice of product, allowing a range of insulation values
to be obtained as required.

Blocks are stacked by hand to form the walls, they commonly have indentations in the
top and bottom surfaces to ensure correct placement (Davies 2006). Rebar is often not
required but can be added if necessary; it is positioned using the ties which hold the
leaves of the block apart. Concrete is poured into the cavity in the centre of the blocks
and gently vibrated where rebar has been placed to ensure compaction (Insulating
Concrete Formwork Association, ICFA 2009). The concrete pouring stage is generally
carried out at a maximum of one storey per day to avoid over loading the formwork.
During the concrete pour additional support may be required in the form of bracing and
trestles to ensure the walls remain true, this is removed once the concrete has set

69
(Davies 2006, Evans 2006). Construction of the following layer or roof can begin the
next day (ICFA 2009).

ICF systems require cladding to protect the insulation from mechanical damage and
the impact of UV radiation which can result in degradation of the foam (ICFA 2009). A
wide range of finishes can be used. Plaster, render and cladding can be applied
directly to the ICF (Barista 2009, ICFA 2003). For this work a brick skin has been used
to maintain the traditional appearance used in housing construction in England and
Wales. Wall ties to connect the brickwork to the ICF are inserted in the ICF before the
concrete pour (Davies 2006, Evans 2006). A small air cavity is maintained between the
ICF foam and the brickwork. Internally the wall is finished with plasterboard, a plaster
skim and paint. This construction method is shown in Figure 4.3. Table 4.9 gives the
material specification used for the ICF construction considered in this work.

Figure 4.3: Axonometric view of ICF construction

Party walls are constructed using the ICF firewall 313 system. The walls are finished
on both sides using one layer of plasterboard, a plaster skim and paint.

70
Table 4.9: Material specification for ICF construction

Component Specification
Brick Fired clay red brick, dimensions- 102.5 x 65 x 215mm
Mortar 1:3 cement sand mix, 10mm joints
2
Wall ties Stainless steel, 2.5 ties per 1m wall
ICF system Beco Wallform 313 Firewall
Concrete Pumpable RC 25 concrete, 10mm aggregate
Plasterboard 12.5mm thickness
Plaster skim 6mm skim
Paint Double coat of water based paint

4.6.2. U value calculation

ICF has good thermal properties with U values of 0.10-0.30 W/m2K given by ICFA
(2009). A value of 0.25 W/m2K without finishes is given by Funke (2006) and 0.25-
0.13W/m2K for filled but unclad blocks depending on size by Evans (2006). Greeley
(1997) notes that the good thermal performance of the foam insulation compensates
for the poor performance of concrete, resulting in a good overall value.

ICF blocks can be provided with a range of insulating options to suit the project
requirements, insulation leaves can be of different thicknesses to achieve desired
values (ICFA, 2009). Additional insulation can also be added to either face of the ICF
after the concrete pour, before finishings are applied to increase performance. Table
4.10 gives the values of thickness, thermal conductivity and thermal resistance used to
calculate the U value for the specified system, values were taken from Beco Products
Ltd. (2008), British Standards Institution (2007), Anderson (2006) and Chudley and
Greeno (2005).

Table 4.10: Values of L, λ and R used to calculate the U value for the ICF wall system

L λ R
2
Component (m) (W/mK) (m K/W)
Mortared brickwork 0.1025 0.840 0.122
Air space 0.0150 0.180 0.083
Concrete filled ICF 0.3130 0.0657 4.764
Plasterboard 0.0125 0.160 0.078
Plaster skim 0.0060 0.570 0.011

71
A U value of 0.204W/m2K was calculated for ICF construction using the data in Table
4.10, Table 3.1 and the wall tie correction given by Chudley and Greeno (2005).

The solid construction of the party wall allows it to have a U value of 0W/m 2K based on
Part L1A of the Building regulations (HM Government, 2010). Therefore heat loss
through the party walls was not considered in this work.

4.6.3. Embodied energy

Concrete has a high value of embodied energy due to the energy required to create its
components, in particular cement. Calkins (2009) notes the high energy inputs and
associated carbon dioxide emissions of this process as a disadvantage of using
concrete in construction. This relates directly to the use of ICF as large volumes of
concrete are required. Pierquet et al. (1998) also state that the concrete component of
ICF results in the system having a high embodied energy and that this results in a long
payback period; i.e. the time it takes for the operational energy savings to outweigh the
greater embodied energy cost is long.

A number of measures can be taken to reduce the embodied energy and therefore the
negative environmental impact of concrete use. Calkins (2009) and Johnston and
Gibson (2008) suggest the use of concrete containing elements such as fly ash to
reduce the cement content and hence lower the embodied energy. The use of recycled
aggregates in the concrete is also suggested by Calkins (2009) and Anink et al. (1996)
as a way to improve the environmental impact of concrete. These measures rely on
availability of materials and test results to prove the acceptability of the materials for
use.

Material quantities and embodied energy coefficients for 1m2 of external and party wall
constructed using the ICF system are shown in Table 4.11 and Table 4.12
respectively.

Material volumes are based on the specification given in Table 4.9, values for the
embodied energy coefficients were taken from the Inventory of Carbon and Energy
V2.0 database developed by Hammond and Jones (2011).

72
Table 4.11: Material quantities and embodied energy coefficients for 1m2 ICF external
wall

2
Volume Density Mass EE EE per 1m
wall area
3 3
Component (m ) (kg/m ) (kg) (MJ/kg) (MJ)
Brick 0.085 1700 144.330 3.00 432.99
Mortar 0.018 1750 30.800 1.33 40.96
Wall tie N/A N/A 0.513 56.70 29.09
Polystyrene N/A N/A 5.180 87.40 452.73
Concrete N/A N/A 335.620 0.78 261.78
Metal spacers N/A N/A 3.210 36.00 115.56
Plasterboard 0.013 950 11.875 6.75 80.16
Plaster skim 0.006 1300 7.800 1.80 14.04
Paint N/A N/A N/A N/A 21.00
1448.31

Summing the values from Table 4.11 gives an embodied energy value of 1448.31MJ
for 1m2 of ICF wall to the specification in Table 4.9. Components which have a high
impact on the embodied energy value of SIP construction are bricks, polystyrene,
concrete and the metal spacers which form the ICF units. If a lower fire resistance is
acceptable polystyrene spacers can be used which would reduce the embodied energy
of this system.

Table 4.12: Material quantities and embodied energy coefficients for 1m2 ICF party wall

2
Volume Density Mass EE EE per 1m
wall area
3 3
Component (m ) (kg/m ) (kg) (MJ/kg) (MJ)
Paint N/A N/A N/A N/A 21.00
Plaster skim 0.006 1300 7.800 1.80 14.04
Plasterboard 0.013 950 11.875 6.75 80.16
Polystyrene N/A N/A 5.180 87.40 452.73
Concrete N/A N/A 335.620 0.78 261.78
Metal spacers N/A N/A 3.210 36.00 115.56
Plasterboard 0.013 950 11.875 6.75 80.16
Plaster skim 0.006 1300 7.800 1.80 14.04
Paint N/A N/A N/A N/A 21.00
1060.47

Party walls constructed using the specified ICF system are calculated to have an
embodied energy value of 1060.48MJ per 1m2. The impact of removing the brick and
the associated energy is high; indicating that alternative cladding options may be more
competitive in terms of embodied energy. The embodied energy value remains high for

73
ICF party walls due to the energy requirement for making polystyrene, concrete and
metal spacers.

4.6.4. Airtightness, wall thickness and fire performance

The continuous nature of the concrete when ICF is used has the potential to produce a
very airtight building. A number of authors note the high level of airtightness achieved
by the use of ICF (ICFA 2009, Barista 2009, Mosey et al. 2009 and Ross 2005). This is
quantified in examples given by Moedinger-Clay (2007) of 0.8ACH and Kośny et al
(1998) of 0.257-0.051ACH with an average of 0.147ACH. Information provided by
Beco, manufacturers of the Wallform system specified here, give the achievable
airtightness as 0.5 to 1.0 m3/hr.m2 at 50Pa (Miller 2012). For this work the higher value
of 1.0m3/hr.m2 at 50Pa has been used, an overestimation was felt to be fairer for
comparison than an underestimation.

Wall thickness was calculated from the material specifications given in Table 4.9 and
the thicknesses given in Table 4.10. The ICF system being examined was found to
have a wall thickness of 0.449m.

The fire performance of ICF is considered to be very good. Fire retardants are used in
the foam to limit the risk of burning. Concrete will not burn and the continuous nature of
the concrete achieved using this method means there is little risk of building collapse in
the event of a fire (ICFA, 2003). Design details are also important for reducing fire risk.
Internally lining the ICF with gypsum plasterboard improves the fire performance of the
building (ICFA 2009, Barista 2009). Barista (2009) also notes that the foam should not
be continuous across floors to limits risks. Beco class the 313 Wallform Firewall
system as F90 –AB, indicating a fire resistance of 90 minutes (Beco Products Ltd.
2008). The firewall version of the Wallform 313 system was selected for this work in
order to achieve this higher fire rating. The standard 313 system has a slightly lower
value of λ, resulting in a better U value for the system; however, it only achieves 30
minutes of fire resistance, making it unsuitable for buildings over 5m.

74
4.7. Prefabricated straw bale and lime plaster panels
4.7.1. Construction details

Straw bale buildings can be constructed using a number of methods, including load
bearing bales and the use of bales as an infill for a timber frame. For this work, the
system created by ModCellTM was used as it was felt this produces a more
standardised construction than the load bearing method and presents a different
method to the timber framed construction considered later in this work. The system
uses timber framed panels which are filled with bales and covered with lime plaster, or
render, off site. Construction of the units takes place under cover, near to the final
building site, often in a large barn or similar building. The units are then transported to
site and installed as prefabricated panels with a finishing coat of lime plaster on both
sides after placement (Modcell 2012a). This construction method is shown in Figure
4.4.

Figure 4.4- Axonometric view of prefabricated bale wall construction

Party wall construction can be carried out either by using an insulated timber stud wall
or by use of the same method as for external walls. For this work, the prefabricated
bale panel system was selected for use in party walls. Material specification for the
prefabricated straw bale system is given in Table 4.13.

75
Table 4.13: Material specification for prefabricated bale wall construction

Component Specification
Timber framing 480mm wide,100m thick, length as required. PEFC timber
Straw bale 420mm wide straw bale
Lime plaster Lime:Sand 1:3, 27mm thick
Lime plaster skim 3mm

4.7.2. U value calculation

Straw bale construction is considered by many authors such as Arnaud et al (2009),

Calkins (2009) and Lawrence et al. (2009b) to have very good thermal properties.
Jones (2007) and Yates (2006) both give a U-value of 0.13W/m2K for a 450mm wall.
The low U values achievable are credited to the large amount of air trapped in the
bales which act as insulation.

ModcellTM provide a U value of 0.13 to 0.19W/m2K for the 0.48m straw bale system
being used in this work (Modcell 2012a). The higher value, 0.19 W/m2K, was used in
the optimisation process as it was felt over estimating the energy requirement was
fairer that under estimating. As the party wall construction is solid it is considered to
have a U value of 0 W/m2K (HM Government, 2010). Therefore heat loss through party
walls is considered to be zero and is not calculated as part of the operational energy
requirements.

Information supplied by Modcell, including thermal imaging of an existing building

constructed using this method, demonstrates that thermal bridging through the timber
framing is not an issue (White 2011).

4.7.3. Embodied energy

Straw bale construction has a low value of embodied carbon. During growth of the
plant material, carbon is stored in the stems which ultimately become straw. This will
be retained until it is released, for example by decomposition or burning. This carbon
storage is considered to be one of the potential benefits of straw bale construction by
Arnaud et al (2009), Lawrence et al (2009) and Apte et al (2008). The stored carbon
can be offset against the production, transportation and installation energy carbon
emissions, resulting in a low net value of embodied energy. However, Calkins (2009)
notes that while carbon storage is a benefit, straw cannot be considered carbon neutral

76
as energy is put into its production, for example, to bale the straw. Energy is also
required for the other materials used. As a result of this the embodied energy value of
the system is low, but not zero or a negative value.

The material quantities for 1m2 of wall constructed using prefabricated, lime plastered,
bale panels and relevant embodied energy coefficients are given in Table 4.14.

Values for volumes of material are based on the specification given in Table 4.13,
values for the embodied energy coefficients were taken from the Inventory of Carbon
and Energy V2.0 database developed by Hammond and Jones (2011).

Table 4.14: Material quantities and embodied energy coefficients for 1m2 prefabricated
bale wall construction

2
Volume Density Mass EE EE per 1m
wall area
3 3
Component (m ) (kg/m ) (kg) (MJ/kg) (MJ)
Lime plaster 0.027 1650 44.550 1.386 61.74
Straw 0.420 304 127.680 0.240 30.64
Timber 0.074 529 39.220 7.400 290.23
Lime plaster skim coat 0.027 1650 44.550 1.386 61.74
Lime wash 0.006 1650 9.900 1.386 13.72
458.06

Summing the values from Table 4.14 gives an embodied energy value of 458.61MJ for
1m2 of prefabricated, lime plastered, straw bale panel.

It can be seen from Table 4.15 that the main energy input for the construction of
prefabricated bale panels is the timber used to make the frame. The straw bales have
a very low energy requirement. Although the lime plaster accounts for a small volume it
is a dense and energy intensive material, resulting in a significant value of embodied
energy.

As party walls are constructed using the same technique this value is used for the
embodied energy of party walls as well as external walls.

77
4.7.4. Airtightness, wall thickness and fire performance

Yates (2006) discusses the need for a good finish with a high degree of airtightness to
help prevent moisture entering the walls and hence avoid the high moisture levels that
would lead to rotting. The Canada Mortgage and Housing Corporation (2000a) notes
that the use of a plaster finish aids airtightness. This is particularly the case with the
prefabricated panel system as the final plaster skim applied onsite creates an
opportunity for a continuous plaster barrier to limit airflow. The airtightness value given
for the straw bale system is 0.86m3/hr.m2 at 50Pa (Modcell 2012b).

The wall thickness of the prefabricated straw bale system is dictated by the width of the
bales used in the construction of the panels. Plaster adds a small amount to the width,
both in the prefabricated panels and the additional plaster layer applied on site. The
system selected for use in this work has a wall thickness of 0.48m (Modcell 2012a).

Straw bales are perceived to burn easily, presenting a considerable fire risk if used in
construction. However, this perception has been shown to be incorrect for plastered
bales in a number of tests. Testing by Apte et al. (2008) showed the importance of
render in restricting ignition by limiting air availability. Values were also obtained for
risk of ignition at various levels of fire intensity. Lower levels of fire intensity showed
temperatures too low to cause ignition, however higher intensity fire resulted in
temperatures capable of causing ignition after 40 minutes. Ignition did not occur due to
the low levels of oxygen inside the plastered bale; however smouldering did occur,
resulting in ignition when the plaster was cut open and the straw exposed to the air.

Testing discussed by Jones (2007) into fire resistance of straw bale walls shows that
plastered bales have a resistance of at least 2 hours 40 minutes under standard
construction material fire tests of up to 1000°C. Although these test results, combined
with tests and anecdotal evidence from other countries such as that produced by Apte
(2008), have been sufficient to allow construction of straw bale buildings within the UK
it is noted by Jones (2007) that a lack of funding has resulted in an inability to produce
a British Standard for the fire performance of lime plastered straw bales. This is
because the testing rig used was smaller than that required to obtain results to British
Standard level. A fire resistance of 135 minutes was used for the optimisation work;
this value was taken from data published by Modcell (Modcell 2012a).

78
4.8. Thin joint block work
4.8.1. Construction details

Thin joint block work is similar in construction to standard brick and block work. The
potential for differences between the two methods, in terms of Embodied energy due to
reduced mortar, increased construction acceptability due to perceived faster build time
and expected minimal impact on consumer acceptability due to similarities with brick
and block construction led to its inclusion in this work.

A double skin construction is used with an external leaf of mortared red brick. An air
cavity and insulation separate the leaves. The internal leaf is constructed from large,
lightweight blocks with a maximum of 3mm mortar between them. The mortar bed is
reinforced with a thin stainless steel or nylon mesh placed on the block before the
mortar is applied. This mesh compensates for the reduced strength of thin mortar.
Internally the walls are finished with plasterboard, a plaster skim and paint. The internal
leaf can be constructed prior to the external leaf. Wall ties between the two leaves are
inserted into the blocks after construction of the internal leaf as the external brickwork
is placed. This construction method is shown in Figure 4.5.

Figure 4.5- Axonometric view of thin joint block work construction

Party wall construction uses a thin mortared double layer of lightweight concrete blocks
with mortar reinforcement. A 75mm layer of mineral wool insulation is used between
the layers. On both sides the walls are finished with plasterboard, a plaster skim and
paint. The material specification used in this work is given in Table 4.15.

79
Table 4.15: Material specification for thin joint block work construction

Component Specification
Brick Fired clay red brick, 102.5 x 65 x 215mm
Mortar 1:3 cement sand mix, 10mm joints
2
Wall ties Stainless steel, 2.5 ties per 1m wall
Insulation 100mm mineral wool
Thin joint blocks Durox top block System 600, 100 x 299 x 620mm
Thin joint mortar 1:3 cement sand mix, 3mm joints
Reinforcement mesh Galvanised, 90mm wide, 1.5mm flat wire. Murfor
EFS/Z used
Plasterboard 12.5mm thickness
Plaster skim 6mm skim
Paint Double coat of water based paint
Party wall insulation 7mm mineral wool

4.8.2. U value calculation

Values for material thickness and thermal conductivity were taken from Tarmac
Topblock Ltd. (2006), British Standards Institution (2007), Anderson (2006) and
Chudley and Greeno (2005), these, along with the calculated value of thermal
resistance, are given in Table 4.16.

Table 4.16- L, λ and R values for thin joint block work components

L λ R
2
Component (m) (W/mK) (m K/W)
Mortared brickwork 0.1025 0.84 0.12
Air space 0.025 0.18 0.14
Mineral wool 0.1 0.038 2.63
Mortared block work 0.1 0.16 0.63
Plasterboard 0.0125 0.16 0.08
Plaster skim 0.006 0.57 0.01

The values from Table 4.16 and Table 3.1, adjusted for the effect of wall ties as
recommended by Chudley and Greeno (2004) give U = 0.272W/m2K for the thin joint
block work wall using the materials specified in 4.16.

80
The U value of the party walls is taken as 0W/m2K as it is fully filled and sealed (HM
Government, 2010). Therefore heat loss through the party walls is considered to be
zero.

4.8.3. Embodied energy

The embodied energy of thin joint block work walls was expected to be high as a result
of the energy intensive materials used. Calkins (2009) notes the high energy
requirements of both bricks and concrete blocks. Table 4.17 gives values of material
quantities and embodied energy coefficients for 1m2 of external thin joint block work
wall. Table 4.18 gives values of material quantities and embodied energy coefficients
for 1m2 of thin joint block work party wall.

The values for volumes of material are based on the specification given in Table 4.15,
values for the embodied energy coefficients were taken from the Inventory of Carbon
and Energy V2.0 database developed by Hammond and Jones (2011).

Table 4.17: Material quantities and embodied energy coefficients for 1m2 thin joint
block work external wall

2
Volume Density Mass EE EE per 1m
wall area
3 3
Component (m ) (kg/m ) (kg) (MJ/kg) (MJ)
Brick 0.085 1700 144.330 3.00 432.99
Mortar 0.018 1750 30.800 1.33 40.96
Wall tie N/A N/A 0.513 56.70 29.09
Insulation retainer N/A N/A 0.061 76.70 4.67
Block 0.099 600 59.280 3.50 207.48
Reinforcement mesh N/A N/A 0.622 22.60 14.06
Mortar 0.002 1750 2.348 1.33 3.12
Plasterboard 0.013 950 11.875 6.75 80.16
Plaster skim 0.006 1300 7.800 1.80 14.04
Paint N/A N/A N/A N/A 21.00
847.56

Summing the values from Table 4.18 gives an embodied energy value of 847.56MJ for
1m2 of wall using the thin joint block work system as specified in Table 4.16. Bricks are
the main contributor to the embodied energy. The lightweight blocks used also make a
significant contribution to the value as they have a higher value of embodied energy
per kilogram than standard blocks. However, the low mass of the blocks prevents this
value being too high.

81
Table 4.18: Material quantities and embodied energy coefficients for 1m2 thin joint
block work party wall

2
Volume Density Mass EE EE per 1m
wall area
3 3
Component (m ) (kg/m ) (kg) (MJ/kg) (MJ)
Paint N/A N/A N/A N/A 21.00
Plaster skim 0.006 1300 7.800 1.80 14.04
Plasterboard 0.013 950 11.875 6.75 80.16
Block 0.099 600 59.280 3.50 207.48
Mortar 0.002 1750 2.348 1.33 3.12
Reinforcement mesh N/A N/A 0.622 22.60 14.06
Wall tie N/A N/A 0.513 56.70 29.09
Insulation 0.075 25 1.875 16.60 31.13
Reinforcement mesh N/A N/A 0.622 22.60 14.06
Mortar 0.002 1750 2.348 1.33 3.12
Block 0.099 600 59.280 3.50 207.48
Plasterboard 0.013 950 11.875 6.75 80.16
Plaster skim 0.006 1300 7.800 1.80 14.04
Paint N/A N/A N/A N/A 21.00
739.92

The value of embodied energy calculated for 1 m2 of party wall built using the thin joint
block work system was 739.93MJ. Removing the brick skin caused a reduction in
embodied energy. However, including a second layer of lightweight blocks increased
the value. Lightweight blocks have a higher embodied energy per kilogram than
standard concrete blocks as a result of the additional manufacturing processes.

4.8.4. Airtightness, wall thickness and fire performance

The reduced number of joints resulting from the use of larger blocks was expected to
improve the airtightness achievable with this method when compared with the
traditional use of small blocks. Product information provided by the manufacturer
indicates an airtightness for the Durox blocks of 0.12m3/hr.m2 at 50Pa (Tarmac
Topblock Ltd. 2006). However, testing of a building constructed using the method as
specified in Table 3.16 showed an airtightness of 4m3/hr.m2 at 50Pa (Zero Carbon Hub
2011), this higher value was used in the optimisation as it is based on actual
construction values and includes all elements of the wall construction, not just the thin
joint blocks. This is lower than the value achieved by traditional brick and block
construction. It is important that the quality of workmanship is high to allow this value to
be achieved.

82
The value of wall thickness was calculated from Table 4.16. For the system as
specified for this work the wall thickness was 0.346m.

Fire performance of thin joint block work was expected to be similar to that shown by
standard brick and block construction. Information supplied by the manufacturer
indicates that 100mm thickness of System 600 blocks, as specified for this work, will
provide two hours of fire resistance for load bearing walls (Tarmac Topblock Ltd.
2006). A fire resistance time of 180 minutes was used during the optimisation
procedure.

4.9. Timber frame with brick cladding

Timber frame construction is the second most commonly used method of construction,
used for 7% of the houses constructed in England from 1990 to March 2009 (DCLG
2010). Historically this construction method experienced increased popularity during
the 1960s and 1970s, however, bad publicity resulting a television programme stating it
was at risk from rotting and fire severely damaged the public acceptability of this
construction method and reduced its use (Cavill 1999).

It was included as an alternative method that is considered to have had some level of
mainstream exposure.

4.9.1. Construction details

Timber frame construction uses a load bearing timber frame with external cladding and
infill to provide insulation and internal finishing.

For this work an external finish of mortared brickwork was used to maintain a
traditional appearance and provide the necessary waterproofing for the building. To
control the movement of moisture and heat the internal system is composed of an air
filled cavity, a waterproof membrane, wood boarding, insulation, a breathable vapour
control membrane and plasterboard, finished internally with a plaster skim and
paintwork. Details for this construction method are based on those provided by Knight
et al (2009) and Hastings (2010). This construction method is shown in Figure 4.6.

83
Figure 4.6- Axonometric view of timber framed construction

Party walls are constructed using a double thickness layer of frame and infill with
additional insulation between the two layers. Material specification used in this work is
given in Table 4.19, unless stated specification applies to external and party walls.

84
Table 4.19- Material specification for timber frame construction

Component Specification
Brick Fired clay red brick, 102.5 x 65 x 215mm
Mortar 1:3 cement sand mix, 10mm joints
Wall ties Stainless steel, plastic insulation retainer, 4.4 ties per
2
1m wall
Air space 60 mm ventilated
Waterproof membrane Glidevale Protect TF200 Breather Membrane
Wood sheet OSB sheet, 9mm thickness
Insulation Mineral wool insulation, 120mm between studs
Breathable membrane Glidevale VC foil ultra with cavity clips to reduce thermal
bridging
Timber frame 140 x 45mm timber studs at 500mm spacings
Services void 20mm, interrupted by studs to allow placement of cavity
clips. Services can also be run in these voids.
Plasterboard 25mm thickness, two sheets of 12.5mm
Plaster skim 6mm skim
Paint Double coat of water based paint
Party wall timber frame 90 x 45mm timber studs at 500mm spacings
Party wall cavity 60mm
Party wall insulation Mineral wool insulation, 70mm between studs, 60mm in
cavity

4.9.2. U value calculation

Values of material thickness and thermal conductivity for the timber framed
construction method shown in Table 4.20 were taken from Kingspan (2009), British
Standards Institution (2007), Anderson (2006) and Chudley and Greeno (2005). The
impact of the vapour barrier and waterproof membrane are ignored due to their low
thickness (1mm) as suggested by the manufacturer (Hastings 2010). The U value is
calculated for the insulated section of wall.

85
Table 4.20- L, λ and R values for timber frame construction components

Thermal Thermal
Thickness conductivity resistance
2
Component (m) (W/mK) (m K/W)
Mortared brickwork 0.1025 0.84 0.12
Air space 0.060 0.18 0.14
Wood sheet 0.009 0.03 0.30
Mineral wool 0.120 0.038 3.16
Services void 0.020 0.18 0.11
Plasterboard 0.025 0.16 0.08
Plaster skim 0.006 0.57 0.01

Values from Table 4.20 and Table 3.1 were summed then adjusted for the effect of wall
ties as recommended by Chudley and Greeno (2004) to give U= 0.249W/m2K for the
insulated section of timber framed construction as specified in this work.

Timber framed construction carries the risk of thermal bridging, where heat can be
transferred from inside the building, through the timber to the outside, bypassing the
insulation layer and resulting in significant heat loss. The selection of Glidevale VC foil
ultra with cavity clips as the vapour control membrane was intended to reduce this
issue to an acceptable level. The cavity clips cover the internal face of the timber studs
and reduce the heat transfer into the timber (Hastings 2010). Although some depth of
insulation is lost to allow for the clips the reduced thermal bridging is considered to be
of greater importance.

The specified construction of party wall enables the U value of the party wall to be
considered as 0W/m2K (HM Government, 2010). Therefore heat loss through the party
walls is considered to be zero.

4.9.3. Embodied energy

Table 4.21 gives quantities of material components for 1m2 of external timber framed
wall to the specification given in Table 4.19 and the associated embodied energy
values. Table 4.22 contains material and embodied energy values for 1m2 timber
framed party wall.

86
The values for volumes of material are based on the specification given in Table 4.19,
values for the embodied energy coefficients were taken from the Inventory of Carbon
and Energy V2.0 database developed by Hammond and Jones (2011).

Table 4.21- Material quantities and embodied energy coefficients for 1m2 timber
framed external wall

2
Volume Density Mass EE EE per 1m
wall area
3 3
Component (m ) (kg/m ) (kg) (MJ/kg) (MJ)
Brick 0.085 1700 144.330 3.00 432.99
Mortar 0.018 1750 30.800 1.33 40.96
Wall tie N/A N/A 0.903 56.70 51.19
Insulation retainer N/A N/A 0.107 76.70 8.22
Waterproof membrane N/A N/A 0.140 155.00 21.70
OSB sheet 0.009 680 6.120 15.00 91.80
Mineral wool 0.109 25 2.730 16.60 45.32
Breathable membrane N/A N/A 0.100 95.89 9.59
Timber studs 0.013 529 6.665 7.40 49.32
Plasterboard 0.025 950 23.750 6.75 160.31
Plaster skim 0.006 1300 7.800 1.80 14.04
Paint N/A N/A N/A N/A 21.00
946.45

Summing the values from Table 4.21 gives an embodied energy value of 946.45MJ for
1m2 of wall using the timber framed wall construction specified in Table 4.17. Brickwork
was the main contributor to the total embodied energy for this method of construction.
Using an alternative method of cladding could reduce the total embodied energy of the
construction.

87
Table 4.22- Material quantities and embodied energy coefficients for 1m2 timber
framed external wall

2
Volume Density Mass EE EE per 1m
wall area
3 3
Component (m ) (kg/m ) (kg) (MJ/kg) (MJ)
Paint N/A N/A N/A N/A 21.00
Plaster skim 0.006 1300 7.800 1.80 14.04
Plasterboard 0.025 950 23.750 6.75 160.31
Breathable membrane N/A N/A 0.100 95.89 9.59
Timber studs 0.008 529 4.285 7.40 31.71
Mineral wool 0.064 25 1.593 16.60 26.44
OSB sheet 0.009 680 6.120 15.00 91.80
Mineral wool 0.090 25 2.250 16.60 37.35
OSB sheet 0.009 680 6.120 15.00 91.80
Mineral wool 0.064 25 1.593 16.60 26.44
Timber studs 0.008 529 4.285 7.40 31.71
Breathable membrane N/A N/A 0.100 95.89 9.59
Plasterboard 0.025 950 23.750 6.75 160.31
Plaster skim 0.006 1300 7.800 1.80 14.04
Paint N/A N/A N/A N/A 21.00
747.12

The value calculated for the embodied energy of 1m2 of timber framed party wall was
747.12MJ. The removal of the brick outer skin reduced the embodied energy.
However, the second layer of insulated timber frame required to construct a suitable
party wall caused a significant increase in the value.

4.9.4. Airtightness, wall thickness and fire performance

High quality of workmanship is particularly important in achieving an airtight building

with timber framed construction. The nature of the method, with multiple layers and
components increases the potential for gaps which will allow the passage of air.

The selection of a vapour barrier with cavity clips helps to improve the airtightness of
the system by creating a continuous layer within the wall. Joints must be sealed with
tape to achieve this (Hastings 2010). The estimated airtightness for a timber framed
wall following the above specification is 3 to 5m3/hr.m2 at 50Pa (Glidevale 2012).
Hastings (2011) gives an example of a perfectly sealed system achieving 3m3/hr.m2 at
50Pa in testing. The higher value of 5m3/hr.m2 at 50Pa has been used in the
optimisation process as it was considered achieving a perfect seal in site conditions
was unlikely.

88
To calculate wall thickness the component thicknesses given in Table 4.19 were
summed. This gives a value of 0.295m for the wall constructed to the specification
used in this work.

The presence of wood in timber framed construction is often considered to present a

high fire risk. The fire resistance of timber framed buildings is discussed by Lennon et
al. (2000) with the conclusion that good workmanship is required to achieve good fire
performance. The testing of the TF2000 timber framed housing discussed by Lennon
et al (2000) showed that timber framed construction meets the Building Regulations
fire performance standard of 60 minutes, this value was used in the optimisation. Two
layers of plasterboard, as specified in Table 4.17 are typically required to achieve this.

4.10. Summary

Six construction methods were chosen to be studied in detail- brick and block,
structural insulated panels (SIPs), insulating concrete formwork (ICF), straw bale and
lime plaster panels, thin joint block work and timber framed construction. Brick and
block construction was included as the “typical method”. Alternative techniques were
selected to cover a range of available options. All methods considered had been
previously used in England or Wales, meaning they are capable of passing the
Building Regulations for England and Wales and can be constructed and maintained in
the temperate climate experienced. Details of the method and performance associated
with each of the six construction methods have been given. Values for U value,
embodied energy and performance for the construction method specifications studied
are summarised in Table 4.23.

89
Table 4.23- U values, embodied energy and performance values for each of the
methods of construction and Building Regulations

(m /(h.m ) at 50Pa)
Embodied energy

Embodied energy

Fire performance
Wall thickness
Construction

External wall

Air tightness
Party wall

(minutes)
2
(W/m K)
method

U value
2

(MJ)

(MJ)

(m)
Brick and block 0.277 744.30 713.46 4.50 0.346 240
SIP 0.167 1464.12 834.77 1.00 0.326 73
ICF 0.204 1448.31 1060.47 1.00 0.449 90
Prefabricated bale unit 0.190 458.61 458.61 0.86 0.480 135
Thin joint block work 0.272 847.56 739.92 4.00 0.346 180
Timber frame 0.249 946.45 747.12 5.00 0.295 60
Building Regulations 0.300 N/A N/A 10.00 N/A 60

A visual comparison of Table 4.23 shows that all the options perform better than the
requirements of the Building Regulations where applicable. A comparison of the
alternative methods of construction with the “typical” construction method of brick and
block and each other is difficult to achieve visually as each material performs better on
some elements, but worse on others. Because of this the optimisation carried out later
in this work (Chapters 7 and 8) is particularly suitable for identifying the best option.
Identifying the values in Table 4.23 enabled the calculation of embodied energy,
operational energy and performance characteristics during the optimisation procedure.

It can be seen from the embodied energy calculation table for each construction
method that bricks contribute significantly to the total embodied energy for those
methods that include them. It is possible that alternative methods of cladding would
reduce the embodied energy of construction. However, if a brick finish is desired other
methods of cladding may be unacceptable. Materials such as brick slips, which provide
a brick appearance but use significantly less material and hence have a lower
embodied energy, may be a solution to this issue. This has been identified as an area
for further work as this work focused on maintaining a brick appearance where
possible. Foam insulation, such as that used in SIPs and ICF has a higher than
expected embodied energy, however, these methods of construction achieve high
performance in terms of U value and airtightness, therefore the embodied energy cost
can be balanced by the potential operational energy savings.

90
5. Public acceptability of construction methods
5.1. Introduction

It was considered important to determine the public acceptability of alternative

construction methods. An option that was not supported by the public would be unlikely
to be a successful construction option, even though it may have benefits such as low
embodied energy or good thermal performance. In this chapter the acceptability of the
end user, the public who will live in the finished houses was examined. The opinions of
the public on the construction methods being studied were examined by the use of a
survey. Views were collected on factors which the public consider to be important
when choosing a house, as well as the acceptability of each method.

5.2. Questionnaire results and discussion

The view of house purchasers was considered to be of particular importance to this

work. In order for a type of housing construction to present an economical solution it
must be acceptable to the purchaser. If a house is constructed using a method in
which people are not prepared to invest it is unlikely to sell. Even if such a house
performs well in terms of embodied and operational energy it cannot be considered to
be a viable solution. The survey aimed to evaluate this acceptability and to identify
factors about which purchasers were concerned when selecting a house.

5.2.1. Response rate

A total of 593 responses were received. Of these, twenty one responses were fully
blank, with only the permission to use information question completed; these were
removed. For the analysis 573 responses were used; partial responses were retained
as it was felt they could provide information about those questions which were
completed.

5.2.2. Geographical representativeness of responses

The distribution of respondents by region was examined and compared with the
geographical population distribution for England and Wales. This is shown in Figure
5.1.

91
 25
Percentage in England
and Wales
20
Percentage of population (%)
Percentage in survey
responses
15

10

Figure 5.1- Percentage distribution of respondents and population of England and

Wales by region

Visual inspection of Figure 5.1 indicates a higher percentage of responses from the
East Midlands, South East and South West than exist in the actual population, with
lower than actual values for all other regions. A Chi squared test of the values shows
the distributions are not sufficiently similar for the values to be used nationally (Χ2=
39.88, Χ0.952= 16.92. Therefore, sampling was carried out within the survey responses
to mirror the geographical distribution of the population. The sub-samples created were
used for statistical analysis of the responses regarding importance of factors when
choosing a house. As each region is considered separately for acceptability it was not
necessary to sample within the data for this.

5.2.3. Sampling of data

Sampling was carried out from within the survey responses to create subsamples that
correctly mirrored the geographical distribution of England and Wales.

SPSS was used to draw five different samples from the data; the use of multiple
samples allowed a comparison between the results. The number of respondents from

92
each region was calculated to mirror the geographical distribution of population in
England and Wales.

It was desirable to maximise the number of responses in the sample; therefore, the
region with the smallest population to total response ratio was used as the limiting
region. This was identified as Wales; all subsamples used all responses from Wales
and a reduced number from each of the other regions.

To calculate the number of samples required from each region, the number of
responses for Wales was divided by the percentage of the actual population who live in
Wales. This value was multiplied by the percentage in the population for each region.
All values were rounded to the nearest whole number, as survey numbers are a
discrete value. The number of responses from each region is shown in Table 5.1.

Table 5.1- Number of responses taken from each region to form the sample

Percentage of
England and Total number Number of Percentage
Wales of survey responses of the
Region population responses in sample sample

East Anglia 10.35 36 17 10.35

East Midlands 8.02 110 13 8.02
London 13.78 38 22 13.78
North East 4.83 10 8 4.83
North West 12.93 52 21 12.93
South East 15.37 96 25 15.37
South West 9.47 96 15 9.47
West Midlands 10.12 43 16 10.12
Yorkshire and Humberside 9.54 39 15 9.54
Wales 5.58 9 9 5.58
Totals 100 529 161 100

Five subsamples containing the correct number of responses from each region to
mirror the actual population distribution of England and Wales were drawn randomly
from the survey responses. Analysis of importance factors was carried out on these
subsamples.

93
5.2.4. Importance of characteristics in house purchase

The importance of each factor when considering buying as house is displayed as a

percentage score in Figure 5.2. A score of 1 was equivalent to not at all important, a
score of 5 meant very important.

80

70

60

50
Response (%)

40

30

20

10

0
1 2 3 4 5
Score
Fire risk Price

Low energy use Need for maintenance

Price of insurance Mortgage availability

Environmental impacts of construction

Figure 5.2- The importance of each factor when considering a house purchase, from
Sample 1.

Values are shown for Sample 1 only as the graphs produced for all samples were
similar.

The most frequent value (modal) and the middle value when all responses were
arranged in numerical order (median) values are given in Table 5.2. These values were
the same for all samples.

94
Table 5.2- Modal and median values of importance for the characteristics considered

Importance score

Characteristic Mode Median

Fire risk 3 3
Price 5 5
Low energy use 4 4
Need for maintenance 4 4
Price of insurance 3 3
Mortgage availability 5 4
Environmental impacts
3 3
of construction

From visual inspection of the graphs and consideration of the modal and median
values, the factors can be given the following importance ranking (most to least):

1) Price
2) Need for maintenance
3) Mortgage availability
4) Low energy use
5) Price of insurance
6) Environmental impacts of construction
7) Fire risk

It can be seen from Figure 5.2 that all factors have some importance to the public. The
impact of any changes such as wall construction method on these should be
considered carefully as a positive impact, such as reducing price, may result in good
public support for the method. A negative impact, for example increasing price, will
reduce the attractiveness of the method and the potential for its use.

Principal component analysis was used to reduce the seven categories scored using
Likert scales into a smaller number of themes. These identify areas about which the
public are concerned, for example financial issues, rather than focusing on specific
aspects, such as house prices. Connections between the factors are identified by the
use of a rotated component matrix.

For each of the five samples drawn from the data, principal component analysis was
carried out on the seven sets of data produced by the Likert scales for the
questionnaire. This allowed linked factors to be grouped into a smaller set of issues

95
that should be considered when designing houses. Each question asked was
considered to be a variable. To determine the suitability of the data from the
questionnaire for principal component analysis, two tests are carried out, the Kaiser-
Mayer-Olkin measure and Bartlett’s test of spherecity. The Kaiser-Mayer-Olkin
measure for each sample was in excess of 0.6, which is considered acceptable (Field
2009). The correlation of items was tested using Bartlett’s test of spherecity, the values
for this are shown in Table 5.3.

Table 5.3- Values for Bartlett’s test of spherecity for each sample

Sample Values for Bartlett’s test of spherecity

2
1 χ (21)= 169.703, p < .001
2
2 χ (21)= 213.911, p < .001
2
3 χ (21)= 194.821, p < .001
2
4 χ (21)= 212.300, p < .001
2
5 χ (21)= 214.802, p < .001

Values of p < .05 indicate there is a relationship between the variables that can be
explored by principal component analysis (Field 2009). The values in Table 5.3 show
that all samples satisfy this requirement, therefore, the data is suitable for this analysis
as there is a significant relationship between the variables.

Eigenvalues for each “theme”, or component, were calculated. However, only those
considered to be significant are retained. This dictated by the minimum eigenvalue that
is accepted. All samples produced two factors with eigenvalues greater than 1, which
would satisfy Kaiser’s criterion, however, it was decided to use Joliffe’s criterion of
factors with eigenvalues above 0.7 being retained (Field 2009). Using Joliffe’s criterion
resulted in four factors for all subsamples and created a more logical division of
categories when examined. The four components are initially labelled 1-4;
determination of their meaning is carried out after examination of the results.

Scree plots can be used to examine the eigenvalues for each component identified. A
plot of the eigenvalue calculated against the component number can be used to
identify the number of components to retain. Typically such plots will fall sharply and
demonstrate a noticeable point of inflexion. Components with eigenvalues greater than
this point of inflexion are retained; those with eigenvalues below the point of inflexion
below are discarded. The point of inflexion itself may be retained or discarded,
depending on preference and the results achieved (Field 2009). Scree plots were

96
generated for each sample in this work. However, they produced ambiguous results,
with the point of inflexion difficult to identify. Therefore, Joliffe’s criteria was used.

When a principal component analysis is carried out it is desirable that the components
identified account for the maximum possible variance. The variance represents the
spread of the data. Each component identified will account for some percentage of the
variance. The cut off point for the number of components should be such that an
acceptable amount of variance is accounted for. The variance accounted for by the use
of four factors for each sample was in excess of 78% for all five samples, considered to
be acceptable.

To improve the clarity of results by making each variable impact most strongly on a
single component, rather than on many, a technique called rotation can be used. The
results are referred to as the rotated component matrix. For this work Orthagonal
rotation (varimax) was used (Field, 2009). The results of the principal component
analysis for Samples 1 to 5 are shown in Tables 5.4 to 5.8 respectively.

97
Table 5.4- Rotated component matrix for Sample 1 Likert scale principal component
analysis

Component
Question 1 2 3 4
When considering a house for purchase
how important are the environmental
impacts of its construction to you? 0.868
When considering a house for purchase
how important is low energy use to you? 0.785
When considering a house for purchase
how important is mortgage availability to
you? 0.861

When considering a house for purchase

how important is price to you? 0.793
When considering a house for purchase
how important is fire risk to you? 0.855

When considering a house for purchase

how important is the price of insurance to
you? 0.698 0.483
When considering a house for purchase
how important is the need for
maintenance to you? 0.916

Consideration of the variables which affect each component leads to the components
being identified as:

 Component 1- Environmental factors

 Component 2- Financial factors
 Component 3- Risk
 Component 4- Running costs

98
Table 5.5- Rotated component matrix for Sample 2 Likert scale principal component
analysis

Component
Question 1 2 3 4
When considering a house for purchase
how important is the need for
maintenance to you? 0.811

When considering a house for purchase

how important is the price of insurance to
you? 0.773
When considering a house for purchase
how important is fire risk to you? 0.644
When considering a house for purchase
how important are the environmental
impacts of its construction to you? 0.913
When considering a house for purchase
how important is low energy use to you? 0.795
When considering a house for purchase
how important is mortgage availability to
you? 0.933
When considering a house for purchase
how important is price to you? 0.960

Consideration of the variables which affect each component leads to the components
being identified as:

 Component 1- Risk and associated costs

 Component 2- Environmental factors
 Component 3- Mortgage availability
 Component 4- Purchase price

99
Table 5.6- Rotated component matrix for Sample 3 Likert scale principal component
analysis

Component
Question 1 2 3 4
When considering a house for purchase
how important are the environmental
impacts of its construction to you? 0.811
When considering a house for purchase
how important is low energy use to you? 0.773
When considering a house for purchase
how important is mortgage availability to
you? 0.644
When considering a house for purchase
how important is price to you? 0.913
When considering a house for purchase
how important is the price of insurance to
you? 0.795
When considering a house for purchase
how important is fire risk to you? 0.933
When considering a house for purchase
how important is the need for
maintenance to you? 0.960

Consideration of the variables which affect each component leads to the components
being identified as:

 Component 1- Environmental factors

 Component 2- Financial factors
 Component 3- Risk
 Component 4- Maintenance

100
Table 5.7- Rotated component matrix for Sample 4 Likert scale principal component
analysis

Component
Question 1 2 3 4
When considering a house for purchase
how important is fire risk to you? 0.720

When considering a house for purchase

how important is price to you? 0.973
When considering a house for purchase
how important is low energy use to you? 0.811
When considering a house for purchase
how important is the need for
maintenance to you? 0.915
When considering a house for purchase
how important is the price of insurance to
you? 0.735
When considering a house for purchase
how important is mortgage availability to
you? 0.930
When considering a house for purchase
how important are the environmental
impacts of its construction to you? 0.865

Consideration of the variables which affect each component leads to the components
being identified as:

 Component 1- Environmental factors

 Component 2- Running costs
 Component 3- Mortgage
 Component 4- Price

101
Table 5.8- Rotated component matrix for Sample 5 Likert scale principal component
analysis

Component
Question 1 2 3 4

When considering a house for purchase

how important are the environmental
impacts of its construction to you? 0.888
When considering a house for purchase
how important is low energy use to you? 0.819
When considering a house for purchase
how important is the price of insurance to
you? 0.857
When considering a house for purchase
how important is fire risk to you? 0.742
When considering a house for purchase
how important is price to you? 0.861
When considering a house for purchase
how important is mortgage availability to
you? 0.814
When considering a house for purchase
how important is the need for
maintenance to you? -0.636

For the results of Sample 5 maintenance achieved a negative value. As a result of this
it was removed from the analysis of this sample. Consideration of the variables which
affect each component leads to the components being identified as:

 Component 1- Environmental factors

 Component 2- Risk
 Component 3- Price
 Component 4- Mortgage

The four components identified from each of the five samples above were compared to
identify the “themes” about which it was considered the public are concerned when
considering a house for purchase. Energy use and environmental impact of
construction are always grouped into a single factor; therefore, it is considered that
environmental factors can be considered as a “theme” about which the house
purchasing public are concerned. Purchase price and mortgage appear either grouped
together or individually, with no other factors. Therefore, a second “theme” about which
it can be said the purchasing public is concerned is financial factors. Other elements

102
appear grouped in a variety of ways across the samples; they all indicate a level of
concern for different types of risk and the associated costs. Risk management is
considered to be the best way to categorise the remaining elements into a single
“theme” about which the house purchasing public is concerned. Environmental factors,
financial factors and risk should all be considered when making decisions about
housing construction. At this point in the survey the questions were about house
purchase in general, rather than the specific topic of wall construction method.
Therefore these themes can be applies to any housing related decision.

5.2.5. Awareness of alternative construction methods

The respondents were asked if they had heard of each of the six construction types
being considered. At this stage no further details were given about the methods. The
results from this question are shown in Figure 5.3.

100
90
80
Yes
70
Response (%)

60 No
50
40
30
20
10
0
Brick and SIPs ICF Straw bale Thin joint Timber
block and lime block work frame with
plaster brick
cladding
Construction method

Figure 5.3- Percentage of respondents who have heard of each type of construction.

There is a high level of awareness of brick and block amongst the public; this was
expected as it is the most commonly used housing construction method. For example,
brick and block was used to construct 88% of dwellings in England between 1990 and
March 2009 (DCLG, 2010).

Timber frame had the second highest level of awareness. A high level of awareness
was expected as timber frame is the second most common method of housing

103
construction. From 1990 to March 2009 7% of houses built in England were built using
timber frames (DCLG, 2010).

Awareness of straw bale and lime plaster construction was higher than expected. Later
in the questionnaire a number of respondents made comments about having seen this
type of construction on the Channel Four television programme “Grand Designs”. It is
felt this, combined with the unusual nature of the method making it memorable, has
increased public awareness. ICF and SIPs have both also featured on “Grand
Designs”, but have a lower awareness.

SIPs, ICF and Thin joint block work all have low levels of awareness. It is possible that
as no illustration of the wall types was given when this question was asked in the
survey a number of respondents may be aware of the construction technique, for
example through media exposure, but be unfamiliar with the name. It would be
desirable to increase the awareness of these methods through education as this is
viewed as the first step to their potential increased usage.

5.2.6. Acceptability of Brick and block

The acceptability of each method is considered by region as the data indicated the
respondents' regional structure was not a reflection of that found in England and
Wales. Visual inspection of the acceptability data when sorted by region indicated
there was a strong variation in the acceptability of construction methods by location
which supported the argument for considering the data for each region separately. In
addition, later use of the data in optimisation procedures required a regional separation
to make it compatible with other data. Comments made expressing views and
concerns about each option were not divided by region as it was felt they could be
applied nationally.

In the survey, the question regarding the acceptability of brick and block was presented
first with the intention that later alternative options would not bias respondents against
brick and block by feeling they may have been shown a better option. It also allowed
respondents to familiarise themselves with the layout of the questions and diagrams on
a type of construction that it was more likely they would recognise.

Percentage responses to the question “Would you buy a house built from this
material?” for the brick and block question are shown in Figure 5.4. Numerical values
for the responses can be found in Appendix C. Answers to this question are sorted by

104
region; no subsamples were taken as acceptability remained divided by region
throughout the work.

100

90

80

70
Yes
Response (%)

60
No
50
Maybe
40

30

20

10

Figure 5.4- Acceptability of brick and block construction by region

Figure 5.4 shows that brick and block has a high acceptability for purchase in all
regions, with the highest in the West Midlands at 98%. A high acceptability for this
construction method was expected as it is the most commonly used type of housing
construction; used for 88% of dwellings in England between 1990 and March 2009
(DCLG, 2010). This data was intended to act as a control group, enabling acceptability
of the “typical” method to be compared with that of alternative options. It also
demonstrates that there is not 100% acceptability of brick and block in any region.

The lowest acceptability for this type of construction was seen in Wales and the North
East; these values may have been artificially lowered by fairly low response rates from
these regions. All other regions had “yes” responses in over 85% of cases.

Of the remaining respondents, “maybe” answers make up the majority, with very few
“no” responses seen in any region. The highest values of “no” responses were seen in

105
the North East and Wales. Again this may have been impacted by low response rates
for these regions.

The acceptability question was followed by an optional open question allowing

respondents to add any comments they wished regarding the construction type. These
comments have not been divided by region as it is felt the views expressed can be
applied nationally.

The most common comment relating to brick and block construction was that it was the
typical housing method. The high acceptability combined with the number of comments
describing brick and block as “typical” or “traditional” indicates that the respondents
have a significant preference for wall types with which they are familiar.

Factors such as insulation, strength and durability received positive comments. The
fact that it is a proven method was also presented as a positive feature by a number of
respondents. The high acceptability of this type of construction may indicate that these
are factors which must be emulated by alternative methods of construction in order for
them to be accepted.

Few negative comments were presented by respondents for this method of

construction. The respondents who said they would not purchase this type of building
expressed a preference for other types of construction, such as stone, or a dislike of
brick as an exterior material. Alternative exterior skin materials may have been
acceptable to these respondents; this provides an opportunity for alternative
construction methods such as SIPs and ICF, where a range of finishes can be
achieved more easily than with a dual skin system.

106
5.2.7. Acceptability of Structural Insulated Panels (SIPs)

Percentage responses to the question “Would you buy a house built from this
material?” for SIPs are shown in Figure 5.5, divided by region. Numerical values can
be found in Appendix C.

100
90
80
70
Yes
Response (%)

60
No
50
Maybe
40
30
20
10
0

Figure 5.5- Acceptability of SIP construction by region

Compared with brick and block, SIPs have a significantly lower acceptability, ranging
from 43% to 69%. However, the percentage of “maybe” responses is much higher, with
a range of 6-16%. If the issues which result in a “maybe” response, rather than a “yes”
response could be solved then there is the potential for this method to move from a
medium to a high level of acceptability. The level of “no” responses ranges from 6% to
14%, indicating the method is unlikely to be as universally accepted as brick and block
even if the perceived issues can be solved. These results may relate to the lack of
awareness indicated in 5.2.4, those respondents who had not previously heard of the
construction method may have selected “maybe” or “no” due to a lack of prior
knowledge.

SIPs received positive comments in two areas. Respondents felt that the appearance
of the finished building would be similar to that achieved by brick and block
construction and indicated that this was a benefit. As with brick and block, a few

107
respondents indicated a dislike of brick as a finish. Although not discussed in the
survey, SIPs can be finished in a range of ways, which may appeal to those who
indicated they would not purchase this type of construction because they disliked the
appearance of brickwork. The presence of insulation in the construction method and
the associated reduction in energy use resulting from this was the other positive aspect
raised by respondents.

In contrast to brick and block, a number of negative points were raised. This was
expected due to the unusual nature of the construction method, and presents an
opportunity to see which issues concern the respondents. The most frequently
expressed negative point was that the method would not be sufficiently strong to
perform the duties of a house wall, in particular when compared to the previously
viewed option of brick and block. The strength of SIPs has been tested, both in
laboratory conditions and by use in existing structures, it is found to be better than that
for comparable timber frame structures in work done by Mosey et al. (2009), Hairstans
and Kermani (2007) and Morley (2000).

Durability, in particular with respect to the effects of moisture on the timber elements of
the construction, was the second most commonly stated negative point. It is noted by
the Department for Communities and Local Government (DCLG) BRE authors (2008)
that there is little long term evidence for the durability of SIP construction in the UK.
However, anecdotal evidence for the durability of SIPs exists in the USA, where early
examples of SIPs from the 1950s are cited by Morley (2000) and Cathcart (1998).
More recent examples which are still in use after thirty years are cited by Hairstans and
Kermani (2007). These examples suggest a significant durability; however, the
presence of timber will present a risk that must be managed by the use of suitable
finishes and good workmanship.

The presence of timber in the structure led to concerns regarding the fire performance
of the construction method. Barista (2008) and Griffen et al. (2006) note that timber
elements of the construction method result in a susceptibility to fire. To achieve the
necessary level of fire protection, it is necessary to use a plasterboard lining. Bregulla
and Enjily (2004) state that this allows SIPs structures to satisfy Part B of the England
and Wales Building Regulations. Good workmanship must be achieved to ensure this
is the case, as gaps would greatly reduce the fire performance of the structure.

Some respondents felt that this type of construction would result in difficulties attaching
fixtures and fittings to the walls. In fact, the OSB skin allows for placement of small
loads such as picture hooks and light weight shelving across the face, giving a greater

108
flexibility. However, it is noted by DCLG BRE authors (2008) that it is desirable to
minimise damage to the OSB layer as damage potentially reduces its strength. Larger
items, such as kitchen units, should be attached to framing or freestanding, rather than
fixed to the OSB.

5.2.8. Acceptability of Insulating Concrete Formwork (ICF)

Percentage responses to the question “Would you buy a house built from this
material?” for ICF are shown in Figure 5.6, divided by region. Numerical values for the
results can be found in Appendix C.

100
90
80
70
Yes
Response (%)

60
No
50
Maybe
40
30
20
10
0

Figure 5.6- Acceptability of ICF construction by region

The acceptability for purchasing a house built using ICF is similar to that of SIPs, with
fewer “yes” responses, but more “maybe” responses than seen with brick and block.
There appears to be a regional variation, with the “yes” percentages ranging from 44%
to 72%, the lowest in Wales and the highest in the West Midlands. This provides an
argument for considering construction methods by region rather than adopting a
national view. As with other questions, the low questionnaire response rates from
Wales and the North East may have skewed the results for these regions, however,
they have similar rates to those seen in other areas such as East Anglia and the South

109
East. Results seen for the acceptability of ICF might be connected the level of
awareness of the method shown in 5.2.4. As with SIPs, this construction method had a
low level of awareness, which may have resulted in more “maybe” or “no” responses
as respondents were not familiar with the method.

In the open ended question, the respondents were divided in their opinions on the
durability of ICF. Some felt that the concrete would produce a durable structure,
whereas others were concerned about the lifespan, in particular with respect to the
foam insulation. Past examples from the USA provide anecdotal evidence for
durability, demonstrating a lifespan of thirty years (ICFA, 2003). Predictions for lifespan
vary from sixty years (ICFA 2009) to two hundred years (Griswold, 2007) which is
comparable with traditional construction. Another divided issue was the potential for
attaching fixtures and fittings to walls made from this type of construction. Some
respondents felt that the concrete would easily support any loads applied, whereas
others were concerned about the ease of fixing items to the walls.

Positive factors commented on by questionnaire respondents included the presence of

insulation; the strength of the construction resulting from the use of concrete; and the
appearance. That the finished building looks like brick and block construction was
given as a positive aspect. In a few cases respondents indicated a dislike of a brick
finish, as with SIPs, it is possible to use alternative finishes with this construction
method to produce a range of appearances to suit the consumer. This has the potential
to increase acceptability.

The main concern expressed by respondents was that the use of concrete has a
negative environmental impact. Although the production of cement requires large
amounts of energy and results in the release of carbon dioxide, the environmental
impacts should be balanced with other factors such as lifespan and performance data.
In addition, it is possible to reduce the negative environmental impacts of concrete
production by the use of alternative component materials such as fly ash and recycled
aggregates (Calkins 2009; Johnston and Gibson, 2008; Anink et al., 1996).

The potential for ICF to be more expensive than other construction methods was
noted. Research by Denzer and Hedges (2007) and Al-Homoud (2005) indicates that
ICF is indeed more expensive than other forms of construction, quantified as 5-20%
more by Griswold (2007). As with environmental impacts, the additional cost of
construction should be balanced against performance data to determine if the initial
investment is economical. However, unlike environmental issues which are balanced

110
across the entire lifespan, the savings must balance the increased purchase price in
the potentially shorter time period of ownership, which may only be a few years.

The thickness of walls constructed from ICF was also a concern to some respondents.
Other methods of finishing, such as rendering, have the potential to reduce the wall
thickness as the air gap and brick skin would be eliminated. This was not examined in
the work as it was decided to maintain a traditional finish with the methods examined
where possible.

5.2.9. Acceptability of Prefabricated straw bale and lime plaster panels

Percentage responses to the question “Would you buy a house built from this
material?” for straw bale and lime plaster construction are shown in Figure 5.7, divided
by region. Numerical values are given in Appendix C.

100
90
80
70
Yes
60
Score (%)

No
50
Maybe
40
30
20
10
0

Figure 5.7- Acceptability of Prefabricated straw bale construction by region

Straw bale construction received the greatest proportion of “no” responses of all the
construction methods examined, with a range of “yes” responses at only 11-40%. This
option scored particularly low in London and Wales; a low acceptability was expected
in London due to its urban nature, however, Wales, being more rural, was expected to

111
have a higher acceptability. As with other methods, this may have been a result of an
insufficient response rate or due to other, unexplored, factors for example, farming in
Wales tends to be less crop-focussed due to the terrain.

The South West was the only region to have more “yes” responses than “no”. It is
thought that this indicates a significant regional divide in acceptability, and further
supports the case for considering each region separately. It is considered that this
greater acceptability may be a result of the South West being a rural area, with a large
amount of farming, which may have made a more easily identifiable connection
between straw bale construction and support for the local economy.

The levels of “maybe” responses is not dissimilar to that seen for SIPs and ICF,
supporting the case that a large number of purchasers would be interested in
alternative methods of construction, of all sorts, even methods perceived as unusual, if
purchasers’ concerns are dispelled. A large number of comments were given in the
open ended question for this method which provides valuable information on the
concerns held by the respondents. Identifying these concerns and addressing them
has the potential to greatly improve the acceptability of the method, in particular for
those respondents who answered “maybe”.

Comments indicated that opinion was divided over the appearance of lime plastered
straw bale construction. This was the only option in the survey that did not have a
brickwork appearance, as the typical construction method does not allow for this.
Views on appearance are largely a matter of personal preference; however, if a greater
range of finishes was available, then the acceptability might increase. Lime based
plaster is typically considered an integral part of the straw bale construction and the
most effective solution to weatherproofing, whilst still allowing moisture to exit the wall.
A solution may be to create a “fake brick” finish using coloured plaster. However, this
has significant implications for increasing costs and maintenance requirements.

Several comments indicated respondents had seen straw bale construction on the UK
Channel Four Television programme “Grand Designs”. This can be credited with
raising awareness of the method and some of its advantages, as the majority of
respondents reflected that their views were positive from this experience. Positive
comments on the use of straw in house building mainly focused on the environmentally
friendly nature of the construction method. The lower impact of the materials used
compared to those in other types of construction was commented on. The word
“sustainable” was used frequently by respondents who identified the rapidly renewable
nature of straw and perceived it as a benefit. The potential for straw bales to act as a

112
good insulator was commented on by many respondents, which has implications for
both environmental and financial factors. Positive financial implications were identified
with respect to cost, in that it may be a cheaper construction option than the others
presented due to the use of low-price components.

Straw bale construction received a large number of comments classed as negative,

either because they expressed a direct dislike or because they indicated a concern that
the respondent held regarding the method. This was expected, due to the unusual
nature and the limited exposure the method has had, particularly in relation to
performance details. The “Three Little Pigs” story, in which the house built from straw
is blown down, received a number of mentions; fortunately this was balanced to a
degree by more recent media exposure. However, neither gives a great deal of reliable
performance data to the mildly interested consumer. To obtain quality information, it
would be necessary to research the method, adding an extra step to the already long
process of house purchase.

The main focus for concern when respondents considered straw bale construction was
fire performance and the perception that straw is highly susceptible to fire damage. In
the case of plastered straw bale, the fire resistance of plastered straw bale is given by
Jones (2007) as two hours and forty minutes which satisfies the England and Wales
Building Regulations for fire performance. The tightly-packed nature of the bales
combined with the plaster skin limits the air available for combustion and reduces the
risk of burning. However, to convince people of this would be a substantial task. The
greatest fire risk is during the construction phase as loose straw will burn fairly easily
and there is a greater risk with unplastered bales. Good site practices should minimise
the risk and this should not affect the end user.

Another frequent area of concern was the durability of the walls, in particular with
respect to moisture and the risk of bales rotting. This is a genuine concern, as high
levels of moisture can result in rotting as discussed by Carfrae et al. (2009), Lawrence
et al. (2009), Goodhew et al. (2007), Summers (2006), Goodhew et al. (2004) and
Canada Mortgage and Housing Corporation (2000b). To limit the risk of this, it is
essential that good workmanship and quality control is maintained throughout the
construction phase. The impact on the end user should be minimal, although there
would be the option of monitoring the moisture content of the wall using moisture
meters. In the USA straw bale buildings between fifty and one hundred years old exist,
they are presented by Seyfang (2010), Smith (2007) and Yates (2006) as evidence of
durability. In the UK there is much less of a tradition of using straw bales in

113
construction; Jones (2007) states that the earliest examples date from 1994. Although
this is not as long as the evidence from the USA, it still demonstrates the durability of
the method if good construction practices are followed. The possibility of damage by
pests was another way respondents suggested durability could be compromised;
however, Jones (2007) indicates there is minimal risk of this as the straw contains little
nutritional value and once the bales are plastered access for pests is difficult.

The third most frequent concern was that the bales would have insufficient strength to
support applied loads. In the prefabricated unit, construction loads are carried by a
combination of the timber and plaster, with the straw acting largely as infill and
insulation. However, King (1997) demonstrates that even without the timber frame the
composite nature of the bales and plaster skin are sufficient to carry the loads. The
strength of plastered bales when used in this way has also been examined by the
Canada Mortgage and Housing Corporation (2008); their conclusion was that it is
possible to carry the loads associated with construction even when quite low quality
bales are used.

Wall thickness was noted by a few respondents. The nature of the construction means
that thick walls are unavoidable; however, thick walls provide the opportunity for
architectural features such as window seats, as demonstrated by Lacinski and
Bergeron (2000). In addition, as the thickness of insulation in brick and block
construction is increased to achieve better U values, the difference in wall thickness
reduces. One respondent stated that the additional thickness would make the method
unsuitable for use in London where space is a premium. This manner of regional
variation should be considered as in other areas wall thickness may be less of a
concern.

Issues raised by smaller numbers of respondents included a high need for

maintenance and potential difficulty relating to financial matters. The unusual nature of
the construction means it is more difficult to obtain mortgages and insurance for the
building. Specialist companies will provide the required financial assistance, however
the lack of competition has the potential to increase costs. Since price and other
financial issues rate highly in consumer concerns, this is an issue that would need to
be addressed to make this method of construction viable on a large scale. It is true that
the lime plaster requires a greater level of maintenance than other construction
methods as the external surface must be repainted to maintain its condition. This is an
unavoidable factor which is likely to reduce the acceptability of the method to a number
of consumers.

114
5.2.10. Acceptability of thin joint block work

Percentage responses to the question “Would you buy a house built from this
material?” for thin joint block work are shown in Figure 5.8, divided by region.
Numerical values are given in Appendix C.

100

90

80

70
Yes
Response (%)

60
No
50 Maybe
40

30

20

10

Figure 5.8- Acceptability of thin joint block work construction by region

A high acceptability of this construction method was expected due to its similarities
with brick and block construction. However, it can be seen that it has a slightly lower
acceptability than brick and block; it is felt that this may be an indication of purchasers’
reluctance to consider new and alternative options, preferring to remain with the
familiar.

Most regions had very low levels of “no” responses, the main exception being the
North East. The low response rate in this region may have affected the data. For other
regions the level of “no” responses was similar to that seen with brick and block, this is
expected due to the similarities between the two methods.

The number of “maybe” responses was a little higher that that seen for brick and block.
Comments in the open-ended questions mostly expressed a desire for more
information regarding the benefits and disadvantages of this option over brick and

115
block. This is considered to have increased the number of “maybe” responses. Unlike
SIPs and ICF, a lack of awareness of the method did not seem to greatly affect the
acceptability of this method. This may be because of the similarities, both visually and
in the description, to brick and block construction.

The most frequently expressed comment related to the similarities between this type of
construction and brick and block construction. This view was typically expressed by
respondents who stated that they would purchase a house built using this method, so
was taken to be a positive comment. Additional positive comments reflected those
seen in the brick and block question: the presence of insulation, strength, proven
nature, ease of hanging fittings and appearance were all given as positive aspects. It is
considered that these views, combined with the high acceptability levels seen above,
support the case that purchasers’ favour options that are familiar or have high levels of
similarity with standard options, over those that are different and unknown.

The method received very few negative comments. The most frequently expressed
concern was a dislike of concrete; in some cases this was tied to the perceived high
environmental impact of concrete. However, in others it was not tied to any other
specific view, just stated as an issue. The environmental impact of concrete use varies
greatly, depending upon the composition and the specifics of the individual use. In this
case, the blocks used would be light weight, containing a significant proportion of air
and reducing the volume of concrete used in their construction. If alternatives such as
fly ash are used to replace a portion of the cement then the environmental impact can
be further reduced (Calkins, 2009). When considered in combination with performance
factors such as expected lifespan, the environmental impact of concrete use is often
lower than expected. A lack of familiarity was noted, in particular several respondents
stated that more information would be needed to determine the advantages and
disadvantages over standard brick and block construction.

116
5.2.11. Acceptability of timber frame with brick cladding

Percentage responses for the question “Would you buy a house built from this
material?” for timber frame construction are shown in Figure 5.9, divided by region.
Numerical values are given in Appendix C.

100

90

80

70
Yes
Response (%)

60
No
50
Maybe
40

30

20

10

Figure 5.9- Acceptability of timber frame construction by region

Timber frame construction is the second most commonly used method for building
houses in England, with 7% of houses built in England from 1990 to March 2009 using
this method (DCLG, 2010). Considering this, a higher acceptability was expected for
timber frame as a construction technique than was seen in the responses. However, as
noted in the literature review, the popularity of timber frame construction was seen to
fall during the nineteen eighties after bad publicity regarding the potential for rotting
and fire (Cavill 1999), this may have impacted responses from those who were aware
of the programme.

For most regions, the responses for timber frame are comparable with those seen for
SIPs and ICF; there are noticeably fewer “yes” responses than for brick and block, but
more “maybe “ responses, indicating issues which it may be possible to solve in order
to increase acceptability. Numbers of “no” responses are similar to, or slightly higher

117
than, those seen for brick and block. A high level of “yes” responses was seen in the
North East. As previously commented upon, the response rate from the North East
was fairly low, therefore, the results may have been skewed as a result of this.
Alternatively it may be that there is a greater acceptability in this region. Timber frame
is used more extensively in Scotland in the construction of housing which may indicate
a more positive view of the construction method in more Northerly regions, for example
due to differences in the construction process and the way this is impacted by different
climates. However, the noticeably lower acceptability in the North West does not
appear to support this.

This was a method of which a few of the respondents had experience, with both
positive and negative comments. Positive views indicated that the respondents had
lived in timber framed housing and were happy with the performance, while negative
views stated that problems had been experienced with houses built using the timber
framed methods. Often, the cause of problems was identified as poor workmanship;
due to the multilayer nature of this construction method it is essential that the highest
quality standards are achieved.

Positive comments made regarding timber framed construction were similar to those
seen for brick and block, SIPs and ICF. The majority of respondents felt that the
appearance of the construction method and its similarity to “typical” construction was a
benefit. For those who noted a specific dislike of brick as an external finish, it is
possible to use this construction method with other cladding options. Although this was
not stated in the questionnaire, it has a potential to further increase acceptability. That
the method is “tried and tested” was given as a positive aspect by some respondents,
particularly some of those who had positive past experience of houses constructed
using timber frames. The presence of insulation in the construction method was noted
by a number of respondents as a positive aspect. Government schemes such as the
“Warm Front scheme” have focussed on connections between insulation, the
environment and heating costs which has increased public awareness of the need for
insulation. The final positive point for timber framed construction was that several
respondents noted that it appeared to be of high strength.

Durability was the main concern, in particular with respect to moisture and the potential
for rotting of the timber components. This is discussed by Hutton (1992) who
concludes that avoiding the conditions in which rotting occurs reduces the risk. The
possibility of damage to the waterproof membrane and the difficulty of inspecting the
timber once covered by the brick and plaster are valid points raised. However, as

118
stated above, by adopting the highest quality of workmanship, issues associated with
moisture can be avoided. The process of attaching fixtures and fittings to the walls was
identified as an issue due to the potential for damaging the membranes and causing
moisture issues. This should be carried out with care to minimise the risk. The chance
of insect infestation causing damage to the timber was identified as another risk that
could seriously reduce the longevity of the construction. This is of minor concern in the
UK and can be further limited by the treatment of the timber components if required.

The presence of timber in the construction led to many respondents indicating

concerns regarding the fire performance of timber framed construction. The presence
of plasterboard and a plaster skim finish on the internal surface of the construction
provide sufficient protection to meet the England and Wales Building Regulations;
again the importance of good workmanship can be seen as poor quality work would
leave the frame susceptible to fire; this is discussed by Lennon et al. (2000).

5.2.12. Methods of increasing interest in alternative construction methods

Responses to this question were not divided by region as it was considered that
suggestions could be applied nationally to improve the image of alternative methods of
construction.

The possibility of financial savings scored highly on the potential for increasing interest.
Low prices would increase interest for 89% of respondents and 99% of respondents
said that low running costs would increase their interest. Alternative methods of
construction have the potential to save on either, or both, of these elements, which
could greatly increase their appeal. However, these need to be quantified and
publicised to persuade purchasers that they can make savings which would make it
worth trying their something different. Where construction costs are cheaper, it would
be necessary that a significant portion of the saving is passed on to the purchaser,
rather than being retained as a larger profit by the initial investor, allowing a lower price
that would attract purchasers. For running costs the annual savings compared to an
equivalent brick and block house presents an ideal way to demonstrate the potential of
alternative construction techniques.

A low environmental impact had the lowest percentage of “yes” responses, but still
scored highly, with 86% of respondents indicating that this would increase their interest
in alternative construction methods. This is somewhat in contrast with responses to

119
questions ranking the importance of factors when choosing a house, where
environmental impact received low scores, placing it at the bottom of the list. However,
when combined with other factors during factor analysis, environmental impact as a
whole was one element about which it is considered house purchasers are concerned.
Increasing media focus on the environment and climate change has resulted in a much
greater awareness of the need for change which may have encouraged the “yes”
response. In addition, the Government focus on operational energy has led to an
implied connection between low environmental impact and financial savings which may
have led to a favourable response.

The need for more information about the construction methods was identified from both
the 90% “yes” response to this section and from the comments sections across the
questionnaire, where a “maybe” response was often accompanied by a query. These
queries typically related to performance and could often be answered by existing
research. For example, queries relating to the fire performance of straw bale
construction which can be answered from existing research, demonstrating straw bale
construction to be more than satisfactory for Building Regulation requirements (Jones,
2007). Having that information could turn a maybe into a yes and greatly increase the
acceptability of alternative construction options.

The open ended question offered respondents a chance to state what they felt would
increase their interest. Many of the comments can be tied to the need for information,
for example, the need for methods to be durable and proven. Information exists
relating to many of the factors queried, both anecdotal and from laboratory testing,
however it is not widely disseminated and is sometimes of restricted availability. In
order to increase the acceptability of alternative construction methods, this information
needs to be shared in an easily understandable manner. Without this, the public will
always favour materials they are familiar with, and therefore trust, as suggested by the
high acceptability and comments relating to the use of brick and block.

Guarantees were suggested in this section as a method of increasing the acceptability

of alternative construction methods. The purchase of a house is a large financial
investment, and a guarantee would reassure not only the purchaser, but also
companies providing mortgages and insurance. This could be a route to greatly
increasing the acceptability of alternative construction techniques and therefore
achieving the associated benefits such as lower embodied energy and better
performance.

120
Respondents also commented that the construction method is not the only important
factor when selecting a house. A wide range of other factors would affect the choice,
for example location. This is not a subject that has been examined as it does not relate
to wall construction methods, which is the focus of this work. It would have been
beneficial to eliminate these comments from respondents’ answers by stating in the
introduction to the construction methods section of the questionnaire that they should
consider the construction method only; as if each option were used to construct an
identical house with identical non-construction related benefits such as location, view,
proximity to services etc.

5.3. Summary
 The top three concerns for house purchasers were: price; need for
maintenance and mortgage availability.
 Factors about which the public are concerned when considering a house
purchase include environmental factors, financial factors and risk management.
 The public has a high awareness of brick and block, timber frame and straw
bale. There is low awareness of SIPs, ICF and thin joint block work.
 Brick and block and thin joint block work construction methods have a high
acceptability.
 There is the potential for good acceptability of SIPs and ICF if concerns are
addressed.
 Straw bale and timber frame construction methods have a low acceptability,
although some respondents found them acceptable.
 The acceptability of construction methods varies noticeably with location.
 Respondents indicated a preference for familiar or traditional options.
 Desirable characteristics of a construction method include: insulation, strength,
durability, proven performance, brick appearance externally.
 Many of the concerns about construction methods identified can be disproved
by existing research and anecdotal evidence.
 More information, financial benefits and good environmental performance
would increase public interest in alternative methods of construction.

121
6. House builder acceptability of construction methods
6.1. Introduction

In addition to determining the existence of a market for houses constructed using

alternative methods it was felt that the views of those who would be producing the
houses were important. A reluctance to use a particular method would make it difficult
to promote. Information from past experience with the construction methods being
examined was also felt to be helpful. Construction industry views were collected by use
of a questionnaire, similar to that used to collect public views, but with an industry
focus.

6.2. Questionnaire results

6.2.1. Response rate

Fifty responses were received for the house builder survey. Of these twenty were fully
blank, with only the permission to use information question completed, these were
removed from the data. Thirty surveys were used in the analysis. Partial responses
were retained as they provided a level of information about the topic.

6.2.2. Representativeness of responses

Many of the respondents work for companies with a wide geographical spread. It was
therefore decided not to divide responses by region. All regions had multiple
respondents.

All areas of the construction process were represented, with multiple respondents in
the design, planning and construction categories. In addition, Building Control was
represented by one respondent.

6.2.3. Importance of characteristics in construction method selection

To determine the importance of a range of factors when choosing a construction

method, respondents were asked to score the importance of each factor on a Likert
scale. The scale ranked from 1- Not at all to 5-Very much. All the factors included can
be affected by the construction method. The percentage scores for each characteristic
are shown in Figure 6.1. Modal and median scores are shown in Table 6.1.

122
 70

60

50
Response (%)

40

30

20

10

0
1 2 3 4 5
Importance score
Fire performance Cost
Thermal efficiency Embodied energy
Construction speed Saleability
Health and safety requirements Waste material disposal
Availability Durability
Environmental impacts of construction

Figure 6.1- Percentage scores for the importance of each characteristic

Table 6.1- Modal and median scores for the importance of each characteristic

Importance score

Characteristic Mode Median

Fire performance 3 3
Cost 5 5
Thermal efficiency 5 5
Embodied energy 3-4 3-4
Construction speed 4-5 4
Saleability 5 4
Health and safety requirements 4 4
Waste material disposal 3 3
Availability 4 4
Durability 4 4
Environmental impacts of construction 4 4

123
From a visual inspection of Figure 6.1 and Table 6.1 these factors were given the
following importance ranking, from most to least.

1. Thermal efficiency
2. Cost
3. Construction speed
4. Saleability
5. Health and safety
6. Availability
7. Durability
8. Environmental impacts of construction
9. Embodied energy
10. Waste material disposal
11. Fire performance

These characteristics were all indicated to be important to the respondents, so should

be considered when making decisions. The top ranked characteristics of thermal
efficiency, cost, construction speed and saleability are of the most importance. The
high ranking of saleability indicates that construction methods supported by the public,
as indicated by the results of the questionnaire, have a good chance of being accepted
by the construction industry.

6.2.4. Awareness of construction techniques

As with the public it was desirable to determine the level of awareness of alternative
construction methods within the construction industry. At this stage in the questionnaire
no details of the methods were given, only the names. Higher levels of awareness
were expected compared to those seen with the householder survey as ongoing
training, press releases and continuing professional development encourages
awareness of changes in the industry such as new construction methods.

The survey results indicated a high awareness of the alternative methods in the
construction industry. ICF was the only method to receive any “no” responses, with 7%
of respondents not having heard of it. All other methods had a 100% “yes” response.
High awareness is beneficial in encouraging greater usage. However, being aware of a
construction method does not necessarily indicate a high level of knowledge of its

124
performance; this was examined to some degree in later questions regarding
acceptability.

6.2.5. Past experience of construction methods

To determine the past use of the construction methods being considered respondents
were asked if their company had used each of the methods in the past. Options
included ``yes- housing'', ``yes-other'', ``no'' and ``don't know''. It was possible for
respondents to select multiple categories for this question. The results of this question
are shown in Figure 6.2.

100
90
80
70
60
Score (%)

50
40
30
20
10
0
Yes- housing Yes- other No Don't know
construction
Has your company ever used this method in construction?

Brick and block SIPs

ICF Straw bale and lime plaster
Thin joint block work Timber frame

Figure 6.2- Past usage of the construction methods by respondents’ companies

Responses indicated that all the methods had been used for both housing and ``other
construction'' in the past by at least one of the respondents' companies.

Brick and block had the highest level of past usage. It had been used for housing
(90%) or “other construction” (27%) by all respondents. This high level of past usage
was expected due to it being the most frequently used construction method in England
and Wales.

125
The level of past usage for SIPs was higher than expected, 43% of respondents had
experience of the method in housing and 17% in other construction. This method
currently has a low level of usage in England and Wales, so a lower level of past
experience was expected.

ICF had a fairly low level of past usage, although this was higher than expected
considering how few buildings are currently constructed using this method in England
and Wales. For housing, 17% of respondents had past experience of this method and
3% in other construction. No past experience was indicated by 60% of respondents.

Straw bale had the lowest past usage. It had been used for housing by one respondent
and “other construction” by a second respondent. Although this is lower than other
methods it is higher than was expected due to the unusual nature of the method. The
remaining 73% of respondents had no past experience of using this construction
technique.

Thin joint block work showed some level of past usage, with 50% having used it in
housing and 10% in other construction. The similarity of this method to brick and block
construction meant some past usage was expected. However, as with SIPs and ICF
this was higher than expected when considered in relation to the methods of
construction used for housing in England and Wales

Timber frame construction had a high level of past usage, 73% said their company had
used it for housing, 13% for other construction. As this is the second most frequently
used construction method for housing in England and Wales a high level was
expected.

All methods of construction had a higher than expected level of past usage. Methods
other than brick and block and Timber frame accounted for only 5% of housing
construction in England and Wales between 1990 and March 2009 (DCLG, 2010) so
past experience of other methods was expected to be low. This past usage indicates
that there is some experience of using alternative construction methods in the building
industry and potentially acceptability for the methods.

The relatively high past usage of alternative construction methods shown by

respondents in this question was felt to be beneficial when considering responses to
the following questions about acceptability. Having past experience with a method may
have given respondents insight into benefits and issues and influenced their decision
on each method’s acceptability.

126
6.2.6. Acceptability of alternative construction methods

To determine the acceptability of the alternative construction methods being

considered the respondents were shown a labelled axonometric diagram of the method
and given a brief description. Respondents were asked if they felt their company would
use this method in construction with answer options of “yes”, “no” and “maybe”. An
optional open ended question allowed respondents to add any further comments they
had about the method. This question was similar to the one used for the public
opinions survey, however the method description was slightly more technically worded.

Respondents were asked to give the views of the company they worked for rather than
their individual views which may have affected the acceptability. For example a timber
frame construction company is unlikely to find brick and block construction acceptable.
Responses to this section were considered in conjunction with the data on past usage
to see if a company with past experience of a particular method would be prepared to
reuse it and if not why. Percentage responses to this question are shown graphically in
Figure 6.3.

100

90

80

70
Response (%)

60

50

40

30

20

10

0
Yes No Maybe
Would your company use this method in construction?

Brick and block SIPs

ICF Straw bale and lime plaster
Thin joint block work Timber frame

Figure 6.3- Percentage responses to the question “Would your company use this
method?” for each option

127
A high level of acceptability was expected with brick and block construction as it is the
most frequently used method for UK housing, this was seen with 89% of respondents
saying they would use this method. Of those who would not use the method, one
respondent stated they work only in the timber construction sector so their acceptability
of other methods was expected to be low. A dislike of the insulation used was given as
the reason for not using the method by the other respondent to say they would not use
it, if an alternative was offered they may find the construction method acceptable.

Positive views expressed in the open ended question included that the method was the
traditional method of house construction, it is proven and affordable. The acceptability
and familiarity of this method for both tradesmen and purchasers was noted. The ability
to meet Building Regulations using it was also listed as a positive.

Only one respondent stated a negative comment. It was noted that the construction
method is “not carbon friendly”. However, this respondent indicated they would still use
the method. Several respondents commented that a brick and block construction is
used because it is specified, indicating there may be scope for other methods if the
designers specify their use.

A lower level of acceptability was expected with SIPs than for brick and block as it is a
non typical form of construction. Figure 6.3 shows that the acceptability is significantly
lower at 48%. If the “yes” and “maybe” responses are considered together it shows
there is the potential for use if the issues causing “maybe” responses can be
addressed. Respondents who would not use the method accounted for 30%
responses. Two respondents who had past experience of the method, either for
housing or other construction indicated they would not use it again; however they
made no further comment to indicate why.

The speed of construction was commented on by several respondents. Having a faster

construction time has the potential to reduce costs by reducing labour requirements
and reaching completion faster, allowing a quicker recovery of investment. The
structural performance, air tightness and thermal efficiency achievable using this
method were noted by respondents who indicated they would use it in construction.

The fact it was not the typical construction method was noted, along with the fact that
this may make it unpopular. One respondent commented that “Developers are terrified
of systems they haven't used before”. Addressing this issue is likely to be one of the
major challenges of encouraging alternative construction techniques.

128
Opinion was divided on the potential for financial savings, some respondents felt there
was currently no potential for savings, others felt it was a cheaper option. It was
indicated that changes to legislation may increase the appeal of this method as it could
become a cheaper solution for meeting more stringent regulations. Another financial
issue noted was the mortgagability of houses constructed using the method. This is
another challenge to overcome to gain the benefits of alternative methods of
construction. Financial providers would need to be encouraged to accept the
alternative methods to enable purchasers to afford houses. This may require proof of
performance, either from anecdotal evidence or laboratory testing.

It can be seen from Figure 6.3 that ICF has a lower level of acceptability than SIPs,
with 33% “yes”, 37% “no” and 30% “maybe”. When ”yes” and “maybe” are considered
in combination ICF shows a medium level of acceptability with the construction
industry. Those respondents who had past experience of the method said either “yes”
or “maybe” they would use the method again.

Positive comments for this type of construction included greater construction speed,
structural performance and the thermal performance achievable. These were mostly
noted by those who had past experience of using the method.

ICF had a greater number of negative comments, including some expressed by those
who had used the method in the past. The most frequently expressed negative was
related to the higher cost of construction. ICF is considered to be more expensive
(Denzer and Hedges 2007, Al-Homoud 2005); however, the better thermal
performance as noted by the respondents may offset this higher cost over time.
Limitations on the construction as a result of using this method were noted by some
respondents. These included limited future adaptability, wall thickness and restrictions
on service penetration of the wall. Although these exist as issues with the method,
careful design at the initial stages should limit their negative effects.

Straw bale construction was expected to have a low acceptability due to its unusual
nature. This was seen in the responses, with 15% saying “yes” they would use it; 63%
“no” and 22% “maybe”. Although this is a low acceptability, it is higher than expected,
particularly if “yes” and “maybe” responses are considered in combination. Of those
who had past experience of the method, one would use it in housing, the other would
not.

Several respondents noted positives of this construction type; including the sustainable
nature; good thermal performance; availability and affordability of the materials used.

129
The respondent with past experience of the method who would consider it for future
use listed sustainability, BREAAM score, availability and cheap materials as reasons.

Perceived financial issues associated with this construction method focused on

difficulties selling the end product and the lack of financial products to assist this
(mortgages etc). As with other alternative methods this is likely to be an issue that
would need to be resolved to take advantage of the method. The public survey results
(Chapter 5.2.8) show there is some acceptability for this method with purchasers, so
there would be a market for the finished houses; however, it is lower than for other
options.

The respondent with past experience who would not consider using the method again
gave wall thickness and its impact on reducing the number of units that can be built in
a given space as the reason why. This was also noted by other respondents. Wall
thickness is an issue with this type of construction. Although it has a negative impact
on the density of housing units it can be used to incorporate design features (Lacinski
and Bergeron, 2000). It should also be noted that as thermal performance
requirements increase, other methods of wall construction will also increase in
thickness, reducing the difference.

The method was viewed by respondents as unproven, with durability and fire
performance of particular concern. Examples exist in the USA that are up to one
hundred years old (Seyfang 2010, Smith 2007, Yates 2006) and in the UK from 1994
(Jones 2007). These examples can be used to provide anecdotal evidence for the
construction method. In addition laboratory and field research has been carried out into
some of the areas of highest concern such as durability (Carfrae et al. 2009, Lawrence
et al. 2009, Goodhew et al. 2007, Summers 2006, Goodhew et al. 2004, CMHC 2000)
and fire performance (Apte et al. 2008, Jones 2007).

Acceptability of thin joint block work was expected to be high as a result of its similarity
to standard brick and block construction. Figure 6.3 shows this is the case, with 67%
saying “yes” they would use this construction method. This is the second highest “yes”
rate. “No” was selected by 15% of respondents and 18% said “maybe”. One
respondent who had past experience said they would not use it again, but gave no
further reason, all other respondents who had past experience of the method would
consider using it again.

The similarity to standard construction was the most frequently expressed positive. The
increased speed of construction and the reduced costs of the build method were also

130
noted. Additional comments made included: good thermal efficiency, reduced wastage
of mortar, and the possibility of improved airtightness if well constructed.

Negative comments for this type of construction focused on the issues of

workmanship. In particular the issues associated with the gauge differences between
the inner and outer leaf and the need to deal with this when building. Appropriate
training in the construction method should avoid these issues. One respondent
expressed a concern that larger blocks can result in health and safety issues on site;
although this is a possibility good site practices should reduce this to a minor risk.

A high level of acceptability was expected for timber frame construction as it is a well
known technique and the second most frequently used method in England and Wales,
this was seen with 63% of respondents saying “yes” they would construct housing
using timber frame; 19% said “no” and 18% said “maybe”.

Reduced construction speed was noted as a benefit by a number of the respondents,

as with SIPs this can result in reduced costs for the project. Another comment was that
it is a fairly traditional method of construction, and as such is accepted by the
construction industry. The presence of insulation and the potential for achieving
airtightness were seen as benefits by some respondents.

One respondent commented that the method is best for large number of repeated
units, such as seen on large speculative housing estates. This is both an advantage,
as the repeated nature allows the mass production of components, and a disadvantage
as it can limit flexibility and reduce the appeal of the method for smaller developments
or those with many styles of construction.

The more expensive nature of this type of construction was given as a disadvantage by
several respondents. Although one respondent stated that this can be balanced by
savings associated with the speed of construction, increased speed cannot be
guaranteed as external factors such as weather can affect it. If high thermal
performance is achieved the savings resulting from this may offset the higher
construction costs over time.

Other negative views expressed included that the method is not popular with clients, a
view which is supported to some extent by findings from the householder survey in this
work (Chapter 5.2.10). The durability of the construction method was also given as a
concern. However, as discussed by Hutton (1992), with high quality workmanship to
avoid the condition which allow rotting to occur this should not be an issue.

131
6.2.7. Methods of increasing interest in alternative construction methods

Respondents were asked what would increase their interest in alternative options. If a
method has high potential for improving environmental standards but low industry
support it would be necessary to encourage its use. This question aimed to identify
ways this could be achieved.

The factor with the greatest impact was low cost, with 96% saying this would increase
interest. Reducing the cost of the construction can result in a greater profit or the ability
to sell at a lower price, increasing saleability and leading to a faster return on
investment.

A response of 81% yes to “market desirability” indicated that support from customers
would increase the interest in alternative methods. The householder survey has shown
that customers would be interested in some of the alternative options, particularly SIPs
and ICF. If this information were communicated to the industry to demonstrate that a
market exists for houses constructed using alternative methods the level of interest
may increase. Low running costs can be linked to market desirability as the ability to
sell a home as “green” and “cheap to run” both have the potential to increase the
attractiveness of the property. Although this may not have a direct benefit to the builder
it increases saleability resulting in faster and potentially higher returns, 87.5% of
respondents said this would increase their interest. Similarly, “if they were
environmentally friendly” was indicted to increase interest (87.5% “yes”). Again the
direct benefits to the builder may be low, but the increased saleability and the option of
marketing their product as “green” can have financial and reputational benefits.

As seen with the public questionnaire responses, more information scored highly (87%
“yes”). Although the industry respondents had heard of the options, and some had past
experience, those who did not may not know a great deal about the methods and their
advantages. By highlighting the benefits of alternative methods their use could be
greatly increased. This would discourage the attitude of sticking with what you know
and encourage the use of a wider range of options.

6.2.8. Comparison of survey results

The acceptability of each construction method was compared for the public and
industry questionnaire results. It was seen that there is some similarity between the
acceptability, with brick and block scoring highly. Both groups show some acceptability

132
for SIP and ICF construction, although this is lower in the construction industry
responses. This lower score supports comments made in the literature review that the
construction industry is reluctant to adopt new methods. Prefabricated straw bale
construction had a low acceptability with both groups; however there was some degree
of acceptability indicating it may have potential for use in some cases. Timber framed
construction appears more acceptable to the construction industry than to the public,
which agrees with responses to the industry questionnaire, which note it is not popular
with the public. As a result of this it is somewhat surprising that this is the second most
common method of housing construction in England and Wales.

6.3. Summary
 The three most important factors when selecting a construction method were
thermal efficiency, cost and speed of construction.
 There is a high level of awareness of alternative construction methods in the
industry.
 Respondents had past experience of all methods examined in this work.
 Brick and block construction has the highest level of acceptability.
 There is a good acceptability of timber frame and thin joint block work.
 The potential exists for a high acceptability of SIPs and ICF if concerns are
addressed.
 Straw bale had a low acceptability, although some respondents would consider
its use.
 A number of concerns were identified for each construction method; many of
these can be disproven by existing evidence.
 Financial benefits and more information would increase industry interest in
alternative methods of construction.
 Saleability would have some effect on industry interest in alternatives, therefore
support from the public (the buyers) may encourage support within the
construction industry.
 Similarities are seen between the public and industry survey results. Both
groups support brick and block strongly, SIP and ICF to a moderate degree and
show little support for prefabricated straw bale.
 Timber framed construction is noticeably more popular with the construction
industry than with the public.

133
7. Development of the optimisation based methodology

To aid the design process, by simplifying the selection of wall construction method, a
design methodology was created. This design methodology uses optimisation to allow
the user to identify the best, or optimal, solution to the design problem; in this case, the
problem of selecting the best wall construction method to use. This chapter describes
the process for developing the fitness function containing the single output equation
and the optimisation based design methodology.

7.1. Design of the fitness function

7.1.1. Input variables

The floor plan of the house was taken to be a rectangle with the dimension
designations shown in Figure 7.1.

(b)

(a)

Figure 7.1- Dimension designations for single objective calculation

User inputs for the generation of the single objective variable were total internal floor
area, number of storeys, height of each storey, number of doors, region, type of
building, construction method, maximum window percentage, door size and the front
dimension of the building (a). In order to maintain integer values, dimensions were set
in millimetres for (a) during the optimisation. The value of (a) is then converted to
metres during the calculation process. Default values were set for all except floor area.
These can be seen in Table 7.1.

134
Table 7.1- Default values for variables in the single objective equation

Factor Value
Total building area No default set, input required
Number of storeys 2
Height of storeys 3.0m
Number of doors 1
Region 3 (London)
Building type 2 (Semi detached)
Construction method 1 (Brick and block)
Maximum window percentage 25
2
Door size 2.42m
Dimension (a) 3.6m

7.1.2. Constraints

Constraints represent a limitation on the optimisation programme. The equation must

be solved but according to these limitations, for example a maximum or minimum value
for some aspect of the equation.

It was considered desirable to maintain the optimisation as unconstrained as

constraints can cause complications with the optimisation. Goldberg (1989) explains
that genetic algorithms are naturally unconstrained. The only constraint on the single
objective equation is the need to have a minimum window area of 20% of the building
floor area as recommended by Part L of the Building Regulations (HM Government,
2010). This allows a sufficient amount of natural light and hence avoids excessive use
of artificial lighting. By satisfying this constraint the operational energy is not negatively
affected by the window area. Although operational energy resulting from heat loss
through windows and doors is not included in the operational energy calculation used
in this work, the area of wall removed to accommodate windows and doors must be
removed for the assessment of embodied and operational energy demands resulting
from the wall materials. This provides a more representative value of the energy
requirements relating to the wall construction method.

The window area must also be of a size that it does not eliminate too much of the wall
area. If too much wall area is discounted due to being glazed, then the value achieved
from the single objective equation will be unrealistically low for large floor areas with
low values of (a). A maximum value of wall area that can be removed was set. The
default value of this is 25%.

135
In order to satisfy the need for a constraint on the window and wall areas a simplified
version of the penalty method discussed by Goldberg (1989) and Bunday (1984) was
used. If a value of dimension (a) caused 20% of the floor area to be greater than 25%
of the wall area (i.e. an unfeasible glazed area) the single objective value was set to
10. This value was selected as it is considerably greater than any values calculated by
the programme. The result of this was a high value of the single objective equation
which was discounted by the optimisation process as it was significantly higher than
the optimal value.

A further constraint is applied to the optimisation by setting upper and lower bound
values for the variables. This indicates limits to the variables that the optimisation
process will examine. The method for applying bounds is contained within the
optimisation programme; it is not part of the single variable equation. It is not desirable
to apply the window area constraint by adjusting the lower bound as this would require
calculations to be carried out before the optimisation methodology can be used. The
method adopted in this work eliminates this step while still allowing the constraint to be
incorporated.

7.1.3. Wall dimensions

For terraced and semi-detached buildings dimension b was always the length of the
party wall.

The value for dimension (a) was a user input, or adopted the default setting from Table
2. During the optimisation process this value was input by the optimisation programme.

To convert the value of dimension (a) from millimetres to metres it was divided by
1000. This allowed the optimisation programme to use millimetres but the single
objective calculation to be carried out using metres. Dimension (b) was calculated
using Equation 7.1.

(Eqn. 7.1)

Aspect ratio was calculated using Equation 7.2.

(Eqn. 7.2)

136
The value of total external wall area was dependent on the type of building. To
incorporate this a switch function was used allowing the correct equation for the
specified building type. Equations 7.3, 7.4 and 7.5 were used for detached, semi
detached (s/detached) and terraced houses respectively.

(Eqn. 7.3)

(Eqn. 7.4)

(Eqn. 7.5)

Window and door areas have a significant impact on the total embodied and
operational energy requirements of the house resulting from the wall construction
method. Windows and doors reduce the area constructed from the wall materials and
hence the volume of materials used. Therefore, the area occupied by windows and
doors must be removed from the total external wall area. Door area was based on user
inputs or default values for number of doors and area of doors. Window area was set
at a minimum of 20% of the floor area as suggested by part L of the Building
Regulations (HM Government, 2010). This allows sufficient natural lighting to limit the
use of artificial lighting and hence reduces operational energy requirements. The
window area was therefore calculated using Equation 7.6.

(Eqn. 7.6)

The net wall area was then calculated using Equation 7.7.

(Eqn. 7.7)

137
It was desirable to keep the window area below 25% of the wall area to avoid a high
glazed area artificially reducing the embodied and operational energy. This limitation
was accommodated by including a constraint in the single objective equation (see
7.1.2 Constraints).

For those building types with party walls, the party wall area was calculated using the
Equation 7.8 and 7.9 for semi detached and terraced houses respectively.

(Eqn. 7.8)

(Eqn. 7.9)

7.1.4. Embodied energy

Using the generated wall areas from Chapter 7.1.3, embodied energy could then be
calculated using the embodied energy data from Chapter 4. For party walls, half the
embodied energy was assigned to the house in question as the value was considered
to be divided between the two adjoining properties. This calculation is shown in
Equation 7.10.

(Eqn. 7.10)

Embodied energy of the whole house was considered in relation to the floor area. To
generate the embodied energy per square metre Equation 7.11 was used.

(Eqn. 7.11)

138
7.1.5. Operational energy

The wall construction method was considered to only impact on the operational energy
requirement relating to heating as a result of heat loss through the wall fabric.
Therefore, only this operational energy is considered. Operational energy associated
with lighting, hot water, appliance use and heat loss resulting from ventilation systems
are not considered as they are not impacted by the external fabric of the building
resulting from the wall construction method. Operational energy requirements resulting
from heat loss through the walls were calculated using the degree day method. Degree
days are the sum of the difference between external temperature and the desired
internal base temperature, in this case 18°C, over the course of a given time period.
Measurements are taken at intervals, for example, daily or hourly, as the intervals
decrease the accuracy improves (CIBSE 2006).

To generate values for heating energy requirements using degree day values they
must be combined with the building’s specific heat loss rate, calculated by multiplying
the U value of the wall and the total wall area built using the wall construction method
in m2. Multiplying by 24 converts the value to days, dividing by 1000 converts the value
to kWh. This calculation is shown in Equation 7.12 (based on CIBSE 2006).

(Eqn. 7.12)

Degree day based calculations are considered to give a less accurate, although widely
accepted, value than a full thermal simulation. In addition the use of annual values for
degree days allows the operational energy to be calculated for a full year, thermal
simulations often calculate the energy requirement for a short time period, such as
twenty four hours. The degree day method has significantly lower calculation times
than a full thermal simulation. For optimisation work such as this, where it is necessary
to run the calculation many times, this is considered to be a benefit and the reduced
accuracy considered to be acceptable.

An internal base temperature of 18°C was selected using the World Health
Organisation (WHO) recommendation that this is an acceptable temperature for
healthy, appropriately dressed adults (WHO, 1979). This was considered to be an
acceptable value as 15.5°C is traditionally used in the UK (CIBSE 2006). The higher

139
value used in this work is considered to be more representative of housing situations.
Higher or lower values could be substituted in future work. The locations for degree
day values were based on the city in each region with the highest population, as this
was considered to be the most likely location for new housing (Office for National
Statistics 2011). Additional gains, for example from occupants and equipment are
ignored for this work as they would be equal across the construction types.

Values of heating degree-days for investigated locations or regions were calculated by

M. Mourshed using the methods described in Mourshed (2012). Interpolated weather
data for the key cities within the selected region were obtained from Meteonorm
software (Meteonorm 2011). Degree-days values used in this research are given in full
in Appendix C.

All windows and doors were assumed to be identical across the construction methods
having equal contribution in the calculation of heating related operational energy
demands. Therefore, these are omitted from the equation. Party wall construction is
taken to be solid or fully filled and sealed, therefore the heat loss through these is
considered to be zero (see Chapter 4).

Operational energy (OE) is considered in terms of the energy per square metre of floor
area, to generate this value Equation 7.13 was used.

(Eqn. 7.13)

7.1.6. The single objective calculation

Data generated in 7.1.2 to 7.1.4 was used in the calculation of the single objective. The
data was combined with values for airtightness, wall thickness, fire performance and
acceptability. These were taken from Chapters 4, 5 and 6 (Table 4.23, Figures 5.4-
5.10, Figure 6.3). Switch functions allowed the selection of the correct value depending
on the input for construction method and region.

To allow the combination of factors with different scales it was necessary to apply
normalisation to the values. This was achieved by dividing the values for each
construction method by a set of benchmark values. Benchmark values for each factor

140
were calculated from the mean of the values for all methods of construction under
consideration.

Published values for embodied energy per square metre of floor area and operational
energy per square metre of floor area were considered as benchmark values.
However, they include other elements of the construction, such as roofing; and
operational energy uses such as lighting. To generate a benchmark value the details of
a “typical” house were determined (see Appendix F). The embodied and operational
energy per square metre were calculated using the typical house dimensions for
detached, semi-detached and terraced house types. Average values of embodied
energy per square metre and U value were used in these calculations. Values of
embodied energy did not change with location, however, operational energy did as the
result of differing degree days for each region. Operational energy values were
calculated for each region. Benchmark values calculated for embodied and operational
energy can be seen in Tables 7.2 and 7.3 respectively. Benchmark values used for
performance factors shown in Table 7.4.

Table 7.2- Benchmark values for embodied energy, divided by house type.

Benchmark
House type Embodied energy
Detached 1532.4
Semi-detached 1169.2
Terraced 805.9

Table 7.3- Benchmark values for operational energy, divided by region and type of
house.

Region Detached Semi-detached Terraced

1 27.0031 16.5898 6.1765
2 27.2418 16.7364 6.2311
3 23.8927 14.6789 5.465
4 27.8301 17.0979 6.3656
5 25.2658 15.5225 5.7791
6 24.7854 15.2273 5.6692
7 25.1047 15.4235 5.7423
8 28.2463 17.3536 6.4608
9 28.3126 17.3943 6.476
10 25.7935 15.8466 5.8998

141
Table 7.4- Benchmark values for performance factors

Variable Benchmark value

3 2
Air tightness 2.73m /hr.m at 50Pa
Wall thickness 0.374m
Fire performance 130 minutes

Normalisation of the acceptability scores was achieved by dividing the value achieved
by one hundred as this was the maximum achievable score.

User input weightings enabled the importance of the different factors to be adjusted
and these to affect the single value output. These have a default setting of 1 for all
variables, giving them equal importance.

To accommodate the constraint of a maximum wall area being taken by windows a

switch function was included in the programme at this point (see Chapter 7.1.2).

If the value of window area required was greater than the maximum allowable, the
single objective value takes a fixed value. This value is considerably larger than the
values of single objective that were generated by the programme for the construction
options. Having a high value of single objective means the optimisation programme will
not select the variables which produce this value. In this situation the single objective
was calculated using 7.14.

(Eqn. 7.14)

If the window area was below the set value for maximum window area, and hence did
not violate the constraint, Equation 7.15 was used for calculating the single objective.

142
(Eqn. 7.15)

The lowest value of single objective indicates the best compromise between the
variables.

The Matlab programming code for the fitness function based on the above process can
be seen in Appendix G.

7.2. Optimisation procedure

Two input variables were selected for the optimisation process; these were building
front dimension (a) and wall construction method. During the optimisation process, the
input values are varied using a genetic algorithm until the optimal solution is identified.
The programme identifies the input values which obtain the lowest score for the single
objective calculation.

7.2.1. Optimisation variables

To reduce the number of variables for the optimisation procedure, those variables
which were to remain fixed throughout were given a set value. Fixed variables were
included in the programme with a single value, rather than being an input during the
optimisation process. This simplifies the process and reduces calculation time. These
values can be altered, and the programme re-run, as required to evaluate different
scenarios. Values selected to be set were total building area, number of storeys, height
of storeys, number of doors, region, building type, maximum window percentage and
door size. The value of fixed variables is set by the user before the optimisation

143
procedure is run; alternatively, default values will be used. Weightings for each
element in the single objective equation can also be altered at this point.

Values which were to be varied during the optimisation procedure were: front building
dimension (a) and wall construction method. This allowed the exploration of varying
aspect ratios and construction method combinations. As the optimisation was to be
carried out with two variables they take the form of a two dimensional vector as
discussed by McKeown et al. (1990).

Although the example given here, and used in the case study presented in Chapter 8,
considers two variables it is possible to use the method created to evaluate a greater
number of variables. Any of the fixed values could be converted to a variable for the
optimisation process. However, in this work, it was intended to demonstrate the
method, so a simplified version with only two variables was used. The output of the
method with two variables can be used to identify the optimal combination of
dimension (a) and construction method.

7.3. Use of the decision making methodology

The design methodology has been created to aid house construction decision making.
The main focus of this work was on the environmental impact of house wall
construction method, in particular achieving the best compromise of embodied and
operational energy use, with consideration given to acceptability and performance. By
altering the factors used in the single objective equation, the method demonstrated
could also be used to consider a wide range of construction aspects, for example roof
construction, foundations, and building services. The potential exists to expand the
methodology. Additional methods of construction can be incorporated, along with
different specifications of the same methods, for example thicker SIP panels. The
number of criteria could also be increased; the same method could be used to assess
cost or other aspects of performance such as lifespan, speed of construction etc. The
factors chosen for this work were intended to demonstrate the range of criteria that can
be included.

The case study presented in Chapter 8 demonstrates the use of the optimisation
methodology to solve the question of the best compromise between method of
construction and size of dimension (a) when consideration is given to energy use,
public and industry acceptability of the method, air tightness, wall thickness and fire

144
performance. This is an example of how it could be used in a design situation where
the best compromise of these elements is required.

The methodology could be used in any situation where construction method selection
is necessary. This is most likely to occur at the design stage of housing construction,
so it will be of most use to designers. The use of optimisation aids the difficult decision
between choices that have no clear single best solution. A compromise is achieved but
it is one that can be justified. Identification of sub-optimal, but good scoring, variables
can also assist with decision making; these may ultimately be the best options when
other criteria are considered. The ability to demonstrate why a particular method is the
best compromise, for example by the use of fairly easy to understand graphical images
such as three dimensional plots and phi arrays, could support the case for a particular
construction method. This could be useful to financial backers, planners, Building
Regulation officers as well as to the end user. Phi arrays are of particular benefit here
as a visual representation is often easier to understand for non specialists. This agrees
with comments made by Li and Shen (2002) regarding the way in which decision tools
can increase the transparency of the decision making process with all parties
understanding the reasons for an option being selected.

The ability to demonstrate the reasoning behind a construction method selection could
be used to encourage Government support. Government backing of alternative
construction methods could increase their use. This may be by financing research into
methods or by promoting methods to increase awareness and acceptability. This would
allow the benefits demonstrated by the alternative methods used in this work, such as
better U values, lower embodied energy and good airtightness to be accessed.

7.4. Summary
 To allow the use of optimisation in aiding construction method selection, a
fitness function was designed.
 The fitness function combines data generated and collected in Chapters 4, 5
and 6 to create a single value that incorporates embodied energy, operational
energy, performance, acceptability and weightings.
 The single value equation can be used to assess a particular design
combination. It is also used as the basis for the optimisation design
methodology.

145
 An unconstrained problem was preferred due to its simpler nature. To achieve
this, but to accommodate the need for a minimum window area and associated
external wall area, a simplified version of the penalty method was used. Values
of (a) which do not allow sufficient window area will result in a high value of the
fitness function. This will not be the optimal value and therefore will not be
selected.
 To simplify the optimisation process, two variables were selected for
consideration; these were construction method and dimension (a), the front
dimension of the house.
 All other variables were set prior to the optimisation.
 An alternative combination of fixed and non fixed variables can be used if
desired.
 Potential uses for the optimisation based methodology were identified as
material selection when designing buildings.
 Phi arrays and graphs produced from optimisation data can be used to
demonstrate the reasoning behind a construction method selection.
 Use of the methodology to demonstrate the best construction option for
England and Wales could provide justification for Government support of a
particular construction method, resulting in increased use and accessing the
environmental benefits it offers.

146
8. Case study
8.1. Introduction

To demonstrate the potential implications of this work, the optimisation based design
methodology developed in Chapter 7 was run for a given scenario. The scenario was
based on housing figures for England and Wales, with the most frequently-constructed
type and size of housing examined. The design methodology was run for the case
study with three sets of weightings for the factors in the single objective equation:
energy use, acceptability and performance. These weightings were: all factors equal,
weightings based on the public survey and weightings based on the construction
industry survey. The use of varied weightings demonstrates the impact weightings can
have on the output of the design methodology. The initial run of the design
methodology used equal weightings to demonstrate the functionality of the
methodology and ways in which the solution space can be examined. Using weightings
based on the results of the questionnaires in chapters 5 and 6 was considered to
provide a more realistic view than equal weightings; therefore this was carried out to
determine which method of construction would be the optimal for the case study
situation. Where an alternative method of construction was identified as optimal,
energy consumption values were generated for constructing the case study house from
the alternative method of construction and from brick and block construction. A
comparison of these allows a demonstration of the effect of using the most suitable
alternative material as identified by the optimisation.

8.2. Case study optimisation results

8.2.1. Solution metrics for equal weighting case study

The solution metrics achieved by running the optimisation programme for the case
study detailed above, when equal weightings for all factors, are given in Table 8.1.

Table 8.1- Solution metrics for the optimisation of the case study

Variable Optimal solution

Construction method 1- Brick and block
Dimension a (mm) 6377

For the case study values discussed above, the optimisation programme identified
option 1- Brick and block as the best option for construction method. Brick and block
has a number of factors which result in a good score for the single objective value. A

147
low embodied energy, combined with high acceptability and fire performance result in
the “typical” construction method being selected as the best option when all factors are
equally weighted.

The best dimension of (a) was identified as 6377mm. This is very close to the minimum
value at which the constraint applied by window area is not violated. The minimum
value of (a) which will give an acceptable external wall area, allowing sufficient window
area to satisfy the Building Regulations, is 6373mm. Increasing values of (a) require
more materials and allow greater heat loss due to increased external wall area. This
will increase their embodied and operational energy values and hence their fitness
function value, making them less desirable. It can be seen that satisfying the
requirement for window area to be equivalent to twenty per cent of the floor area is a
challenge when constructing terraced properties. The reduced external wall resulting
from terraced construction leads to a limited area for window placement. Possible
solutions to this could be to allow a glazed area greater than twenty five percent of the
wall, or to use methods such as sun tunnels which transport light from roof windows
into the living areas of the building. Constraint violation could be permitted, meaning
that the window area is less than twenty percent of floor area, if it is shown that the
housing design is capable of generating sufficient energy to compensate for the
increased operational energy requirements. However, this is undesirable as it places
greater requirements on energy generation and natural lighting is considered to be
better for health reasons.

8.2.2. Fitness landscapes

The fitness landscape is the score of single objective achieved for each combination of
options. To display this values were calculated for the case study using a spreadsheet.
Calculating all possible combinations would be a large and time-consuming task as for
the case study values discussed above there were 5.8x104 possible combinations of
dimension (a) and construction method. Therefore, dimension (a) was increased in
intervals of 0.250m for each construction method; the single objective value was
calculated for each combination. Whilst this allows a graphical representation of the
results, which can be useful for gaining a better understanding of the solution space
and high and low scoring regions, it would not be sufficiently accurate for use in
design. The use of optimisation in determining the best value of (a) greatly simplifies
this in comparison with the use of a spreadsheet. Fitness landscapes can be portrayed
three dimensionally as shown in Figure 8.1.

148
 10

Single objective value

8
6
4
2
0
3.6
4.25
5
5.75
6.5
7.25
8
8.75
9.5
10.25

11

11.75

Timber frame
Thin joint block work
Prefabricated straw bale
12.5

ICF
SIP
Brick and block
Dimension a (m)

Wall construction method

Figure 8.1- Three dimensional representation of the fitness landscape

The region which violates the window area constraint can clearly be seen as the high
section at low values of (a). However, when considering viable values of (a), which do
not violate the constraint, the three dimensional nature causes low values to be
obscured by the higher values surrounding them. For example, low values achieved by
prefabricated straw bale are obscured by the higher scores for ICF. As the variable
wall construction method has a low range of values, the results can also be displayed
on a two dimensional graph with a line for each construction method option. This
allows low values to be seen more easily and the optimal method of construction to be
identified. The optimal method is the one which has the lowest score for single
objective value. It appears on the graph as the lowest point of all the lines. The two
dimensional display of the fitness landscape values for the case study with equal
weightings is shown in Figure 8.2.

149
 12
Brick and block
SIP
10
ICF
Prefabricated straw bale
Single objective value

8 Thin joint block work

Timber frame
6

Dimension a (m)

Figure 8.2- Two dimensional graphical display of the fitness landscape for case study
optimisation with equal weighting

The region which violates the window area constraint can be seen as the high scoring
region from (a)=3600mm to (a)=6500mm. Houses built using dimensions of (a) in this
range would have a low window area, below the 20% of floor area recommended by
Part L1A of the Building Regulations. This would result in higher operational energy
requirements for lighting and so presents an unsatisfactory solution. Although reducing
the value of (a) results in lower values for embodied energy as material requirements
decrease, it is not acceptable for this to lead to increased lighting requirements and
hence greater operational energy.

The optimal solution of brick and block as the construction method can be identified
from the graph, the optimal value of (a) can be seen to occur in the region of 6500. It is
logical that the single objective value increases as dimension (a) moves away from the
optimal value. The external wall area is increasing, which will result in increasing
values for operational and embodied energy. Use of the graph allows sub-optimal
solutions that have low single objective values to be identified. It is noted that, although
brick and block is identified as optimal, thin joint block work and prefabricated straw
bale panels are sub-optimal, but are nevertheless low scoring options in the optimal
region of (a). These may be more suitable if other criteria are considered, for example,
the dimensions achieved by multiple standard brick units.

150
As the value of dimension (a) increases, the single objective value does not increase at
the same rate for all materials. At higher values of (a), different construction methods
may prove to be optimal, for example, at a = 8m the optimal method of construction
would be prefabricated straw bale construction.

Figure 8.2 shows Timber frame and ICF wall construction methods to have particularly
poor performance compared to the other construction methods examined; this
indicates that they may be unsuitable for use. This is of particular interest, as timber
frame is currently the most widely used alternative construction method for the area of
study.

It should be noted that the use of a two-dimensional graph as shown in Figure 3 would
not be feasible if the second variable had a high range of values, for example, floor
area. The number of lines on the graph would cause a low level of clarity and obscure
information. For consideration of this situation, the fitness would need to be displayed
on a three dimensional graph, or by use of a Phi array.

8.2.3. Phi array visualisation

The use of Phi arrays allows a visual representation of the solution space (Mourshed et
al. 2011). For visual clarity when designing the phi array, the single objective value for
each point was subtracted from ten, making the optimal solution have the highest
value. As a result of this, the optimal combinations of material and dimension (a)
appear as large red circles. As the suitability decreases, the circles become smaller
and the colours descend the scale until the small blue circles (representing constraint
violation) are reached. Phi array generation programming created by M. Mourshed,
following methods in Mourshed et al. (2011) was used to create a Phi array for the
case study; this is shown in Figure 8.3.

151
Figure 8.3: Phi array for single objective function for case study, equal weightings.

The results seen from the phi array correspond to those discussed for Figures 8.1 and
8.2. It can be seen that there is a low level of fitness at low values of (a), indicated on
the Phi array by small, dark blue circles. This corresponds to the region where the
external wall area is too small to allow sufficient glazed area to avoid additional artificial
lighting. The region following this demonstrates the best fitness, corresponding to the
values of (a) where the window area is acceptable but (a) is the lowest it can be. This
scores highly as it leads to the lowest levels of embodied and operational energy. As
the value of (a) increases, the fitness decreases- these values are less suitable.

The phi array also demonstrates the difference in single objective value achieved by
changing the construction method. It can be seen that brick and block, prefabricated
straw bale, SIP and thin joint block work are the better options, with large circles in the
red-yellow range throughout. Prefabricated straw bale shows the best score at higher
values of (a), indicated by the larger circles which do not drop below orange on the
colour spectrum. Scores for timber frame and ICF are noticeably worse, as shown by
the smaller circles which reach the lower end of the colour scale.

152
8.2.4. Non equal weightings

Further experiments were carried out using the case study to determine the effect of
altering the weightings on the results achieved and to evaluate the case study on
realistic terms to identify the best construction method for the situation examined. An
equal importance for all factors as used above is unlikely to occur in a real design
situation. The use of equal weightings in the previous sections of this chapter served to
demonstrate the use of the optimisation based methodology, however, equal
weightings is unrealistic. Evidence for this comes from the questionnaires, where
different factors received different scores for importance. This occurred even though
the factors were scored out of five for importance, rather than ranked, in the
questionnaire which would have allowed all factors to demonstrate equal importance if
this were the case. Two scenarios were examined, with focus on different aspects, to
determine the effect this would have on the result of the optimisation process.

Considering the results from the public survey in Chapter 4, the ranking of importance
factors and the results of the principal component analysis, a public importance
weighting was devised, this is shown in Table 8.2.

Table 8.2: Weighting focusing on the views of the public

Factor Weighting Justification

Embodied energy 1.00 Environmental impacts area of concern
Operational energy 1.00 Environment and running cost considered important
Public acceptability 1.00 Ability to sell property on indicated as important
Industry acceptability 0.50 Little concern displayed over this, provide customer
with what they want
Airtightness 0.75 Affects cost and environmental performance
Wall thickness 0.75 Connected to cost as thick walls mean larger area
per housing unit
Fire performance 0.25 Relates to risk but scores low as a concern

It should be noted, that although fire performance is given a low weighting, all the
construction methods examined meet the requirements for the Building Regulations
fire performance as a minimum. This is shown in Table 4.24. Any method which is
incapable of meeting the required fire standard should not be considered as a possible
construction method on safety grounds.

153
Re-running the case study optimisation methodology, but with these weighting
substituted for the original equal values, produced an optimal result of prefabricated
straw bale panels for construction method, and 6376mm for dimension (a). A two
dimensional representation of the fitness landscape is shown in Figure 8.4.

12
Brick and block
SIP
10
ICF
Single objective value

Prefabricated straw bale

8
Thin joint block work
Timber frame
6

Dimension a (m)

Figure 8.4- Two dimensional graphical display of the fitness landscape for case study
optimisation with public-focused weightings

Examination of Figure 8.4 shows that straw bale construction achieves the best single
object score at all values of dimension (a). Straw bale construction shows excellent
values of embodied energy, and good values of operational energy, airtightness and
fire performance. Although the scores achieved for acceptability and wall thickness
were low, the benefits outweighed these when examined with the public-focused
weightings. At values of (a) near the optimal, SIPs achieve sub optimal, but good
scores, this may be a result of the high embodied energy associated with SIPs. If this
were addressed, SIPs may prove to be the optimal construction method. Altering the
weighting has noticeably reduced the suitability of thin joint block work and brick and
block construction. The reduction in importance of fire performance will have affected
these materials. They score very well in terms of fire performance, so reducing the
impact of this will lower their score. The score of timber frame and ICF were largely
unaffected by the alteration in weightings, this indicates they are far from the optimal
solution.

154
A second set of weightings was generated based on the importance of factors found
from the construction industry survey. These are shown in Table 8.3.

Table 8.3: Weighting focusing on construction industry views

Factor Weighting Justification

Embodied energy 0.50 Environmental impacts area of some concern
Operational energy 1.00 Thermal efficiency given as most important concern
Public acceptability 0.75 Saleability of some concern
Industry acceptability 1.00 Method must be acceptable to themselves and
employees
Airtightness 1.00 Affects thermal efficiency
Wall thickness 0.75 Connected to cost as thick walls mean larger area
per housing unit
Fire performance 0.25 Of little concern to industry provided Building
Regulations requirements are met

The results of the case study optimisation when these weightings are substituted were
that the optimal material is SIPs with a value of 6377mm for dimension (a). Figure 8.5
shows the fitness landscape for the optimisation with construction industry weightings.

12
Brick and block
SIP
10
ICF
Single objective value

Prefabricated straw bale

8
Thin joint block work
Timber frame
6

Dimension a (m)

Figure 8.5- Two dimensional graphical display of the fitness landscape for case study
optimisation with construction industry-focused weightings

155
Examination of Figure 8.5 shows that SIPs are the optimal solution at all examined
dimensions of (a) with these weightings. SIPs achieve good values of operational
energy use, wall thickness, acceptability and airtightness, the combination of these
factors result in a good score for single objective value when these are highly ranked
by the weightings set. Prefabricated straw bale also achieved a good score; however
its low acceptability is likely to have reduced its suitability here. While ICF showed a
small improvement in score as a result of the weightings being based on the
construction industry’s interests, it is still outperformed by SIPs and prefabricated straw
bale. Timber frame showed little improvement as a result of the weightings change.
Although it is the most frequently used alternative method of construction in the area of
study these results indicate better alternatives exist in terms of selecting the
construction method based on energy use, acceptability and performance.

It can be seen that the weightings chosen can have a significant impact on the output
of the optimisation. The value of dimension (a) is dictated by the window area
constraint; however the material identified as optimal varies as weightings are
adjusted.

The two sets of weightings examined here were intended to show the impact of altering
the weightings and to highlight the importance of selecting suitable values. There are
many other parties who may suggest different weightings, for example, health and
safety, financial parties, and planners. Comparing the results of the optimisation run
with each of these different points of view has the potential to find the best compromise
in terms of interested parties, in addition to in terms of the factors included in the
optimisation. It is unlikely the optimal construction method for all interested parties
would be the same, due to different areas of concern. However, the selection of sub-
optimal but good options can allow the best compromise to be identified. Identifying
weightings which would represent the views of these parties is identified as an area for
further work in the development of the design methodology.

8.3. Comparison of case study results with brick and block construction
8.3.1. Equal weighting case study

No comparison was carried out for the results of the equal weighting case study as
brick and block construction was identified as the optimal method of construction. In
addition, as the equal weighting was considered unrealistic any comparison would not

156
have represented a realistic situation had an alternative method of construction been
identified.

8.3.2. Public weighting case study

Values were calculated for embodied and operational energy for both brick and block
and the recommended prefabricated straw bale panel versions of the case study. The
results of this are shown in Table 8.4. Performance and acceptability data for each are
also shown in Table 8.4, values taken from table 4.24.

Table 8.4- A comparison of prefabricated straw bale construction with the “typical”
method used in England and Wales for the case study house

Industry acceptability
Construction method

Public acceptability
Operational energy

(m /(h.m ) at 50Pa)
Embodied energy

Fire performance
Wall thickness
(kWh/m /year)

Air tightness

(minutes)
2
2
(MJ/m )
2

(m)

Brick and block 744.88 11.77 4.50 0.346 240 95 91

Prefab. Bale panels 467.91 8.0756 0.86 0.480 135 44 26
Building Regulations N/A N/A 10.00 N/A 60 N/A N/A

It can be seen from Table 8.4 that straw bale construction shows many benefits over
traditional brick and block construction. Values for embodied energy, operational
energy and airtightness all outperform those seen for traditional construction. The fire
performance is significantly less that that seen for brick and block construction,
although it outperforms that required by the Building Regulations, with almost double
the required time. This high level of performance was considered to be acceptable.
Wall thickness is greater than for brick and block; this can have impacts on the cost of
construction as fewer houses could be built in a given area. However, if the insulation
in the brick and block were increased to achieve comparable operational energy
values, the greater depth required would reduce the advantage of brick and block in

157
terms of wall thickness. The low acceptability of straw bale construction, both public
and industry, is the main challenge for this form of construction. Increasing awareness
of the benefits and proving that commonly held beliefs are incorrect could raise the
acceptability and improve the viability of this method of construction. For example, the
belief that straw bale construction burns easily when, in fact, the fire performance value
for prefabricated straw bale construction in Table 8.4 demonstrates this is not the case
(see Chapter 5 for further examples). Achieving this would allow the embodied energy
and operational benefits to be achieved, resulting in reduced energy use and
associated emissions. This could then help to achieve the emissions reduction targets
set out in the Climate Change Act (2006).

8.3.3. Construction industry weighting case study

Consideration was also given to the effect of using SIPs as suggested by the
construction industry focused case study run. Table 8.5 presents a comparison of SIP
construction with brick and block construction for the case study building, values taken
from Table 4.24.

Table 8.5- A comparison of the SIP construction method with the “typical” method used
in England and Wales for the case study house.

Industry acceptability
Construction method

Public acceptability
Operational energy

(m /(h.m ) at 50Pa)
Embodied energy

Fire performance
Wall thickness
(kWh/m /year)

Air tightness

(minutes)
2
2
(MJ/m )
2

(m)

Brick and block 744.88 11.77 4.50 0.346 240 95 91

SIP 1197.7 7.10 1.00 0.326 73 74.5 59
Building Regulations N/A N/A 10.00 N/A 60 N/A N/A

It can be seen that brick and block performs much better than SIP construction in
terms of embodied energy and fire performance. As the importance of the fire
performance falls due to the adjusted weightings, the other aspects in which SIP
construction performs well become more important. As a result of this, the single
objective score is lower and SIP construction becomes optimal. The high embodied

158
energy of SIP construction is undesirable; however, it may be that alternative cladding
methods, such as the use of brick slips could reduce this to a more comparable, or
lower, level.

The SIP version of the case study had a much lower annual operational energy per
square metre of floor area, a saving of 4.67MJ/m 2/year is seen. This would be
desirable for members of the construction industry who are looking towards the
requirement for building zero carbon homes from 2016. Airtightness achieved by SIPs
is much better, again something that would appeal to the construction industry who
must meet targets, Building Regulations as a requirement, but ideally lower ones set
by best practise guidelines. Wall thickness is somewhat improved when SIPs are
compared with brick and block. This would appeal to the construction industry as it
allows more units to be built in a given area; alternatively it could result in larger homes
which are more desirable and more appealing to the purchaser. SIPs show a fairly high
level of acceptability, both with the public and the construction industry. Although this is
lower than brick and block construction, education could improve this value by
publicising the benefits and dispelling those concerns stated in the survey responses
(see Chapter 5). For example, SIP construction was considered by some survey
respondents to be insufficiently strong for housing, in fact the strength of SIPs has
been tested and found to be acceptable by authors such as Mosey et al. (2009),
Hairstans and Kermani (2007) and Morley (2000).

8.4. Case study recommendations

The method considered to be most suitable for constructing the case study building is
SIPs. This is the optimal method when the construction industry weightings are used,
and sub-optimal, but good scoring when the public weightings are used. This was
considered to be the best compromise between the two interested parties. The main
disadvantage to SIP construction as examined in this work is the relatively high value
of embodied energy when compared with brick and block construction. The use of
alternative cladding, such as brick slips, may improve this situation by reducing the
embodied energy. The impact of this change on the acceptability of the method would
need to be examined as based on comments, in the public questionnaire the use of
alternative cladding is likely to negatively impact the acceptability, a brick external
appearance was desirable. In addition, some questionnaire respondents felt SIPs did
not appear to be sufficiently strong. Removing the brick outer skin may be perceived to
lower the strength of the system to a level the public would not trust, despite evidence

159
that SIP construction is able to perform sufficiently (Mosey et al. 2009, Hairstans and
Kermani 2007, Morley 2000).

As the data used in the design methodology was based on realistic values and the
case study was based on construction figures for England and Wales it is considered
that SIP construction may have the potential for use in speculative housing on a large
scale. As a result of increased use, savings would be seen in terms of operational
energy, and if alternative cladding were used in terms of embodied energy, contributing
to reduced emissions.

8.5. Summary
 A case study was designed based on housing figures for England and Wales.
 The optimisation based design methodology was used to identify the best
solution to the case study problem presented.
 The case study allowed the exploration of population size, number of
generations and crossover, and hence mutation, probabilities. These were set
at Population = 100, Number of generations = 200, PC= 0.4 and PM= 0.6.
 The solution identified by the methodology as optimal for the case study
scenario with equal weighting for all factors was brick and block with dimension
(a)= 6374mm.
 Fitness landscapes were examined both in three and two dimensional graphical
form. These can allow sub-optimal, but good, solutions to be identified and
allow for additional criteria to be considered.
 A phi array was also used to demonstrate the fitness landscape. This presents
a method for visualising the results. Sub optimal solutions can also be identified
from the phi array.
 Both fitness landscapes and Phi arrays identify the variable values which
violate the constraint applied by window area. These occur at low values of (a)
(below 6.5m). These values of (a) are undesirable as they would result in low
natural lighting and increased use of artificial light.
 The importance of weighting was demonstrated by running the case study
optimisation with different weightings.
 There was no impact on the optimal value of dimension (a) as weightings were
altered.

160
 A public focused weighting indicated that prefabricated straw bale construction
was optimal. SIP construction was sub-optimal but scored well.
 Industry focused weightings suggested the use of SIPs would be the most
effective compromise.
 A comparison of the embodied energy, operational energy and performance of
the suggested solution with traditional methods was carried out.
 Prefabricated straw bale panels outperformed brick and block construction in
terms of embodied and operational energy and airtightness. Wall thickness and
fire performance was lower than the traditional method, but still good. The very
low acceptability of this method is the major challenge it presents.
 SIP construction requires a greater amount of embodied energy for its
manufacture, it is suggested that alternative methods of cladding may solve
this. Areas that concern the construction industry such as operational energy
use and airtightness outperform the traditional method making it a good choice.
A reasonable level of acceptability is seen, this could be further improved with
education about the benefits of SIP construction.
 The combination of SIP construction as optimal for the industry weightings and
sub-optimal but good based on the public weightings supports its use in
housing construction in England and Wales.

161
9. Conclusions
9.1. Findings from the work

To examine the potential of alternative methods of wall construction to reduce energy

requirements associated with housing, six methods of construction were investigated in
detail. These included the “typical” method of brick and block construction, and a range
of alternative methods. A design methodology was created, using optimisation based
techniques to help identify the construction method and building size which give the
best compromise between energy use, acceptability and performance.

9.1.1. Embodied energy of construction methods

It was initially expected that alternative methods of construction may have lower values
of embodied energy, due to the presence of timber and reduced quantities of concrete.
However, it was found that materials such as foam insulation require large amounts of
energy to manufacture. The high performance achieved using these materials may
ultimately pay the high energy cost of production back. However, the opportunity to
save energy and the associated emissions at the manufacturing stage is lost if higher
embodied energy materials are used. Prefabricated straw bale panels were the only
method studied to demonstrate a lower embodied energy than brick and block
construction.

The high embodied energy of brick contributed significantly to the total embodied
energy of all those methods that included it as a material. Bricks accounted for 58% of
the embodied energy of brick and block, 30% for SIPs and ICF, 51% for thin joint block
work and 46% for timber frame. Using alternative methods of cladding, such as timber,
has the potential to reduce the embodied energy of new build housing. However,
comments from the public questionnaire strongly support a brick appearance.
Satisfying both the need for lower embodied energy and the desire for a “traditional”
appearance is a significant challenge. Materials such as brick slips may provide a
middle ground. They are most suitable for methods of construction such as ICF and
SIPs. Although they can be used with other methods of construction, such as timber
frame, extra materials required to apply the slips onto may negate the embodied
energy savings achieved by eliminating brickwork.

162
9.1.2. Acceptability of alternative methods of construction

The acceptability of the six methods of construction was examined to determine their
potential for use. Both public and construction industry opinions were examined via
questionnaires. Both groups showed a preference for the traditional method of
construction (brick and block) but some acceptability was seen for alternative methods.

The level of acceptability varied significantly across the methods. Thin joint block work,
SIPs and ICF showed a similar, fairly high, level of acceptability. Methods with high
acceptability are most likely to be successful in speculative construction as they have
greater appeal to the house buying market as a whole.

Timber frame and straw bale construction showed low acceptability. That there is some
acceptability indicates they do have the potential to be used in housing construction.
However, they may be more suited to individual houses, constructed to order with the
client who will live in the house being part of the consultation process.

A noticeable variation in acceptability was seen with geographical location. For

example, the acceptability of prefabricated straw bale panels was significantly lower in
Wales and London, it was higher than average in the South West, it was considered
that the rural nature of the South West may be responsible for this trend. The potential
for regional variation in acceptability should be considered when selecting a
construction method to use.

9.1.3. Areas of concern relating to housing

Concerns held by the public when considering a house to buy were examined in the
questionnaire. These areas were categorised as environmental factors, financial
factors and risk management. The highest scoring individual concerns were price,
need for maintenance and mortgage availability. For the construction industry the
factors with the greatest influence on construction method choice were identified as
thermal efficiency, cost and speed of construction. These factors should be considered
when making decisions related to the construction of housing, they are relevant to all
areas of the process, not just to wall construction methods.

163
9.1.4. Increasing acceptability

Questionnaire responses indicated that alternative methods which emulate the

properties of “traditional” construction they are more likely to be acceptable. Desirable
characteristics were identified as a traditional appearance i.e. brickwork outer, good
levels of insulation, high strength, good durability and proof of performance.

Education is key to increasing the use of alternative methods of construction. Many of

the concerns listed in the survey responses have already been disproven by existing
research and anecdotal evidence, as shown in Chapter 4. The desirable characteristics
listed above can also be achieved by alternative methods of construction. This
information needs to be demonstrated to purchasers and the construction industry.
Increasing awareness of the benefits of alternative methods could increase their
acceptability.

Financial benefits scored highly as a way of increasing interest in alternative methods

of construction. This may be from reduced construction costs or lower heating bills due
to better thermal performance. Low materials cost was identified as attractive to the
construction industry. If this can be achieved by alternative methods, they have a
higher likelihood of being adopted.

Guarantees were identified by survey respondents as something that would increase

their interest in alternative methods of construction. In addition to reducing concerns
relating to the construction method this provides a guarantee of the workmanship and
ensures houses are built to an acceptable standard. This was identified as an area
where the Government could show support for alternative methods of construction, by
backing guarantees for houses built using the alternative methods.

9.1.5. The optimisation based design methodology

It was noted that no single construction method examined presented a clear best
option; all methods had benefits and disadvantages. The key to identifying the optimal
solution is identifying the best compromise.

A fitness function, which combines values for embodied energy, operational energy,
acceptability and performance with weightings, was created to assess combinations of
construction method and building front dimension (a). Optimisation using genetic
algorithms was applied to the fitness function to allow a wide range of options to be

164
examined and the optimal solution to be identified. This represents the best
compromise, and the most suitable choice for the design. Sub-optimal solutions may
be better than the optimal solution in some cases, visual representations of the solution
space, such as Phi-arrays, help to identify these.

Possible uses for the methodology were identified as building design, where
construction method and building layout selection is necessary. Outputs from the
methodology can be used to explain the reasoning behind a construction method
selection. The methodology was designed to demonstrate the potential of the method,
there is wide opportunity for expansion, allowing more criteria to be examined.
Identifying the optimal solution and clearly demonstrating the reasons why it is optimal
can help to encourage support for the chosen solution. This may be from financial
parties, the Government or end users.

9.1.6. Case study solution

The use of the optimisation based design methodology was demonstrated with a case
study based on the most frequently constructed house type and size for England and
Wales. Case study values were used to determine the best values to use for
crossover, and hence mutation, fraction; for number of generations and size of
population.

From the optimisation of the case study it was seen that the solution which uses the
minimum amount of material will always be selected as favourable using the current
programme design. A small value of dimension (a) minimises the materials used, and
hence the embodied energy. The area which allows heat loss is also minimised,
reducing the operational energy requirements.

The minimum value of dimension (a) must allow a suitable amount of natural lighting.
This was achieved by the use of a simplified penalty constraint in the optimisation
calculation. From the case study results it can be seen that for terraced houses
meeting this constraint requires fairly large values of dimension (a), in excess of 6m.
The reduced wall area available for windows at small values of a makes it impossible
to satisfy both the 20% of floor area requirement and the 25% of wall area restriction.
Larger window areas could be used if they are thermally efficient. Alternatively, some
natural lighting could be provided by other measures, such as sun tunnels. Constraint

165
violation may be allowed if the housing design provides sufficient energy generation to
compensate for the increased artificial lighting, however, this is undesirable.

The case study examined indicates that if the aspects considered have equal
weighting, brick and block construction is optimal. Equal weighting for all factors was
used to demonstrate the functionality of the methodology and the ways in which the
solution landscape can be examined. However, equal weightings are not considered to
be realistic. Two sets of realistic weightings were created based on the questionnaire
findings.

The results from the case study with realistic weightings suggest that prefabricated
straw bale construction was optimal for the public viewpoint, with SIP construction sub-
optimal but achieving a good score. SIPs were optimal for the industry focused
weightings. That altering the weightings produced a different optimal construction
method for each set of values shows the importance of weightings and that they must
be carefully considered.

A major disadvantage of straw bale construction is the low acceptability seen in the
survey responses; this would need to be improved for straw bale construction to be
viable on a large scale. If this could be addressed the potential for low embodied
energy, thermally efficient houses this method offers could be accessed.

The fact that SIP construction was optimal for the construction industry weighting and
sub-optimal, but good scoring, for the public based weighting indicates it may be the
best option for speculative construction in England and Wales. It is the solution which
best meets the requirements of both parties. Energy use, performance and
acceptability are all good. The major disadvantage to this method is its high embodied
energy. It has been noted that the use of bricks as an outer skin contributes
significantly to the embodied energy of this method. If the bricks were replaced with an
alternative cladding, SIP construction may equal or outperform brick and block
construction in terms of embodied energy. The impact on acceptability, both due to
appearance and perceived strength must be carefully considered if this approach is
taken.

The operational energy saving achievable by SIPs, compared to brick and block
construction has been examined using the case study dimensions. An operational
energy saving of 4.67MJ/m2/year was calculated. If this was achieved across all new
build houses constructed in England and Wales the savings in energy use and the

166
associated emissions would play a significant part in achieving emissions reduction
targets.

9.2. Conclusions and recommendations

The following recommendations are made based on the findings of this work.

 Brick work has been identified as a major contributor to the embodied energy of
construction. Replacing this with a less energy intensive material could reduce
the embodied energy of construction and result in significant energy savings
and the reduced emissions associated with them. However, a brick finish has
been identified as strongly desirable to house purchasers. These two factors
must be balanced to achieve the best combination of saleability and embodied
energy. Brick slips may present a solution in some cases.
 Education to improve the understanding of alternative construction methods is
a key step in raising their acceptability and accessing the benefits they present.
Awareness of the existence of alternative methods of construction and the
benefits they offer should be encouraged. This could be by the use of media, as
seen with straw bale construction becoming better known due to the television
programme “Grand Designs”. Greater education would also help to dispel fears,
such as fire performance being poor for straw bale construction.
 Government support for housing developments made using alternative
methods of construction would allow them to become examples which can then
demonstrate the benefits of the alternative methods. For example, actual use
figures for energy requirements, durability and maintenance requirements.
 Guarantees were identified as a method of increasing the level of interest in
alternative methods of construction. These could be offered by construction
firms; however this relies on support from the construction industry for the
alternative method. This is also an area where Government support could play
a role.
 Areas to consider when making decisions related to housing construction
include environmental factors, financial factors and risk management. Price
was of particular importance to the public. Although this was not covered in the
optimisation based design methodology it could be assessed using a similar
method to that demonstrated for embodied energy.

167
  An optimisation based design methodology has been devised which can be
used to aid decision making. Opportunities exist for it to be expanded and
modified to suit a range of purposes. Using phi arrays to identify sub optimal
but high scoring solutions allows other factors to be included in the decision
making process. Phi arrays present a way of showing the solution space that is
easy to understand without specialist knowledge. This would be useful at the
design phase when options are being discussed by a wide range of interested
parties. Potential areas for expansion of the methodology are discussed in
Chapter 10.
 Potential users of the methodology include designers. However it also has
applications as a result of increasing the transparency of the decision making
process. Low environmental impact construction methods can be identified;
these could be used as the basis for Government backing of a particular
construction method.
 SIP construction has been identified as a good choice for achieving lower
operational energy, higher performance, housing construction in England and
Wales. To support this, further research into the method would be beneficial,
accompanied by dissemination of the findings to a wide audience. Making the
public and industry aware of the benefits and disadvantages is key to
increasing the acceptability of this method and hence it’s economic viability.
Government support for the method, for example, by funding research or
backing guarantees could also help the method be used on a wider scale, and
the reduced operational energy requirements it offers to be accessed.

9.3. Contribution to knowledge

The contribution to knowledge provided by the work included in this thesis is:

 The performance and energy associated with the construction methods studied
has been reviewed. In particular brick has been identified as a high embodied
energy material, finding alternatives to this, such as brick slips, has been
discussed.
 The public and industry acceptability of the six construction methods studied
has been examined. Alternative methods of construction with high acceptability
are Structural Insulated Panels, Insulating Concrete Formwork and Thin joint
block work. Prefabricated straw bale panels and timber framed construction

168
 have low acceptability with the public. Timber framed construction has
moderate acceptability with the construction industry.
 Desirable characteristics for wall construction methods were identified as a
traditional appearance i.e. brickwork outer, good levels of insulation, high
strength, good durability and proof of performance.
 Education was identified as the key to improving the acceptability of alternative
methods of construction. Many of the issues identified with them have been
disproven by testing or anecdotal evidence, by disseminating this information
the acceptability of alternative methods may increase.
 An optimisation based design methodology for the selection of construction
methods that considers energy use, acceptability and performance to identify
the best compromise for a given situation has been created. The use of this
was demonstrated with a case study. Areas have been identified for further
work to expand this into a tool which can be used in a design environment

The case study examined identified SIP construction as the most suitable construction
material, balancing the points of view of the public and industry. This construction
method is considered to have the greatest potential for reducing the energy
requirements of housing construction whilst maintaining performance and economic
viability. Further investigation into this is recommended as the case study used in this
work was limited by the number of performance indicators used.

169
9.4. Further work

There is enormous potential for the expansion of the optimisation based methodology
developed in this thesis. This work intended to demonstrate the potential of
optimisation in this situation and some of the methods that could be used to generate
data. Suggestions for expanding the methodology include:

 The inclusion of other construction methods, for example those discarded

during the method selection (see Chapter 4.3).
 Altering specification of construction methods. For example, thicker SIP
panels, greater depth of insulation in thin joint block work. This would not
require additional acceptability information; energy values and performance
would be affected.
 Altering the cladding material has the potential to affect all aspects considered.
To include different cladding options a further survey would have to be carried
out to determine acceptability.
 Further aspects of the construction methods’ performance could be considered
in the design methodology. Examples include, cost, durability, maintenance
requirements, time to construct and construction waste. The methods
demonstrated can be adapted to these factors.
 Embodied energy assessment could be expanded to include transportation
and site activities.
 Alteration of the degree day values used. This may be to focus the
methodology on a particular location or to use a different degree day base
temperature.
 Cooling degree days could be considered if the methodology was to be used in
a country or region which experiences high temperatures during part of the
year.
 The development of weightings for additional parties with an interest in
construction so their views can be incorporated into the decision making
process, for example, health and safety and planners.

It was noted that the use of brick has a significant impact on the embodied energy of
the construction methods. It would be useful to consider the effect of different types of
cladding on the total embodied energy of construction methods. This could be
independent of the optimisation methodology; however the results would apply to the

170
methodology. The acceptability of alternative cladding methods would need to be
examined as many survey comments indicated a preference for a brick appearance
externally.

171
12. References

Abeysundara, U.G.Y., Babel, S. & Gheewala, S., 2009. A matrix in life cycle perspective for
selecting sustainable materials for buildings in Sri Lanka. Building and Environment, 44(5),
pp.997-1004.

Al-Homoud, M.S., 2005. Performance characteristics and practical applications of common

building thermal insulation materials. Building and Environment, 40(3), pp.352-366.

Anastaselos, D., Giama, E. & Papadopoulos, A.M., 2009. An assessment tool for the energy,
economic and environmental evaluation of thermal insulation solutions. Energy and
Buildings, 41(11), pp.1165-1171.

Anderson, B., 2006. Conventions for U-Value calculations. Watford: BRE Press

Anink, D., Boonstra, C. & Mak, J., 1996. Handbook of sustainable building. London: James
and James (Science Publishers) Ltd.

Apte, V., Griffin, G.J., Paroz, B.W. & Bicknell, A.D., 2008. The fire behaviour of rendered
straw bales. Fire and materials, 32(5), pp.259-279.

Arnaud, L., La Rosa, C. & Sallet, F., 2009. Mechanical behaviour of straw construction
following the GREB technique. In: P. Walker, K. Ghavami, K. Paine, A. Heath, M. Lawrence,
E. Fodde, eds. Proceedings of the 11th International Conference on Non-conventional
Materials and Technologies (NOCMAT 2009). Bath, U.K., 6-9 September 2009. Bath:
University of Bath.

Barista, D., 2008. Building Tight with SIPs. [Online]. Available at:
http://www.bdcnetwork.com/building-tight-sips?page=show [Accessed 16 September 2011].

Barista, D., 2009. Tall ICF walls. Building Design and Construction, August 2009, pp.16-17.

Barlow, J., 2000. Private sector housebuilding: structure and strategies into the 21st century.
London: Council of Mortgage Lenders.

Barlow, J., 1999. From Craft Production to Mass Customisation. Innovation Requirements for the UK
Housebuilding Industry. Housing Studies, 14(1), pp.23-42.

Bartlett, J. E. II, Kotrlik, J. W. and Higgins, C., 2001. Organizational research: Determining appropriate
sample size for survey research. Information Technology, Learning, and Performance Journal, 19(1)
43-50.

172
Beco Products Ltd., 2008. BecoWallform Insulating concrete formwork system. April 2008
[Brochure].

Björklund, A.E., 2002. Survey of Approaches to Improve Reliability in LCA. International

Journal of Lifecycle Assessment, 7(2) pp.64-72

BRE Ltd., 2013. About the BRE Watford Innovation Park. [Online] Available at:
http://www.bre.co.uk/innovationpark/page.jsp?id=2985 [Accessed 1 May 2013]

BRE Ltd., 2011. The Passivhaus Standard. [Online] Available at:

http://www.passivhaus.org.uk/standard.jsp?id=122 [Accessed 2 March 2013].

BRE Global Ltd., 2008. Green Guide 2008 ratings, External walls. [Online] Available at:
http://www.bre.co.uk/greenguide/ggelementtype.jsp?buildingType=Housing&category=6
[Accessed 14 September 2012].

Bregulla, J. & Enjily, V., 2004. An introduction to building with Structural Insulated Panels
(SIPs). Watford: BRE Press.

British Standards Institution, 2005. BS5628-3:2005 Code of practice for the use of masonry.
Materials and components, design and workmanship. London: BSI.

British Standards Institution, 2007. BS EN ISO 10456:2007. Building materials and products.
Hygrothermal properties. Tabulated design values and procedures for determining declared
and design thermal values. London: BSI.

Buchanan, A.H. & Honey, B.G., 1994. Energy and carbon dioxide implications of building
construction. Energy and Buildings, 20(3), pp.205-217.

Bunday, B., 1984. Basic Optimisation methods. London: Edward Arnold.

Calkins, M., 2009. Materials for sustainable sites. Hoboken, USA: John Wiley & Sons, Inc.

Canada Mortgage and Housing Corporation, 2000a. Straw Bale House Moisture Research.
Canada: Canada Mortgage and Housing Corporation.

Canada Mortgage and Housing Corporation, 2000b. Moisture Properties of Plaster and
Stucco for Straw Bale Buildings. Canada: Canada Mortgage and Housing Corporation.

Canada Mortgage and Housing Corporation, 2008. Effect of Mesh and Bale Orientation on
the Strength of Straw Bale Walls. Canada: Canada Mortgage and Housing Corporation.

173
Carfrae, J., DeWilde, P., Littlewood, J., Goodhew, S., & Walker, P., 2009. Long term
evaluation of the performance of a straw bale house built in a temperate maritime climate. In:
P. Walker, K. Ghavami, K. Paine, A. Heath, M. Lawrence, E. Fodde, eds., Proceedings of
the 11th International Conference on Non-conventional Materials and Technologies
(NOCMAT 2009). Bath, U.K., 6-9 September 2009. Bath: University of Bath.

Castro-Lacouture, D., Sefair, J.A., Flórez, L. & Medaglia, A.L., 2009. Optimization model for
the selection of materials using a LEED-based green building rating system in Colombia.
Building and Environment, 44(6), pp.1162-1170.

Cathcart, C.M., 1998. SIPs, not studs. Architecture, 87(6), pp.148-152.

Cavill, N., 1999. Timber’s back in the frame. Building, 31. Available online:
http://www.building.co.uk/news/timber8217s-back-in-the-frame/4474.article [Accessed 1
December 2012]

Chinyio, E.A., Olomolaiye, P.O. & Corbett, P., 1998. An evaluation of the project needs of
UK building clients. International Journal of Project Management, 16(6), pp.385-391.

Chown, I., 1999. Houses and flats. In: D.Adler ed. Metric handbook, planning and design
data. London: Elsevier Ltd.. Ch. 33.

Chudley, R. & Roger Greeno, R., 2004. Building construction handbook. 5th ed., Oxford:
Elsevier Butterworth-Heinemann.

CIBSE- Chartered Institution of Building Services Engineers, 2006. Degree-days: Theory

and application, TM41. London: Chartered Institution of Building Services Engineers.

Climate Change Act 2008 (c.27), London: HMSO.

Communities Scotland, 2002. Scottish Housing condition survey 2002. Edinburgh

Davies, C., 2006. Working details/styro stone house. Concrete Quarterly, March 2006,
pp.54-55.

Denzer, A.S. & Hedges. K.E., 2007. Sustainable construction in Yellowstone Park: A case
study. Eds: Chun, Y.-M. Claisse, P. Naik, T.R. & Ganjian, E. Sustainable Construction
Materials and Technologies. Coventry, UK. 11-13 June 2007. Taylor & Francis: London.

Department for Communities and Local Government (DCLG), 2007. Building a Greener
Future: Towards Zero Carbon Development. London: HMSO.

174
Department for Communities and Local Government (DCLG), 2008. Definition of zero
carbon homes and non-domestic buildings. London: HMSO.

Department for Communities and Local Government (DCLG), 2009. The code for
sustainable homes, case studies. London: Department for Communities and Local
Government.

Department for Communities and Local Government (DCLG), 2010. English Housing survey,
Housing stock report 2008. London: HMSO.

Department for Communities and Local Government (DCLG), 2011. Accredited construction
details for Part L. [Online] Available at:
http://www.planningportal.gov.uk/buildingregulations/approveddocuments/partl/bcassociated
documents9/acd [Accessed 15 April 2013].

Department for Communities and Local Government (DCLG), 2012a. Table 209: Permanent
dwellings completed, by tenure and country. [Online] Available at:
http://www.communities.gov.uk/housing/housingresearch/housingstatistics/housingstatisticsb
y/housebuilding/livetables/ [Accessed 11 September 2012].

Department for Communities and Local Government (DCLG), 2012b. Table 254: Permanent
dwellings completed by house and flat, number of bedroom and tenure ENGLAND. [Online]
Available at:
http://www.communities.gov.uk/housing/housingresearch/housingstatistics/housingstatisticsb
y/housebuilding/livetables/ [Accessed 5 September 2012].

Department for Communities and Local Government (DCLG), 2013a. Building Regulations
Approved Documents. [Online] Available at:
http://www.planningportal.gov.uk/buildingregulations/approveddocuments/ [Accessed 15
April 2013].

Department for Communities and Local Government (DCLG), 2013b. Approved Document
L- Conservation of fuel and power. [Online] Available at:
http://www.planningportal.gov.uk/buildingregulations/approveddocuments/partl/approved
[Accessed 15 April 2013].

Department for Communities and local government (BRE authors), 2008. Innovative
Construction Products and Techniques. London: Department for Communities and Local
Government.

175
Department of Energy and Climate Change, 2012. Carbon emissions reduction target
(CERT). [Online] Available at:
http://www.decc.gov.uk/en/content/cms/funding/funding_ops/cert/cert.aspx [Accessed 10
August 2012].

Diakaki, C., Grigoroudis, E., Kabelis, N., Kolokotsa, D., Kalaitzakis, K. and Stavrakakis, G.,
2010. A multi-objective decision model for the improvement of energy efficiency in buildings.
Energy, 35(12), pp.5483-5496.

Dye, A. & McEvoy, M., 2008. Environmental Construction Handbook. London: RIBA
Publishing.

Emmanuel, R., 2004. Estimating the environmental suitability of wall materials: preliminary
results from Sri Lanka. Building and Environment, 39(10), pp.1253-1261.

Emmitt, S. & Yeomans, D.T., 2008. Specifying Buildings, A Design Management

Perspective. 2nd edn. Oxford:Butterworth-Heinemann

Energy Saving Trust, 2011. Enhanced construction details. [Online] Available at

http://www.energysavingtrust.org.uk/Organisations/Innovation/Free-resources-for-housing-
professionals/New-build/Enhanced-construction-details: [Accessed 15 April 2013]

Evans, B., 2006. The house took around 35 trucks of structural-grade ready mix. Concrete
Quarterly, March 2006, pp.52-53.

Field, A., 2009. Discovering statistics using SPSS. 3rd ed. London: Sage Publications Ltd.

Fink, A., 2003. The Survey Handbook. 2nd ed. London: Sage Publications Ltd.

Fink, A., 2006. How to conduct surveys. 3rd ed. London: Sage Publications Ltd.

Funke, H., 2006. Insulated concrete formwork- sustainable, durable and modern. Concrete,
October 2006, pp.27-28.

Gaze, C., 2008. Applying the code for sustainable homes on the BRE innovation park.
Watford: IHS BRE Press.

Gen, M. and Cheng, R., 1997.Genetic Algorithms and Engineering Design. New York: John
Wiley and Sons.

Gillham, B., 2000. Developing a questionnaire. London: Continuum.

176
Glidevale, 2012. Build Tight Ventilate Right. [Online] Available at:
http://www.glidevale.com/buildtight.html Accessed [Accessed 17 August 2012].

Goodier, C. & Gibb, A., 2007. Future opportunities for offsite in the UK. Construction
Management and Economics, 25(6), pp.585-595.

Goldberg, D. 1989. Genetic algorithms in Search, Optimization and Machine learning.

Reading, Massachusetts: Addison-Wesley Publishing Company Inc.

González, M.J. & Navarro, J.G., 2006. Assessment of the decrease of CO2 emissions in the
construction field through the selection of materials: Practical case study of three houses of
low environmental impact. Building and Environment, 41(7), pp.902-909.

Goodhew, S., Griffiths, R., de Wilde, P. & Simmons, T., 2007. Monitoring of strawbale and
non-food-crop based walling systems. Eds: Chun, Y.-M. Claisse, P. Naik, T.R. & Ganjian, E.
Sustainable Construction Materials and Technologies. Coventry, UK. 11-13 June 2007.
Taylor & Francis: London.

Goodhew, S., Griffiths, R. & Woolley, T., 2004. An investigation of the moisture content in
the walls of a straw-bale building. Building and Environment, 39(12), pp.1443-1452.

Greeley, T.R., 1997. A review of expanded polystyrene (EPS) properties, performance and
new applications. Insulation Materials: Testing and Applications: Third Volume. U.S.A.:
American Society for Testing and Materials.

Griffen, G.J., Bicknell, A.D., Bradbury, G.P. & White, N., 2006. Effect of Construction Method
on the Fire Behaviour of Sandwich Panels with Expanded Polystyrene Cores in Room Fire
Tests. Journal of Fire Sciences. 24(4), pp.275-294.

Griswold, M., 2007. Building a following: ICFs picking up fans. Plastics News, 19(10).

Hairstans, R. & Kermani, A., 2007. Briefing: Structural insulated panels in modern
construction. Proceedings of the ICE - Construction materials, 160(3), pp.91-94.

Hamilton-MacLaren, F., Loveday, D.L. and Mourshed, M., 2009. The calculation of
embodied energy in new build UK housing. In: A. Dainty. ARCOM (Association of
Researchers in Construction Management) 25th Annual Conference. Nottingham, UK 7-9
September . Reading: ARCOM.

Hammond, G.P. & Jones, C.I., 2008. Embodied energy and carbon in construction materials.
Energy, 161(2), pp.87-98.

177
Hammond, G.P. & Jones, C.I., 2012. Inventory of carbon and energy (ICE), version 2.0.
[Online] (updated 2011) Available at http://www.siegelstrain.com/site/pdf/ICE-v2.0-summary-
tables.pdf [Accessed 10 September 2012].

Harris, D.J., 1999. A quantitative approach to the assessment of the environmental impact of
building materials. Building and Environment, 34(6), pp.751-758.

Hastings, T., 2010. VC Foil ultra insulating vapour control layer. [Online] Available at:
http://www.glidevale.com/downloads/protect_vc_foil_ultra.pdf Accessed [Accessed 17
August 2012].

TM
Hastings, T., 2011. Buildtight Range. [Online] Available at:
http://www.glidevale.com/downloads/airtightness_buildtight_range.pdf Accessed [Accessed
17 August 2012].

HM Government, 2006. Approved Document to support regulation 7, Materials and

workmanship. Newcastle Upon Tyne: NBS.

HM Government, 2010. Approved Document L1A, Conservation of fuel and power in new
dwellings. Newcastle Upon Tyne: NBS.

HM Government, 2011. Approved Document B (Fire safety), Volume 1- Dwellinghouses.

Newcastle Upon Tyne: NBS.

Hutton, T.C., Lloyd, H. & Singh, J., 1992. The environmental control of timber decay.
Structural Survey, 10(1), pp.5-20.

Insulating Concrete Formwork Association, 2003. High Performance Buildings using

insulating concrete formwork. Camberley: The Concrete Centre.

Insulating Concrete Formwork Association, 2009. Insulating Concrete Formwork, a guide to

use and application. Camberley: The Concrete Centre.

Intergovernmental Panel on Climate Change (IPCC). 2007. In: Solomon, S., D. Qin, M.
Manning, Z. Chen, M.Marquis, K.B. Averyt, M. Tignor, H.L. Miller, eds. Summary for
Policymakers. In: Climate Change 2007: The Physical Science Basis. Contribution of
Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on
Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York,
NY, USA.

Internetworldstats, 2004.United Kingdom Internet usage stats and market profile. [Online]
Available at: http://www.internetworldstats.com/eu/uk.htm [Accessed 16 September 2011].

178
Johnston, D. & Gibson, S., 2008. Green from the Ground Up. Newtown, Connecticut, USA:
The Taunton Press.

Jones, B., 2007. Building with Straw Bales. 2nd ed. Totnes: Green Books.

King, B., 1997. Buildings of Earth and Straw: Structural Design for Rammed Earth and Straw
Bale Architecture. California, USA: Ecological Design Press.

Kingspan, 2009. Kingspan TEK Building System, Product Information. 7th ed. April 2009.
[Brochure].

Knight, T., Singlair, M. and Roberts, T., 2009. Silvatec Design, Timber frame standard
details September 09. [Online] Available at:
http://www.silvatecdesign.com/Technical%20Downloads/Silvatec%20Design%20Standard%
20Details%20-%20September%2009.pdf [Accessed 17 August 2012].

Kośny, J., Christian, J.E., Desjarlais, A.O., Kossecka, E. & Berrenberg, L., 1998. The
Performance Check Between Whole Building Thermal Performance Criteria and Exterior
Wall Measured Clear Wall R-Value, Thermal Bridging, Thermal Mass and Airtightness.
U.S.A.: ASHRAE.

Kothari, C.R., 2004. Research Methodology: Methods and Techniques. 2nd ed. New Delhi:
New age international (P) Ltd.

Lacinski, P. & Bergeron, M., 2000. Serious straw bale. Canada: Chelsea Green Publishing
Company.

Lagaros, N. D., Papadrakakis, M. and Plevris, V., 2005, Multiobjective Optimization of Space
Structures under Static and Seismic Loading Conditions. In: A. Abraham, L. Jain, R.
Goldberg, eds. 2005. Evolutionary Multiobjective Optimization. London: Springer-Verlag. Ch.
12.

Lawrence, M., Heath, A., Walker, P., 2009. the impact of external finishes on the weather
resistance of straw bale walls. In: P. Walker, K. Ghavami, K. Paine, A. Heath, M. Lawrence,
E. Fodde, eds., Proceedings of the 11th International Conference on Non-conventional
Materials and Technologies (NOCMAT 2009). Bath, U.K., 6-9 September 2009. Bath:
University of Bath.

Lecompte, J.G.N, 1987. Airtightness of masonry walls. In: AIVC Ventilation technology-
Research and application. Uberlingen, Federal Republic of Germany 21-24 September
1987.

179
Lennon, T., Bullock, M.J. & Enjily, V., 2000. The fire resistance of medium-rise timber framed
buildings. World Conference on Timber Engineering. Whistler Resort, British Columbia,
Canada, July 31- August 3 2000. [Online]. Available at:
http://timber.ce.wsu.edu/Resources/papers/4-5-4.pdf [Accessed 10 September 2011].

Li, H. & Shen, Q., 2002. Supporting the decision-making process for sustainable housing.
Construction Management and Economics, 20(5), pp.387-390.

Lorenzoni, I., Nicholson-Cole, S. & Whitmarsh, L., 2007. Barriers perceived to engaging with
colmate change among the UK public and their policy implications. Global Environmental
Change, 17(3-4), pp.445-459.

Low Carbon Construction Innovation and Growth Team, 2010. Final report, Executive
summary. London: Department for Business, Innovation and Skills.

Mahaptra, K., Gustavsson, L. & Hemstrom, K., 2012. Multi-storey wood-frame buildings in
Germany, Sweden and the UK. Construction Innovation, 12(1) pp.62-85.

Mackinder, M., 1980. The Selection and Specification of Building Materials and
Components, Research Paper 17. York: University of York Institute of Advanced
Architectural Studies

Mackinder, M. & Marvin, H., 1982. Design Decision Making in Architectural Practice,
Research Paper 19. York: University of York Institute of Advanced Architectural Studies

MathWorks Inc., 2012. Matlab. [Online] Available at:

http://www.mathworks.co.uk/products/matlab/.

McKeown, J.J., Meegan, D. And Sprevak, D., 1990. An introduction to unconstrained

optimisation. Bristol: IOP Publishing Ltd.

McMeeken, R., 2009. Top 150 contractors and housebuilders 2009. [Online] Available at:
http://www.building.co.uk/story.asp?sectioncode=32&storycode=3146125 [Accessed 3
March 2010].

Meteonorm, 2011. Meteonorm plugin. [Online] Available at: http://meteonorm.com/home/. [Accessed

10 July 2011].

Miles-Shenton, D., Wingfield, J. And Bell, M., 2007. Interim Report Number 6 – Airtightness
Monitoring, Qualitative Design and Construction Assessments. Partners in Innovation Project: CI
39/3/663. Leeds, Leeds Metropolitan University.

Miller, R.E., 2012. Beco wallform. [email](Personal communication 11 January 2012).

180
Miller, D.C. & Salkind, N.J., 2002. Handbook of research design and social measurement. 6th
ed. London: Sage Publications Ltd.

Mitchell, M., 1996. An Introduction to Genetic Algorithms. Cambridge, Massachusetts: A

Bradford Book.

Modcell, 2010. Modcell- structural. [Online] Available at:

http://www.modcell.co.uk/page/structural [Accessed 3 March 2010].

Modcell, 2012a. Modcell- technical. [Online] Available at: http://www.modcell.com/technical/

[Accessed 7 August 2012].

Modcell, 2012b. Modcell tech sheet. [Online] Available at:

http://www.modcell.com/files/4913/1047/4999/ModCell_Technical_sheet_2011_07_12a.pdf
[Accessed 7 August 2012].

Moedinger-Clay, M.W., 2007. Everything begins with a tight envelope. Environmental Design
and Construction, July 2007, pp.24-30.

Monahan, J. And Powell, J.C., 2011. An embodied carbon and energy analysis of modern
methods of construction in housing: A case study using a lifecycle assessment framework.
Energy and Buildings, 43(1), pp. 179-188.

Morel, J.C., Mesbah, A., Oggero, M. & Walker, P., 2001. Building houses with local
materials: means to drastically reduce the environmental impact of construction. Building
and Environment, 36(10), pp.1119-1126.

Morley, M., 2000. Building with Structural Insulated Panels (SIPS). Newtown, Connecticut,
USA: The Taunton Press, Inc.

Mosey, G., Doris, E., Coggeshall, C., Antes, M., Ruch, J. & Mortensen, J., 2009. Conceptual
Soundness, Metric Development, Benchmarking, and Targeting for PATH Subprogram
Evaluation. U.S.A.: U.S. Department of Housing and Urban Development (HUD) Partnership
for Advancing Technology in Housing.

Mourshed, M., 2012. Relationship between annual mean temperature and degree-days.
Energy and Buildings, 54, pp. 3737-3746.

Mourshed, M., Kelliher, D. And Keane M.M., 2003. ArDOT: A tool to optimize environmental
deisgn of buildings. In: Building Simulation 2003- Eighth International IBPSA Conference.
Eindhoven, Netherlands: International Building Performance Simulation Association
(IBPSA). Pp 919-926.

181
Mourshed, M., Shikder, S. and Price, A.D.F, 2011. Phi-array: a novel method for fitness
visualization and decision making in evolutionary design optimization. Advanced Engineering
Informatics, 25(4), pp.676-687.

NHBC, 2007. New House-Building Statistics, 2007 Quarter 2. Amersham: NHBC.

NHBC Foundation 2013. Lessons from Germany’s Passivhaus experience. [Online]

Available at:
http://www.nhbcfoundation.org/Researchpublications/LessonsfromGermanysPassivhausexp
erienceNF47/tabid/527/Default.aspx [Accessed 2 March 2013]

North Kesteven District Council, 2012. NK Straw Houses. [Online] Available at: http://www.n-
kesteven.gov.uk/nk-straw-houses/106433.article [Accessed 1 May 2013].

Office for National Statistics, 2011. [Online] Available at:

http://www.ons.gov.uk/ons/index.html [Accessed 10 February 2011].

Oppenheim, A.N., 1992. Questionnaire Design, Interviewing and Attitude Measurement. 2nd
ed. London: Continuum.

Osmani, M. & O’Reilly, A., 2009. Feasibility of zero carbon homes in England by 2016: A
house builder’s perspective. Building and Environment, 44(9), pp.1917-1924.

Pan, W., Gibb, A.G.F. & Dainty, A.R.J, 2007. Perspectives of UK housebuilders on the use
of offsite modern methods of construction. Construction Management and Economics, 25(2),
pp.183-194.

Pierquet, P., Bowyer, J.L. & Huelman, P., 1998. Thermal performance and embodied energy
of cold climate wall systems. Forest Products Journal, 48(6).

Rao, V.R., 2006. A material selection model using graph theory and matrix approach.
Materials Science and Engineering, A 431, pp.248-255.

Ravetz, J., 2008. State of the Stock- What do we know about existing buildings and their
future prospects? Energy Policy, 36(12), pp.4462-4470.

Rees, R., 2011. Completions statistics. [email] (Personal communication 8 July 2011).

Ross, K., 2008. Learning the lessons from systemic building failures. Watford: IHS BRE
Press.

Roy, R. And Cochrane, S.P., 1999. Development of a customer focused strategy in

speculative house building. Construction Management and Economics, 17(6), pp.777-787.

182
Rudd, A. & Chandra, S., 1994. Side-by-Side Evaluation of a Stressed-Skin Insulated-Core
Panel House and a Conventional Stud-Frame House. Cape Canaveral: Florida Solar Energy
Center.

Schnettler, S., 2009. A structured overview of 50 years of small-world research. Social

Networks, 31(3), pp.165-178.

Scott Wilson Ltd., 2010. Dwelling size survey. [pdf] CABE. [Online] Available at:
http://webarchive.nationalarchives.gov.uk/20110118095356/http:/www.cabe.org.uk/files/dwel
ling-size-survey.pdf [Accessed 21 August 2012].

Seo, S., Aramaki, T., Hwang, Y. & Hanaki, K., 2004. Fuzzy Decision-Making Tool for
Environmental Sustainable Buildings. Journal of Construction Engineering and Management,
130(3), pp.415-423.

Seyfang, G., 2010. Community action for sustainable housing: Building a low carbon future.
Energy Policy, 38(12), pp.7624-7633.

Smith Jr. C.R., 2007. High-performance straw bale. Environmental Design and Construction,
May, pp.60-68.

Stubbs, M., 2002. Car Parking and Residential Development: Sustainability, Design and
Planning Policy, and Public Perceptions of Parking Provision. Journal of Urban Design, 7(2),
pp.213-237.

Summers MD (2006). Moisture and decomposition in straw: implications for straw bale
construction. In: King B, editor. Design of straw bale buildings. San Rafael (CA): Green
Building Press; 2006. p. 162–72.

Survey monkey, 2011. [Online] Available at: https://www.surveymonkey.com/. [Accessed 10

January 2011].

Tam, C.M., Tong, T.K.L. & Zhang, H., 2007. Decision making and Operations research
techniques for construction management. Hong Kong: City University of Hong Kong Press

Tarmac Topblock Ltd., 2006. Durox System- thin joint blockwork. [Brochure].

Tas, E., Yaman, H. & Tanacan, L., 2008. A building material evaluation and selection model
for the Turkish construction sector. Engineering, Construction and Architectural
Management, 15(2), pp.149-163.

183
The Brick Industry Association, 2008. Technical notes on Brick Construction- Fire resistance
of Brick Masonry.

Tricker, R., 2004. Building Regulations in Brief. 2nd ed. Oxford: Elsevier Butterworth-
Heinemann.

UKSIPS, 2012. This is UK SIPS. Available online: http://www.uksips.org/ [Accessed 3

January 2013].

United Nations, 1998. Kyoto Protocol to the United Nations Framework Convention on
Climate Change.

White, C., 2011. Modcell information. [email](Personal communication 23 March 2011).

White, L., 2007. Structural insulated panels offer rapid low-cost building and cut energy use.
Urethanes technology, 23(6).

World Health Organisation (WHO), 1979. WHO Technical Report series, No. 225, 1979,
Public Health aspects of housing: First report of the expert committee. Geneva: WHO

Yates, T., 2006. The use of non-food crops in the UK construction industry. Journal of the
Science of Food and Agriculture, 86, pp.1790-1796.

Zero Carbon Hub, 2011. LZ Carbon profile, Profile 028, February 2011. [Online] Available
at: http://www.zerocarbonhub.org/examplepdfs/P028_Rowner_2011228_11239125.pdf
[Accessed 26 August 2012].

Publications produced from this work

Hamilton-MacLaren, F., Loveday, D.L.L & Mourshed, M., 2012. Public opinions on
alternative lower carbon wall construction techniques for UK housing. Habitat International,
37, pp.163–169

Hamilton-MacLaren, F., Loveday, D.L.L & Mourshed, M., 2009. The calculation of embodied
energy in new build UK housing. IN: Dainty, A.R.J. (ed). Proceedings of the 25th Annual
ARCOM Conference, Nottingham, September 2009. Reading: ARCOM

184
Appendix A- Sample survey, house holder

185
186
187
188
189
190
191
192
193
194
195
196
197
Appendix B- Sample survey, house builder

198
199
200
201
202
203
204
205
206
207
208
209
Appendix C- Acceptability of construction methods

210
Results from the public survey in response to the question “Would you buy a house
built from this material are presented below in Table C.1- C.6. A separate table is
presented for each method of construction.

Table C.1- Acceptability of brick and block construction by region

Percentage response to the question

“Would you buy a house built from this
material?”
Region Yes No Maybe
East Anglia 88 0 12
East Midlands 92 2 6
London 86 3 11
North East 80 10 10
North West 94 2 4
South East 90 0 10
South West 90 5 5
West Midlands 98 0 2
Yorkshire and Humberside 87 5 8
Wales 78 11 11

Table C.2- Acceptability of SIP construction by region

Percentage response to the question

“Would you buy a house built from this
material?”
Region Yes No Maybe
East Anglia 61 9 30
East Midlands 57 8 35
London 54 14 32
North East 44 11 45
North West 43 16 41
South East 53 11 36
South West 69 6 25
West Midlands 63 7 30
Yorkshire and Humberside 68 11 21
Wales 44 11 45

211
Table C.3- Acceptability of ICF construction by region

Percentage response to the question

“Would you buy a house built from this
material?”
Region Yes No Maybe
East Anglia 46 12 42
East Midlands 63 9 28
London 60 9 31
North East 45 22 33
North West 52 6 42
South East 48 7 45
South West 65 12 23
West Midlands 72 7 21
Yorkshire and Humberside 58 11 31
Wales 44 22 34

Table C.4- Acceptability of Prefabricated straw bale construction by region

Percentage response to the question

“Would you buy a house built from this
material?”
Region Yes No Maybe
East Anglia 27 46 27
East Midlands 29 41 30
London 15 41 44
North East 22 45 33
North West 27 51 22
South East 27 42 31
South West 40 28 32
West Midlands 26 35 39
Yorkshire and Humberside 30 38 32
Wales 11 56 33

212
Table C.5- Acceptability of thin joint block work construction by region

Percentage response to the question

“Would you buy a house built from this
material?”
Region Yes No Maybe
East Anglia 81 3 16
East Midlands 74 2 24
London 68 6 26
North East 78 22 0
North West 80 2 18
South East 72 2 26
South West 76 5 19
West Midlands 91 0 9
Yorkshire and Humberside 70 8 22
Wales 67 11 22

Table C.6- Acceptability of timber framed construction by region

Percentage response to the question

“Would you buy a house built from this
material?”
Region Yes No Maybe
East Anglia 47 9 48
East Midlands 59 13 28
London 53 6 41
North East 78 11 11
North West 43 25 32
South East 48 12 40
South West 64 10 26
West Midlands 54 14 30
Yorkshire and Humberside 59 14 27
Wales 56 22 22

213
Numerical results to the question “Would your company use this method in
construction?” from the construction industry survey can be seen in Table D.7

Table C.7- Percentage responses to the question “Would your company use this
material in construction?”

Brick and Prefabricated Thin joint Timber

Response block SIPs ICF straw bale block work frame
Yes 89 48 33 15 67 63
No 7 30 37 63 15 19
Maybe 4 22 30 22 18 18

214
Appendix D- Degree day data

215
Degree day data

The city selected for each region and the number of heating degree days used in
calculations is shown in Table D.1.

Table D.1- Number of heating degree days for city in each region of England and
Wales

Number of heating degree

Region City days, base temp 18°C
East Anglia Peterborough 3192.70
East Midlands Leicester 3220.92
London London 2824.94
North East Sunderland 3290.48
North West Liverpool 2987.29
South East Brighton 2930.49
South West Bristol 2968.24
West Midlands Birmingham 3339.69
Yorkshire and Humberside Leeds 3347.53
Wales Cardiff 3049.68

Using a base temperature of 25 based on information produced by the WHO (1979)

the number of cooling days required for the same locations is shown in Table D.2.

Table D.2- Number of cooling degree days for city in each region of England and
Wales

Number of cooling degree

Region City days, base temp 25°C
East Anglia Peterborough 3
East Midlands Leicester 1
London London 2
North East Sunderland 0
North West Liverpool 1
South East Brighton 0
South West Bristol 0
West Midlands Birmingham 1
Yorkshire and Humberside Leeds 0
Wales Cardiff 0

This low number of cooling days indicates it would not be viable to install cooling
systems as there is such a small cooling requirement. In addition England and Wales
do not have a history of installing cooling systems into domestic housing. Therefore
energy requirements associated with cooling were not included in this work.

216
Appendix E- The typical house

217
The typical house

In order to determine benchmark values for the embodied energy and operational
energy relating to wall materials it was necessary to generate the details of a typical
house for the area of study.

The most common method of construction for England and Wales is brick and block.
This was used for 88% of dwellings in England between 1990 and March 2009 (DCLG,
2010). Therefore, the typical house was considered to be built from a double skinned
brick and block system.

The average area of the typical house was based on a weighted average of mean floor
area for number of bedrooms combined with the percentage of houses built with each
bedroom number. This calculation can be seen in Table E.1.

Table E.1- Weighted average calculation for floor area.

Percentage of
Type of Mean completions Area x
house area (m2) 2009-2010 percentage
1 bed 64.30 1 64.30
2 bed 71.20 10 712.00
3 bed 95.60 25 2390.00
4 bed 120.60 19 2291.40
55 5457.70

Dividing the summed values of area multiplied by percentage, by the total percentage
of completions represented by the table gives the weighted average floor area. The
floor area used for the typical house was 99.23m2.

The typical house discussed by Monahan and Powell (2011) shows an aspect ratio in
the region of 1:2. The value of dimension a was therefore set at 5m, giving dimension b
a value of 9.923m; an aspect ratio of 1:1.98. This also accommodates
recommendations by Chown (1999) which indicate that 5.0m is the limit for narrow
fronted housing, with houses below this being difficult to successfully design internally.

Bungalows are considered separately in the data used; therefore two storeys were
selected as the typical number. This is stated to be the typical number by Chown
(1999) and reflects that case study used by Monahan and Powell (2011).

The height of each storey is based on data provided by Chown (1999). Recommended
ceiling height is 2.4m; therefore 3.0m was used as storey height to allow sufficient
space for the ceiling, floor joints and flooring.

218
Appendix F- Fitness function

219
function [singleobjective]=fn_fitness5b(v)

material = v(1);
a = v(2);

% Evaluates singleobjective, reduces number of variables for

optimisation
% process

%specify values that will not change during optimisation

areatotal = 95.6;
nostoreys = 2;
hgtstoreys = 3;
nodoors = 2;
region = 2;
buildingtype = 3;
maxwindowpercentage = 25;
doorsize = 2.42;
% material = 1
% a = 6500

%Calculate embodied energy, operational energy, acceptability,

airtightness

%evaluate dimensions
%a = a;
am = a/1000;
disp(a)
b = (areatotal/nostoreys)/am;
%disp(b)
aspectratio = b/am;
%disp(aspectratio)

%evaluate total external wall area

switch buildingtype
case 1 % Detached
totextwall = 2*(am*hgtstoreys*nostoreys) +
2*(b*hgtstoreys*nostoreys);
case 2 % Semi-detached
totextwall = 2*(am*hgtstoreys*nostoreys) +
(b*hgtstoreys*nostoreys);
case 3 % Terraced
totextwall = 2*(am*hgtstoreys*nostoreys);
%disp(totextwall)
end

%evaluate window area

twentypercentfloor = areatotal*.2;
maxwindow = totextwall*(maxwindowpercentage/100)

%evaluate door area

doorarea = doorsize * nodoors;

%evaluate net external wall area

220
 % if twentypercentfloor > maxwindow

% netextwallarea = 10000000000;

%windowarea = maxwindow;
%else %windowarea =< twentypercentfloor;

netextwallarea = totextwall - twentypercentfloor - doorarea;

%end

%evaluate party wall area

switch buildingtype
case 1 % Detached
partywallarea = 0;
case 2 % Semi-detached
partywallarea = (b*hgtstoreys*nostoreys);
case 3 % Terraced
partywallarea = 2*(b*hgtstoreys*nostoreys);
%disp(partywallarea)
end

%evaluate embodied energy

switch material
case 1 % Brick and block
embenergy = (netextwallarea * 744.3) + (0.5 *
partywallarea * 713.46);
case 2 % SIPs
embenergy = (netextwallarea * 1464.12) + (0.5 *
partywallarea * 834.77);
case 3 % ICF
embenergy = (netextwallarea * 1448.31) + (0.5 *
partywallarea * 1060.47);
case 4 % Lime plastered straw bale
embenergy = (netextwallarea * 458.61) + (0.5 *
partywallarea * 458.61);
case 5 % Thin joint blockwork
embenergy = (netextwallarea * 847.56) + (0.5 *
partywallarea * 739.92);
case 6 % Timber frame
embenergy = (netextwallarea * 946.45) + (0.5 *
partywallarea * 747.12);

%disp(embenergy)
end

embenergym2 = embenergy/areatotal;

%evaluate operational energy

switch material
case 1 % Brick and block
uvalue = 0.277;
case 2 % SIPs
uvalue = 0.167;
case 3 % ICF
uvalue = 0.204;
case 4 % Lime plastered straw bale
uvalue = 0.190;

221
 case 5 % Thin joint blockwork
uvalue = 0.272;
case 6 % Timber frame
uvalue = 0.249;
%disp(uvalue)
end

switch region
case 1 % East Anglia
hdd = 3192.7;
case 2 % East Midlands
hdd = 3220.92;
case 3 % London
hdd = 2824.94;
case 4 % North East
hdd = 3290.48;
case 5 % North West
hdd = 2987.29;
case 6 % South East
hdd = 2930.49;
case 7 % South West
hdd = 2968.24;
case 8 % West Midlands
hdd = 3339.69;
case 9 % Yorkshire and Humberside
hdd = 3347.53;
case 10 % Wales
hdd = 3049.68;
%disp(hdd)
end

openergy = (uvalue * netextwallarea * 24 * hdd )/ 1000;

%disp(openergy)

openergym2 = openergy/areatotal;

%evaluate acceptability
%public
switch region
case 1 % East Anglia
acceptBB = 94;
acceptSIP = 76;
acceptICF = 67;
acceptSB = 40.5;
acceptTJB = 89;
acceptTF = 71;
case 2 % East Midlands
acceptBB = 95;
acceptSIP = 74.5;
acceptICF = 77;
acceptSB = 44;
acceptTJB = 86;
acceptTF = 73;
case 3 % London
acceptBB = 91.5;
acceptSIP = 70;
acceptICF = 75.5;
acceptSB = 37;
acceptTJB = 81;
acceptTF = 73.5;
case 4 % North East

222
 acceptBB = 85;
acceptSIP = 66.5;
acceptICF = 61.5;
acceptSB = 38.5;
acceptTJB = 78;
acceptTF = 83.5;
case 5 % North West
acceptBB = 96;
acceptSIP = 63.5;
acceptICF = 73;
acceptSB = 38;
acceptTJB = 89;
acceptTF = 59;
case 6 % South East
acceptBB = 95;
acceptSIP = 71;
acceptICF = 70.5;
acceptSB = 42.5;
acceptTJB = 85;
acceptTF = 68;
case 7 % South West
acceptBB = 92.5;
acceptSIP = 81.5;
acceptICF = 76.5;
acceptSB = 56;
acceptTJB = 85.5;
acceptTF = 77;
case 8 % West Midlands
acceptBB = 99;
acceptSIP = 78;
acceptICF = 82.5;
acceptSB = 45.5;
acceptTJB = 95.5;
acceptTF = 69;
case 9 % Yorkshire and Humberside
acceptBB = 91;
acceptSIP = 78.5;
acceptICF = 73.5;
acceptSB = 46;
acceptTJB = 81;
acceptTF = 72.5;
case 10 % Wales
acceptBB = 83.5;
acceptSIP = 66.5;
acceptICF = 61;
acceptSB = 27.5;
acceptTJB = 78;
acceptTF = 67;
end

switch material
case 1 % Brick and block
acceptpublic = acceptBB;
case 2 % SIPs
acceptpublic = acceptSIP;
case 3 % ICF
acceptpublic = acceptICF;
case 4 % Lime plastered straw bale
acceptpublic = acceptSB;
case 5 % Thin joint blockwork
acceptpublic = acceptTJB;

223
 case 6 % Timber frame
acceptpublic = acceptTF;
%disp(acceptabilitypublic)
end

%industry
switch material
case 1 % Brick and block
acceptindustry = 91;
case 2 % SIPs
acceptindustry = 59;
case 3 % ICF
acceptindustry = 48;
case 4 % Lime plastered straw bale
acceptindustry = 26;
case 5 % Thin joint blockwork
acceptindustry = 76;
case 6 % Timber frame
acceptindustry = 72;
%disp(acceptindustry)
end

%evaluate airtightness

switch material
case 1 % Brick and block
airtightness = 4.5;
case 2 % SIPs
airtightness = 1.0;
case 3 % ICF
airtightness = 1.0;
case 4 % Lime plastered straw bale
airtightness = 0.86;
case 5 % Thin joint blockwork
airtightness = 4.0;
case 6 % Timber frame
airtightness = 5.00;
%disp(airtightness)
end

%evaluate wall thickness

switch material
case 1 % Brick and block
wallthickness = .346;
case 2 % SIPs
wallthickness = .326;
case 3 % ICF
wallthickness = .449;
case 4 % Lime plastered straw bale
wallthickness = .480;
case 5 % Thin joint blockwork
wallthickness = .346;
case 6 % Timber frame
wallthickness = .295;
%disp(wall thickness)
end

%evaluate fire performance

224
 switch material
case 1 % Brick and block
fireperformance = 240;
case 2 % SIPs
fireperformance = 73;
case 3 % ICF
fireperformance = 90;
case 4 % Lime plastered straw bale
fireperformance = 135;
case 5 % Thin joint blockwork
fireperformance = 180;
case 6 % Timber frame
fireperformance = 60;
%disp(fireperformance)
end

%specify weighting for elements

WEE = 1; % weighting for embodied energy

WOE = 1;% weighting for operational energy
WAP = 1; % weighting for public acceptability
WAI = 1; % weighting for industry acceptability
WAir = 1; % weighting for air tightness
Wthickness = 1; % weighting for wall thickness
Wfire = 1; % weighting for fire performance

%specify benchmark values for normalisation. Average values used,

%alternatively use British Regulation values.
%Switch functions used to apply correct value for building type

switch buildingtype
case 1 % Detached
benchmarkee = 11580;
case 2 % Semi-detached
benchmarkee = 7328.7;
case 3 % Terraced
benchmarkee = 3076.9;
%disp(benchmarkee)
end

switch buildingtype
case 1 % Detached
benchmarkoe = 33.3156;
case 2 % Semi-detached
benchmarkoe = 20.468;
case 3 % Terraced
benchmarkoe = 7.6203;
%disp(benchmarkee)
end
benchmarkair = 10;
benchmarkwt = .346;
benchmarkf = 60;

%evaluate single objective value. Aim for this value to be as low as

%possible indicating best option

%Apply constraint

225
 if twentypercentfloor > maxwindow

singleobjective = 10;

else %windowarea =< twentypercentfloor;

singleobjective = (WEE*(embenergym2/benchmarkee)) +
(WOE*(openergym2/benchmarkoe))...
- (WAP*(acceptpublic/100)) - (WAI*(acceptindustry/100)) +
(WAir*(airtightness/benchmarkair)) + ...
(Wthickness*(wallthickness/benchmarkwt)) -
(Wfire*(fireperformance/benchmarkf));
end

end

226