The current issue and full text archive of this journal is available on Emerald Insight at:

www.emeraldinsight.com/0969-9988.htm

IPD
Integrated project delivery implementation
implementation framework for framework

water and wastewater treatment

plant projects 609

Elie Georges Andary Received 2 February 2019

Revised 27 July 2019
Hazen and Sawyer, New York, New York, USA Accepted 13 August 2019

Caesar Abi Shdid

Department of Civil Engineering, Lebanese American University, Beirut, Lebanon
Arindam Chowdhury
Department of Civil and Environmental Engineering,
College of Engineering and Computing, Florida International University,
Miami, Florida, USA, and
Irtishad Ahmad
Department of Civil Engineering,
American University of Sharjah, Sharjah, United Arab Emirates

Abstract
Purpose – The public nature of water and wastewater construction capital projects has rendered design-bid-
build (DBB) as the delivery method of choice for such projects over the past years. Shortcomings inherent to
DBB have had a negative effect on the key performance indicators (KPIs) of these projects. Numerous studies
have been published about the benefits offered by integrated project delivery (IPD) in improving the delivery
of DBB projects. Links correlating IPD principles to improvements in KPIs of DBB construction projects have
not been established scientifically, thus leaving owners with no guidance on which IPD principle(s) to
implement in order to improve a particular project KPI. The paper aims to discuss this issue.
Design/methodology/approach – Actual data were collected from 43 water and wastewater construction
projects – including two control projects with full implementation of selective IPD principles – and used to
compute major projects KPIs. Regression analysis and a focus group are then utilized to determine the effect
of each implemented IPD principle on various project KPIs.
Findings – Implementation of open communication was found to have a significant effect on reducing project
cost overrun, and the co-location of teams significantly reduced the time to respond to RFIs. Collaborative
decision making significantly reduced the cost of field rework. Other IPD principles showed less-significant
effects on project KPIs and were concluded to be ineffective at improving the projects’ performance.
Practical implications – An implementation framework is developed that can guide utility owners on
which IPD principle(s) to implement in order to improve specific project KPIs.
Originality/value – This study demonstrates that not all IPD principles will result in performance
improvement of a project. A selective and cost-effective implementation of certain IPD principle will have to
be based on the nature of the project and the particular KPIs targeted for improvement.
Keywords Construction, Project management, Novel model
Paper type Research paper

1. Introduction
Water and wastewater treatment facilities remove biological and chemical waste products from
water and wastewater, provide clean water and allow the wastewater to be used for other
Engineering, Construction and
purposes. The complex nature of processes in water and wastewater treatment facilities require Architectural Management
highly specialized expertise in planning, cost estimating, design and construction of such Vol. 27 No. 3, 2020
pp. 609-633
facilities. This type of construction includes rehabilitation, expansion and new construction of © Emerald Publishing Limited
0969-9988
these treatment facilities, and requires a highly skilled team of individuals to ensure DOI 10.1108/ECAM-02-2019-0075
ECAM a successful project undertaken by large construction firms (Abi Shdid and Andary, 2016).
27,3 Water and wastewater construction projects have historically been delivered using the
traditional design-bid-build (DBB) delivery method and have therefore suffered from
shortcomings inherent to DBB (Abi Shdid and Andary, 2016). This is partly due to the public
nature of these projects and the existing public contract laws that make implementation of
integrated project delivery (IPD) in public projects rather challenging. IPD is an emerging
610 construction project delivery system that is becoming increasingly popular and is the subject of
great interest in the architecture/engineering/construction industry. It aims at integrating
project parties, systems, and business structures and practices into one multiparty agreement
that fosters highly effective collaboration commencing at early design stages and proceeding
through to project completion. IPD promises improved delivery of projects through increased
productivity and decreased waste of time and money, while aligning the interests and practices
of all involved parties. Major organizations, construction magazines and journal articles have
described projects that successfully implemented IPD with positive outcomes, such as lower
construction costs, shorter schedule and fewer RFIs. However, the drawn relationship between
IPD and project key performance indicators (KPIs) is shown to be of a general nature and
tenuous at best (Ernstrom et al., 2006; American Institute of Architects California Council, 2007;
Post, 2011; Kent and Becerik-Gerber, 2010). A close look at these studies reveals no existing
guideline to project participants that correlates or links individual IPD principles to particular
project KPIs that may be of interest to the owner. Collection and analysis of actual project data
is necessary to scientifically or statistically validate such widely-accepted claims.
The aim of this study is to use real data correlations between the implementation of
various IPD principles and KPIs of water and wastewater construction projects to develop a
guideline metric for owners. The metric can be used to identify the most effective IPD
principle to implement in order to improve certain desired project KPIs.

2. Literature review
2.1 Delivery methods in water and wastewater projects
Water and wastewater construction projects are public projects that are managed,
procured and funded differently than private projects, and thus have their own local
government procurement regulations (Shane et al., 2013; Molenaar et al., 2004). These
regulations have historically precluded public owners from entering into multiparty
agreements and playing a part in IPD-related financial incentive techniques such
as risk sharing, profit pooling or contingency pooling (NASFA et al., 2010; Singleton and
Hamzeh, 2011). However, several successful water infrastructure projects have
incorporated the concepts of collaboration and accountability through new
technologies, and industry stakeholders are showing a tendency toward creative, non-
traditional deliveries (Molenaar et al., 2004). In studying the delivery methods related to
water and wastewater projects, the Water Design-Build Council (2013) tries to promote
non-traditional delivery methods to facilitate relationships between owners and service
providers in a way to improve the nation’s municipal water and wastewater systems
development. Hence, contract provisions and project procedures can be modified and IPD
principles and practices applied as an overall philosophy, as a way to obtain additional
IPD-related benefits. For example, IPD has been documented to be successfully
implemented in projects that use traditional delivery approaches, and in which owners are
not able to use multiparty contracts (Thomsen et al., 2009; Darrington, 2011; Singleton and
Hamzeh, 2011). In a case study of the Orlando Utilities Commission — North Chiller Plant,
the owner integrated a number of IPD principles to foster a collaborative environment.
This resulted in the project being delivered 10 percent below the guaranteed maximum
price (Singleton and Hamzeh, 2011).
2.2 IPD principles IPD
There is no agreement in the literature on a single set of project delivery practices that implementation
comprise the IPD principles (Thomsen et al., 2009; NASFA et al., 2010). While the American framework
Institute of Architects (AIA) published IPD principles are the most commonly accepted and
used in the industry, several researchers have identified other principles that distinguish
IPD from other delivery methods (AIA, 2010; Singleton and Hamzeh, 2011; Kent and
Becerik-Gerber, 2010). 611
The main attributes that characterize IPD have been identified to include integrated
teams, lean construction principles, collective risk sharing, pain and gain sharing, and profit
pooling (Singleton and Hamzeh, 2011). This spirit of cooperation is credited with
overcoming the traditional adversarial attitudes commonly found in alternative delivery
methods, and is claimed to promise fewer change orders (CO) and claims, shorter schedules,
better cost control and more efficient project documentation (Kent and Becerik-Gerber, 2010;
Baiden and Price, 2011).
Many studies have identified benefits of implementing IPD in construction projects. For
example, Kent and Becerik-Gerber (2010) attributed fewer CO, shorter schedule and fewer
request for information (RFI) to the implementation of IPD principles. Others credited IPD
with minimizing waste, adding value, improving reliability and fostering collaboration
(Thomsen et al., 2009). Mossman et al. (2010) summarized potential benefits of IPD in the
perspective of three entities: clients added value and reduced energy costs of use; designers
reduced design time and design cost; and contractors reduced the cost of rework. Other case
study projects using IPD demonstrated that budget, schedule, design quality, and
sustainability met or surpassed the owner’s expectations, and the projects met the financial
expectations of the engineers, architects and contractors alike (AIA, 2010). More recently,
El Asmar and Hanna (2013) concluded that IPD results in improved project performance in
areas related to schedule, quality and CO.
A careful review of the literature shows that existing studies have been limited to
demonstrating the benefits of the overall implementation of IPD on a project’s KPIs.
However, the correlations between each IPD principle and a project’s performance metrics
have so far remained unexamined. The lack of such linkage leaves owners and contractors
with no guidance as to which IPD principle to implement given their desire to improve the
project success in a key performance area. This study aims to fill this gap by mapping
the correlation between the various IPD principles and KPIs of water and wastewater
construction projects using real data.
There is no agreement in the literature on a single set of project delivery practices that
comprise the whole of IPD principles. For example, some researchers considered building
information modeling (BIM) as an IPD principle while others consider BIM as a practice that
only facilitates the IPD process by recording and sharing project information and helping
the IPD team optimize both product and process (Thomsen et al., 2009; NASFA et al., 2010).
While the AIA published IPD principles as the most commonly accepted and used in the
industry, several researchers, as listed in Table III, have identified other principles that
distinguish IPD from other delivery methods.
Not all IPD principles are applicable to public projects and, more specifically, those using
DBB for project delivery (Azhar, 2014). Currently, the Federal Acquisition Regulation (FAR)
does not permit the government to participate in IPD-related financial incentive techniques
such as risk sharing, profit pooling or contingency pooling. The FAR also does not allow for
relational contracts or multiparty agreements and required competitive bidding for
construction contracts with limited exceptions (Singleton and Hamzeh, 2011). Other IPD
principles are not applicable to public projects but can, however, be modified to allow their
integration in public projects (Azhar et al., 2014). The IPD principles measured for the data
set projects and implemented in the control projects are described in the next section.
ECAM 2.3 Project performance scoring
27,3 Measuring project success is however not an easy task, but rather a complex process that
involves several success criteria and corresponding performance factors that need to be
accounted for. Success criteria are the dimensions of a project by which the project success
or failure will be judged, while performance factors are those measures which contribute to
achieving success in each criterion (Chan et al., 2004). A set of KPIs are normally used to
612 measure the performance factors for each success criteria. Time, cost and quality are
recognized in the literature as the main success criteria of a project (Chan and Chan, 2004).
In addition to these basic criteria, other criteria have also been identified such as
interpersonal relations between the project participants (Pinto and Pinto, 1991) and absence
of legal claims (Pocock et al., 1996). More recent studies identified KPIs of projects
that can be used to quantify project success such as schedule, profit, environment,
specifications, safety, effectiveness of process, level of dispute, staff satisfaction and
innovation (Alarcon and Ashley, 1996; Lam et al., 2007; Liu et al., 2014).
Existing studies have fell short of quantifying project success into a performance score
that is obtained from a mathematical model that incorporates all performance factors and
uses raw data for baseline comparison with past projects. Furthermore, water and
wastewater construction capital projects do not currently have an established performance
scoring model. This study bridges this gap by developing and using the first project
performance scoring method for water and wastewater construction projects using a single
index using raw project data.
In order to measure the project success in the five success criteria identified in existing
literature (cost, time, quality, customer satisfaction, and early involvement), nine
performance factors were used: cost overrun, CO cost, time overrun, RFI, errors and
omissions cost, claims, RFI response time, field rework and owner-requested changes.
The use of these nine performance factors is justified next.
A successfully managed and delivered project is one which gets completed on or under
budget. The main consideration when evaluating project cost is the measurement of the
contractor’s performance against the bid cost of the contract. CO typically alter the original
contact amount and cost overrun tracks costs incurred in excess of the contract amounts.
Cost overrun and CO costs are thus used as critical performance factors for project cost
(Abi Shdid et al., 2019).
Project schedule reflects the timely and efficient preparation, implementation and
closeout of tasks, meeting key program milestones and contractual delivery dates, recovery
from delays, and appropriateness of response to changes. On-time completion of a project is
an indicator of many successful aspects of the project delivery, such as efficiency and proper
project management. A project’s performance with respect to schedule can be measured
using time overrun as a critical performance factor (Abi Shdid et al., 2019).
A good quality design typically includes detailed approach in design concepts, analysis
and detailed execution procedures, thoroughness and accuracy. A large number of RFIs and
high errors and omissions cost indicate low quality of the design and construction
documents. Therefore, RFIs and error and omission costs are used as critical performance
factors for quality (Abi Shdid et al., 2019).
A successfully managed and delivered project is one which results in customer
satisfaction and repeat business. A project’s performance with respect to owner satisfaction
can be measured by the amount of claims resulting from that particular project. Claims are
thus used as a critical performance factor for customer satisfaction, since an owner is not
likely to hire the same contractor in the future if the current project resulted in claims
(Abi Shdid et al., 2019).
Early involvement of project participants – including the owner – promotes well-
developed relationship among project participants and are essential at all levels where
decisions can be made and results achieved. Frequent and honest communication improves IPD
project efficiency through reduced RFI response time, improved labor relations, reduced implementation
owner-requested CO and field rework through improved planning, organizing and framework
managing of all program elements. Early involvement helps to achieve and sustain a high
level of communication and collaboration, thus resulting in recognizing critical problem
areas and ensuring integrated operational efficiency. Error and omissions typically initiate
CO to pay for the correction needed due to poor design quality (Abi Shdid et al., 2019). 613
3. Research methodology
3.1 Research design
The research methodology used in this study combines both quantitative and qualitative
approaches, and consists of first determining the project performance factors and
corresponding KPIs used to mathematically quantify the projects’ performance factors. This
is followed by presenting the collection of raw data used to quantitatively compute the
projects’ performance scores along with the qualitative scoring mechanism used for the IPD
principles. Two control projects with full IPD implementation are then described and
evaluated. The IPD-KPI mapping is developed based on quantitative statistical regression
analysis and a qualitative focus group. Finally, a guideline metric based on results is
presented and discussed. This research design is graphically depicted in Figure 1.

3.2 Data collection

Project performance raw data were collected from water utility construction projects
delivered using the traditional method of DBB in the South Florida geographic region. All
water utilities in South Florida were contacted in order to identify the population of water
and wastewater projects completed since the year 2000. Older projects that were completed
before the year 2000 were not considered to be relevant anymore due to the recent changes
in technology, delivery systems and currency inflation. The population consisted of
60 individual projects of varying sizes. The performance data of these projects were
carefully examined and projects with insufficient data were excluded from the analysis. The
resulting raw data set comprised forty three projects of largely varying sizes when
categorized by their total base contract value, thus making up a highly representative and
unbiased sample of the population. Figure 2 shows the project size categories and the total
number and percentage of projects falling in each category.
In order to quantify project performance, a set of KPIs were used to measure each of the
nine performance factors using the collected raw data. The success criteria, the
corresponding performance factors and KPIs are summarized in Figure 3 and explained
in Table I. KPIs were normalized by the total project cost in order to eliminate any bias
resulting from the project size.
The values for the nine KPIs listed in Table I are obtained from the expressions shown in
the following equations:
Actual Project CostAward Price
Cost Overrun % ¼  100; (1)
Award Price
P
Cost of Change Orders
Change Order Cost % ¼  100; (2)
Actual Total Project Cost

Actual Project Duration

Time Overrun % ¼  100; (3)
Scheduled Project Duration
ECAM Identify project
27,3 performance
factors(PF)

Select IPD
principles and Identify KPIs for each
614 IPD scoring PF and formulate the
mechanism PF score

Collect raw data

Use 2 control projects Compute the KPIs
from 43 water
and implement all for the 43 projects
and wastewater
selected IPD principles of the dataset
projects

Compute IPD and Compute the IPD Compute the PF

PF scores for the scores for the 43 scores for the 43
control projects projects of the projects of the
dataset dataset

Perform
regression analysis

Establish links
Input from
between IPD
focus group
principles and PFs

Develop IPD
Figure 1. implementation
Research methodology framework

Total N umber of RFI s

RFI s per U nit Price ¼ ; (4)
Actual Total Project Cost

P
Cost of Errors and Omissions CO
Errors and Omissions Cost % ¼ P  100; (5)
Cost of Change Orders
 12 IPD
implementation
10
20.93% framework
18.60%
8
No. of projects

13.95%
6
11.63% 11.63% 11.63% 615
4
6.98%
4.65%
2

0
Figure 2.
$(0–1)

$(1–2)

$(2–6)

$(6–10)

$(10–14)

$(14–22)

$(22–40)

$(40–120)
Number and
percentage of projects
for specified range
of contract
Base contract amount range (millions)

P
Direct Cost of Claims
Total Claims Cost % ¼  100; (6)
Actual Total Project Cost

Average RFI Response Time

RFI Response Time per U nit Price ¼ ; (7)
Actual Total Project Cost

Total Direct Cost of Field Rework

Total Cost of Field Rework % ¼  100; (8)
Actual Total Project Cost

Total Cost of Owner 0 s CO

Owner 0 s CO Cost % ¼ P  100: (9)
Cost of Change Orders
The means (μ) and standard deviations (σ) of the data set were then computed for each of the
nine KPIs, as summarized in Table II.
Since some of the KPIs used in this study have dimensional units while others are non-
dimensional, values could not be simply added together, and standardization was necessary.
A standardization method was used that transforms any set of numbers to their equivalent
values in standard deviation units by subtracting the mean and dividing by the standard
deviation, as expressed in the following equation (Kreyszig, 1979):
xj mj
Zj ¼ where j ¼ 1; 2; 3; . . .; 9; (10)
sj
where Zj is the standardized value of a KPI, xj is the value of the KPI as calculated from
Equations (1)–(9), and μj and σj are, respectively, the mean and standard deviation of the KPI
as calculated from the data set of past projects ( from Table II).
Negative values of Zj indicate above average performance while positive values indicate
below average performance, and 0 indicates the average project performance with respect to
a specific performance factor. The measurement units of the standardized value Zj of a
particular KPI are therefore the number of standard deviations (N) above or below the mean
ECAM
27,3 Project
perform
ance Succes
s
criteria Perform
ance
factors
KPIs

616 Cost ov
errun
Project
cost Cost ov
errun %

Change
order
(CO) co Change
st order
Project cost %
schedu
le Time ov
errun
Time ov
errun %

Project Reques
t for
perform Quality informa RFIs pe
ance tion r unit
price
Errors a
nd
omissio Errors a
ns omissio nd
Custom ns cost
er %
satisfac
tion
Claims
Total cla
ims
cost %
RFI res
ponse
time RFI res
ponse
time pe
r unit
Early price
involvem
ent Field re
work
Total co
s
field rew t of
ork %
Owner
Figure 3. requeste
d Owner
Project performance change cha
s orders c nge
evaluation ost %

of the data set for that KPI (μj). The following equation was then developed and used to
calculate the score for each performance factor, designated by aj, using the standardized
value Zj of its corresponding KPI:
 
Z j þN
Perf ormance Factor Score ¼ aj ¼ 10  where –N pZ j pN: (11)
2N

Equation (11) is designed to yield a score that is based on a variable scale between 0 and 10;
0 being the lowest performance score and 10 being the highest. N is a user-specified value
which represents the number of standard deviations from the mean that designate the
boundaries of the performance measurement scale. For example, and as shown in Figure 3,
an owner choosing a value of N ¼ 2 is basically choosing to have the performance score of a
particular performance factor to be zero when Zj value is above 2σj. Similarly, by choosing a
value of N ¼ 2, a project will have the performance score of a particular performance factor
Success Performance
IPD
criteria factors KPIs implementation
framework
Cost Cost overrun Cost overrun as a percentage of the total project award price
CO cost Change order cost as a percentage of total project cost
Time Time overrun Time overrun as a percentage of the original project schedule
Quality RFIs Number of RFIs per unit price
Errors and Errors and omissions cost as a percentage of total cost of change orders 617
omissions cost
Customer Claims Total claims cost as a percentage of the total project cost. The direct cost
satisfaction of claims is the amount paid by the owner, contractor or engineer as a
resolution of the dispute as ruled by mediation, litigation or arbitration
Early RFI response RFI response time per project cost Table I.
involvement time Success criteria and
Field rework Cost of field rework as a percentage of total project cost corresponding
Owner-requested Owner-requested change orders cost as a percentage of total cost of performance factors
changes change orders and KPIs

KPI Mean (μj) SD (σj) Range

Cost overrun % 7.38 6.59 0.00–38.84

Change order cost % 5.83 5.40 −10.77 to 27.98
Time overrun % 23.25 34.51 −7.62 to 125.56
RFIs per unit pricea 16.40 33.88 2.03–166.11
Errors and omissions cost % 27.68 25.23 0.00–100.00
Total claims cost % 4.85 14.21 0.00–69.30
RFI response time per unit pricea 6.13 14.85 0.00–66.47 Table II.
Total cost of field rework % 0.93 0.058 0.72–1.11 KPI mean and
Owner-requested changes % 55.08 35.29 0.00–116.09 standard deviation of
Note: aPer million $ award price the data set

to be 10 when Zj value is below −2σj. It is important to point out that due to the nature of the
KPIs, above average performance score values of Zj is an indication of unsatisfactory
performance of the project. For example, an above average performance for cost and
schedule overruns should result in a low performance score. Hence, the performance factor
score values are obtained by multiplying Zj in Equation (11) by −1. Figure 4 summarizes the
use of Equation (11) using as an example a chosen value of N ¼ 2. A higher N value
increases the score of a performance factor for a positive value of Z and decreases it for a
negative value of Z. Choosing a value for N is thus subjective and should be determined by
the owner depending on the project.
The IPD principles used in this study are qualitative metrics that rely on the evaluator’s
opinions and impressions of performance quality (Table III). A commonly-used scoring
system that was developed by Azhar (2014) was used to assess the implementation level of
each IPD principle in the projects of the data set. This scoring system is based on a numerical
scale that varies from 0 to 10, where a 0 means that the IPD principle was not implemented in
a project and a 10 means it was fully implemented. For example, the rating scale rubric
presented in Table IV is developed and used for evaluating mutual respect and trust.
The IPD scores of the data set projects were then determined through meetings with
utility owners and their consultants, and with the help of in-house documents and databases
that were available to the authors through a secure intranet system. The degree of IPD
implementation was determined for each project based on the subjective score scale, with
ECAM 10
10
27,3 9
8

Performance score aj
7 N=2
6
618 5
10×((–Zj +N)/2N )
4
3
2
1
Figure 4.
Comparison of 0
0
performance –3 –2 –1 0 1 2 3
score to Z
Z-score Zj

Author(s) IPD principles

AIA (2007) Mutual respect and trust, mutual benefit and reward, collaborative innovation and
decision making, early involvement of key participants, early goal definition,
intensified planning, open communication, appropriate technology, organization
and leadership
AIA (2010) Early involvement of key participants, shared risk and reward, multiparty contract,
collaborative decision making and control, liability waivers among key
participants, jointly developed and validated project goals
Forbes and Syed (2011) Multiparty contract, close team collaboration for optimizing the project
Kent and Becerik- Multiparty agreement, early involvement of all parties, shared risk and reward
Gerber (2010)
Kim and Dossick (2011) Integrated form of agreement (IFOA), lean construction and using BIM
Matthews and Howell Multiparty contract, shared risk and profit
(2005)
NASFA et al. (2010) Multiparty agreement, trust and mutual respect, mutual benefit and reward,
collaborative decision making, early involvement of key project participants, early
goal definition and intensified planning, open communication within the project
team and ability to address issues, liability waivers between key participants,
jointly developed project target criteria, key participants bound together as equals,
shared financial risk and reward, transparent financials between key participants
Singleton and Hamzeh Integrated teams, lean construction techniques, lean principles, collective risk
Table III. (2011) sharing, pain sharing and gain sharing, profit pooling, contingency sharing, goals
IPD principles and incentives and performance evaluations

the scores assigned according to the best suited point on the scale. Using the collected raw
data, the IPD principles scores were determined for each of the 43 projects. The mean IPD
scores of the data set were then computed for each of the eight IPD principles, as
summarized in Table V.
Two wastewater construction control projects were delivered with full implementation of
selective IPD principles within a DBB project delivery framework. Those projects were Final
Site Work (Contract S-810) and Screening System Improvements (Contract S-863) at the
South District Wastewater Treatment Plant Capital Project Expansion, both at Miami Dade
Water and Sewer Department. Both projects were delivered by infusing selective IPD
principles within the traditional DBB delivery method. The projects organization consisted
Score for mutual respect and trust
IPD
10 7.5 5 2.5 0 implementation
framework
Trust-building Trust-building Trust-building Trust-building Trust-building
workshops were workshops were workshops were not workshops were not workshops were not
conducted during conducted during conducted during conducted during conducted during
the project phases, the project phases, the project phases, the project phases, the project phases
team has high prior team has high prior team has medium team has medium team has no 619
working experience working experience prior working prior working working experience
and trust and trust experience and trust experience and trust and trust Table IV.
competence was competence was not competence was competence was not competence was not Scoring mechanism
considered as considered as considered as considered as considered as for mutual respect
selection criteria selection criteria selection criteria selection criteria selection criteria and trust

IPD principles Mean score

Open communication within project team 3.69

Integrated and collaborative teams 0.34
Lean principles 1.48
Co-location of teams 4.49
Project’s staff performance evaluations 1.92
Mutual respect and trust 1.82 Table V.
Jointly developed project target criteria 0.51 IPD principles score
Collaborative decision making 1.3 means of the data set

of owners, engineers, construction managers, contractors and accountants. The first author
of this study served as the construction manager on these projects, giving the research team
access to critical data related to the projects parameters, while at the same time having the
ability to coordinate, implement, and manage IPD principles into the delivery of the projects.
While the authors recognize that certain bias may arise as a result of this role, the focus
group of owners, engineers and contractors that was utilized to review the data
interpretations and verify the identified relations between KPIs and IPD principles
minimized any such bias in interpreting the data in a way that supports the hypothesis or
expectations and would influence the research and conclusions.
The selected IPD principles that were implemented in the two control projects by the
project team are described next.
Open communication within the project team. Essential team meetings and collaboration
among project participants help in opening venues for more communication. Weekly
progress meetings were held with the contractor, consultant, owner and other stakeholders
who had the ability to address issues. Meetings allowed open discussions and updates
related to the progress of each element of current work, schedule revisions, milestone dates
and updated contract cost and time. A brief summary of the progress of the work was
documented in meeting minutes and distributed to all projects participants.
In addition to weekly meetings, informal daily meetings among the projects’ team
members were organized by the first author of this study, regardless of the presence of any
issues. These meetings were intended to help stay updated with the progress and to foresee
any upcoming conflict. Open communication through phone conversations between the
engineers and the general contractors were conducted on a daily basis. This facilitated
faster notification and efficient processing of documents requested by the Miami Dade
Building Department officials. This also allowed an expeditious shop drawing review and
ECAM approval process by conveying messages right on the spot. Phone calls between different
27,3 parties resolved problems and addressed issues expeditiously.
Integrated and collaborative teams. The projects participants were led by the construction
manager to collaboratively work as team members. This was accomplished by emphasizing
team engagement for maintaining integrity among project participants. Team engagement
provides the team members with a greater vision and makes them feel like their team cares
620 about them, values their work and has their best interests in mind. The past experience on
previous construction projects performed by the contractor’s project manager, consultant’s
construction manager, and the owner’s representatives assisted in promoting this
collaborative team environment on both control projects. The construction manager also
prompted integration and collaboration by introducing social activities outside the
workplace in an effort to bring people together and promote friendship.
Lean principles. Before the construction manager joined the control projects, lean
principles were integrated in the design process and were focused on maximizing value and
eliminating waste. During the execution phase, the construction manager focused on
increasing the efficiency of lean principles through monthly reviews of the construction
schedule updates that were required and approved by the engineer. The construction
manager also required detailed two-week “look ahead” schedules to be prepared and
thoroughly reviewed during the weekly progress meetings. The process of eliminating
waste, meeting or exceeding all projects requirements, focusing on the entire value stream,
and pursuing perfection in the execution of the projects was continuous and was monitored
closely by the construction manager throughout the projects execution with corrective
actions taken immediately.
Co-location of team. Co-location helps to increase the level of collaboration and
innovation, and to promote meeting project goals and commitments. Co-location of teams
was implemented on the two control projects, after the projects were awarded to the
contractors. The owner provided for approximately one acre of contiguous physical space
that included pre-engineered metal building for the owner, engineers, and construction
manager (research team), field office parking and a field office trailer city for contractors and
subcontractors. In addition to the engineers and contractors, the onsite teams included
schedulers, accountants, inspectors, state inspectors general, document control staff, safety
officers and auditors.
Project’s staff performance evaluations. Evaluations are useful in determining lapses in
performance and in providing the information needed for improvement. On the control
projects, the authors developed performance evaluation forms for evaluating the projects’
participants. The owner filled quarterly performance evaluations for both the contractors
and the consultants. In addition, the consultants evaluated their personnel including
engineers and construction managers.
Mutual respect and trust. A successful IPD implementation requires developing a high
level of trust among the project participants. In the control projects, some preexisting trust
was present among the owner, consultant and contractors due to having repetitive work and
good long-lasting work relationships from previous projects. This was further reinforced
through planned communication and group assignments where team members were forced
to meet face-to-face rather than communicating via e-mail in order to develop confidence as
they become aware of each other’s capabilities. Also, group assignments to develop certain
project-related tasks required the teams to constantly interact, communicate and give each
other feedback to help develop trust. The success of this strategy was evident on many
occasions where the contractors proceeded with performing needed extra work prior to
receiving executed CO to ensure compensation for the extra work; which is an indication of
trust and respect.
 Jointly developed project target criteria. Carefully defining project performance criteria IPD
early in the design phase with the input, support and acceptance of project participants implementation
ensures that maximum attention be paid to the project. For the control projects, jointly
developed project criteria were planned and agreed upon with the project participants
framework
during the design and preconstruction phases. During the execution phase, the construction
manager held monthly meetings with the contractors, consultants, owner and other
stakeholders to monitor and update the jointly developed project target criteria. For 621
example, the control projects were among other projects that were required to be in
compliance with a Consent Order with the Florida Department of Environmental Protection.
One of the main project target criteria was to meet the milestone dates and satisfy all other
consent order requirements. Structured jointly developed project criteria meetings were key
in meeting those requirements.
Collaborative decision making. On the control projects, the construction manager formed
leadership teams including the contractors, consultants, owner and other stakeholders for
decision-making purposes. The teams held monthly meetings and provided recommendations
on decision-making priorities and activities and communication tools toward enhancing
system efficiencies for the project. The teams also assigned specific tasking to develop options
for potential opportunities that could be beneficial for the projects.
Following the completion of the two control projects, their KPIs (xj) were computed,
followed by the standardized value, Zj, for each KPI, and the performance factors scores (aj)
based on a value of N ¼ 2, as summarized in Table VI. The implementation levels for the
various IPD principles were scored for the control projects as summarized in Table VII.

KPI value (xj) Zj PF score (aj)

Performance factor Control Project Control Project Control Control Control Control
(PF) 1 (%) 2 (%) Project 1 Project 2 Project 1 Project 2

Cost overrun 3.41 4.36 −0.60 −0.46 6.51 6.14

Change order cost 8.15 3.64 0.43 −0.40 3.93 6.01
Time overrun 0.00 −0.71 −0.67 −0.69 6.68 6.74
RFIs 3.21 4.93 −0.39 −0.34 5.97 5.85
Errors and
omissions cost 7.13 23.19 −0.81 −0.18 7.04 5.44
Claims 0.00 0.00 −0.34 −0.34 5.85 5.85
RFI response time 0.96 2.13 −0.35 −0.27 5.87 5.67 Table VI.
Field rework 0.00 0.00 −15.77 −15.77 10.00 10.00 Control project KPIs
Owner-requested and performance
changes % 53.21 39.39 −0.05 −0.44 5.13 6.11 factors scores

IPD score
IPD principles Control Project 1 Control Project 2

Open communication within project team 7.5 7.5

Integrated and collaborative teams 7.5 7.5
Lean principles 7.5 5.0
Co-location of team 7.5 5.0
Project’s staff performance evaluations 5.0 5.0
Mutual respect and trust 5.0 5.0 Table VII.
Jointly developed project target criteria 7.5 7.5 Control project IPD
Collaborative decision making 7.5 7.5 principles scores
ECAM 3.3 Data Analysis
27,3 Regression analysis was used on the data set of 43 projects and the two control projects to
identify possible correlation links that may exist between the implemented IPD principles
(explanatory variables) and the performance score of projects with respect to each KPI
(response variables). Such correlation, if present, can be valuable for owners wishing to
identify which IPD principles to implement for improving a specific desired KPI of their
622 project. Regression analysis provides the correlation of determination (R2) that represents
how accurately the regression model represents the existing data. R2 determines the
proportion of the variation in the response variable explained by changes in the
independent, categorical variable (Newbold, 1991).
The p-value is used to identify the level of significance of the variables; small p-values
indicate the statistically significant variables. For numerical variables, correlation
coefficients indicate the degree of association between variables. Correlation coefficients
can range from −1 to +1. Larger coefficient values, either positive or negative, indicate a
stronger linear association between variables. Coefficient values of zero indicate no
association between variables. Positive coefficients indicate that the independent variable
and response variable vary in the same direction (positive sloped line); negative coefficients
indicate the independent variable and response variable vary in opposite directions
(negative sloped line) (Ling et al., 2004).
IPD principles that correlate significantly to one or more KPI(s) were identified through
the Pearson correlation coefficient, R. The predictive power of the model is weighed through
the coefficient of determination, R2. The R2 value provides an explanation of the proportion
of variation in the dependent variable (performance score) due to a change in the
independent variable (IPD score). In general, models that have R2 larger than or equal to 0.7
are considered to have high predictive powers with statistically significant relationships
(Ling et al., 2004). Presented henceforth are the regression models with high predictive
powers (R2 ⩾ 0.7).

4. Results
4.1 Open communication
The data were analyzed to determine the Pearson correlation coefficients and coefficients of
determination between implementation of open communication within the project team and
the performance scores of various project KPIs. The results summarized in Table VIII
indicate that there is a statistically significant relationship that exists between open
communication and only one of the project KPIs, namely, cost overrun.
In Table VIII, the Y intercept of the regression is denoted by “Constant” and the
independent variable is denoted by “X Variable.” β is the regression coefficient and t-value
represents the t-statistic to be compared to the theoretical t-distribution for significance.
The p-value is the calculated probability obtained in the t-test results and is used to
identify the level of significance of correlation of the variables. Small p-values ( p ⩽ 0.05)
indicate that the null hypothesis β ¼ 0 can be rejected and that there is a statistically
significant correlation with the variables.
Cost overrun of water and wastewater projects shows to be strongly associated with the
response variable, with R for the numerical variable being 0.930. The R2 value indicates that
86.5 percent of the variation in the cost overrun is predictable from the implementation level
of open communication. Based on the results of one-way analysis of variance, the predictor
variable shows a high level of significance because the p-value is 0.000. It can therefore be
concluded that the independent variable, open communication, has a statistically significant
effect on the cost overrun of a project. The remaining low values of R2 in Table VIII indicate
that links do exist but with relatively weak effect.
KPI R R2 β t-value p-value
IPD
implementation
Strong effect framework
Cost overrun % 0.930 0.865
Constant 2.387 10.869 0.000
X variable 0.753 15.009 0.000
Weak effect 623
Change order cost % 0.723 0.523
Constant 2.815 7.329 0.000
X variable 0.6593 6.760 0.000
Total cost of field rework % 0.514 0.264
Constant 6.845 11.263 0.000
X variable −0.5436 −3.142 0.000
Error and omissions cost % 0.431 0.186
Constant 4.029 6.2462 0.000
X variable 0.298 2.092 0.042
Total claims cost % 0.392 0.154
Constant 6.101 13.832 0.000
X variable −0.222 −2.208 0.032
Owner-requested changes % 0.270 0.073
Constant 5.731 7.961 0.000
X variable −0.186 −1.132 0.264
Time overrun % 0.192 0.037
Constant 5.757 9.093 0.000
X variable −0.141 −0.975 0.335
RFIs per unit price 0.192 0.037
Constant 5.433 12.363 0.000
X variable −0.0538 −0.377 0.708 Table VIII.
RFI response time per unit 0.176 0.031 Correlation between
Constant 5.576 11.440 0.000 open communication
X variable −0.087 −0.783 0.438 and project KPIs

Figure 5 shows a plot of open communication IPD score and cost overrun performance score
(a2) for the 43 projects and the two control projects. It can be seen that as the implementation
level of open communication increases in a project, the performance score related to cost
overrun increases and, subsequently, the better the cost performance of the project is.

8
Control projects
7 Data set projects

6
Cost overrun score (a1)

2 y = –0.0004x 3 – 0.0657x 2+1.2205x +1.902

R 2 = 0.8652
1 Figure 5.
Open communication
0 vs cost overrun
0 1 2 3 4 5 6 7 8
performance score
Open communication
ECAM This strong effect of communication on cost overrun is expected, since regular status
27,3 meetings typically include reviewing project cost reports and detailing project cost
estimates to help overcome major cost obstacles.
The resulting regression equation shown in Figure 3 is a polynomial third degree, in
which x is the IPD score of open communication (independent variable) and y is the
performance score (aj) of the cost overrun KPI (dependent variable). This equation can be
624 used by water utilities to predict in a rather simple but yet reliable way the cost performance
of a project based on a presumed IPD implementation level of open communication.

4.2 Co-location of teams

The Pearson correlation coefficients and coefficients of determination between co-location of
teams and the performance scores of various KPIs presented in Table IX indicate that there
is a statistically significant relationship between locating project teams in close proximity
on the jobsite and only one of the project KPIs, namely, RFI response time per unit price. The
R value of 0.987 for the numerical variable indicates a strong association with the response
variable, and the R2 value indicates that 97.4 percent of the variation in the RFI response
time per unit price is predictable from the co-location of teams’ score. The extent to which a
project implements the IPD principle of co-location of teams seems to significantly affect the
prediction of the project’s average response time to RFIs, but a rather weak effect on the
prediction of other project KPIs.

KPI R R2 β t-value p-value

Strong effect
RFI response time per unit price 0.987 0.974
Constant 2.696 7.518 0.000
X variable 0.567 7.928 0.000
Weak effect
RFIs per unit price 0.740 0.548
Constant 3.600 8.680 0.000
X variable 0.375 4.540 0.000
Time overrun % 0.415 0.172
Constant 3.468 5.226 0.000
X variable 0.385 2.911 0.005
Change order cost % 0.355 0.126
Constant 6.033 9.891 0.000
X variable −0.217 −1.785 0.081
Total claims cost % 0.303 0.092
Constant 4.936 9.305 0.000
X variable 0.072 0.680 0.500
Cost overrun % 0.292 0.085
Constant 6.224 10.380 0.000
X variable −0.225 −1.880 0.067
Total cost of field rework % 0.272 0.074
Constant 3.980 5.349 0.000
X variable 0.270 1.822 0.075
Owner-requested changes % 0.265 0.070
Constant 4.824 5.768 0.000
X variable 0.045 0.271 0.788
Table IX. Error and omissions cost % 0.105 0.011
Correlation between Constant 5.582 7.478 0.000
CLT and project KPIs X variable −0.095 −0.637 0.527
It can be seen from Figure 6 that as the implementation level of co-location of teams IPD IPD
principle increases in a project, the performance score related to RFI response time per implementation
unit price increases. This correlation is expected because when key participants of owner, framework
engineer, and contractor are present onsite, they can respond to RFIs and provide
information more efficiently. Figure 6 also shows that when this specific IPD score is
higher than 5, further improvement in RFI response time per unit price is rather minimal.
A score of 5 is given to this IPD principle when the key participants of owner, engineer and 625
contractor are located onsite and are dedicated full time to the project. These key
participants have an interest in the project and can influence the execution plan of the
project. Higher scores of this particular IPD principle are given when additional
participants are located onsite such as accountants, consultants, schedulers, legal experts
and building officials. It was observed that an increase in the number of participants in
addition to the key participants present on site and in close proximity tends to add stress
to the project and influences the project’s complexity level, which limits the improvement
in this KPI. It was also observed that in such cases, significant time and resources of the
project are dedicated to identifying, understanding and managing the various
participants’ expectations.

4.3 Collaborative decision making

When examining the data related to collaborative decision making, only one project KPI
(cost of field rework) is deemed to be significantly affected by the implementation of this IPD
principle. The Pearson correlation coefficients and coefficients of determination between the
implementation of collaborative decision making and the performance scores of various
KPIs shown in Table X indicate that 78.8 percent of the variation in the KPI is predictable
from the IPD implementation score. The remaining project KPIs are not strongly affected by
the extent this IPD principle is implemented in the project delivery.
Figure 7 shows that as the implementation of collaborative decision making IPD
principle is increased in a project, the performance score of cost of field rework score
increases and, subsequently, the better the cost performance of the project is. The influence

6
RFI response time per unit price score

4
(a7)

Control projects
3
Data set projects

1 y = 0.0378x 3 – 0.6204x 2 + 3.2786x + 0.3153

R 2 = 0.9739
Figure 6.
CLT vs RFI response
0
time per unit price
0 1 2 3 4 5 6 7 8
performance score
Co-location of teams
ECAM KPI R R2 β t-value p-value
27,3
Strong effect
Total cost of field rework % 0.888 0.788
Constant 3.919 19.758 0.000
X variable 0.885 11.758 0.000

626 Weak effect

Change order cost % 0.549 0.301
Constant 5.506 17.829 0.000
X variable −0.312 −2.662 0.011
Cost overrun % 0.550 0.302
Constant 5.690 18.910 0.000
X variable −0.330 −2.894 0.006
Total claims cost % 0.293 0.086
Constant 5.010 18.389 0.000
X variable 0.173 1.678 0.100
Error and omissions cost % 0.268 0.072
Constant 4.916 12.613 0.000
X variable 0.165 1.114 0.271
Owner-requested changes % 0.263 0.069
Constant 5.015 11.362 0.000
X variable 0.009 0.051 0.959
RFI response time per unit price 0.239 0.057
Constant 5.183 17.507 0.000
X variable 0.044 0.389 0.699
RFI per unit price 0.219 0.048
Constant 5.104 19.564 0.000
X variable 0.129 1.306 0.198
Table X. Time overrun % 0.205 0.042
Correlation between Constant 4.931 13.138 0.000
CDM and project KPIs X variable 0.188 1.316 0.195

10

8
Cost of field rework score

7
Control projects
6
Data set projects
5

3 y = 0.0237x 3– 0.3081x 2 +1.8079x + 3.7705

R 2 = 0.7885
2

Figure 7. 1
CDM vs cost 0
of field rework
0 1 2 3 4 5 6 7 8
performance score
Collaborative decision making

of having several different stakeholders participate in decision making helps to ensure that
the decisions made are ones that incorporate the input of all parties. This can explain the
reduced amount of activities that have to be done more than once in the field to satisfy
the involved parties.
 The association between implementing various IPD principles in the delivery of a IPD
project and the project’s KPIs as identified by the regression analysis is summarized implementation
in Figure 8. framework
In addition to the statistical approach, the study conducted a focus group interview
session to validate the statistically identified correlations between improvement in KPIs of
the control projects and the respective IPD scores. The focus group involved participants
from the owners, engineers, and contractors of the control projects. The participants were 627
asked to rank the implemented IPD principles from 0 to 5 (0 being no effect and 5 being the
most effect) based on their perception of how influential these IPD principles were on each
KPI of the control projects. The participants were also asked to indicate their opinions
regarding the regression analysis results based on their experience with the control
projects. The majority of the identified high influence of the IPD principles on the KPIs
based on the focus group rankings correlated to the findings of the statistical analysis.
The focus group participants were in agreement with the findings that were obtained
through the regression analysis.

5. Discussion
When analyzing the individual links between the IPD principles and the KPIs displayed in
Figure 8, several interesting facts can be pointed out. While open communication was
expected to have an effect on the KPI of cost of field rework, the results show no such effect.
According to Abi Shdid et al. (2019), the implementation of IPD principles – including
open communication – in the delivery of a case study project resulted in high level of
improvement in cost of field rework. This is most probably due to the fact that field rework
is highly influenced by other factors such as inadequate project supervision, inadequate
design specifications and lack of proper inspection by the designer. It is important to note
that IPD principles implemented during the construction phase of a project show no effect
on design-related KPIs. For example, open communication, collaborative decision making

Open communication within

the project team and ability
to address issues

Integrated and collaborative

teams

Lean principles

Co-location of teams

Project’s staff performance

evaluations

Mutual respect and trust

Collaborative decision
making

Jointly developed project

target criteria

Cost Change Time RFIs per Errors and Total RFI Total cost Owner- Figure 8.
overrun % order cost overrun % unit price omissions claims cost response of field requested
% cost % % time per rework % changes % Links between IPD
unit price principles and KPIs
Link with a high Link with a low
identified by
No
predictive power predictive power
correlation
regression analysis
(R 2.0.7) (R 2 < 0.7)
ECAM and project’s staff performance evaluations had no effect on cost of errors and omissions or
27,3 on the RFIs per unit price. Implementing such IPD principles during the project execution
phase cannot affect such things as contract documents being incomplete, conflicting or
erroneous, thereby requiring revisions and clarifications to be provided by the designers.
CURT (2004a, b) stated that in order to achieve the desired IPD outcomes, owners must
collaborate through information sharing early in the design process.
628 Co-location of teams was also not implemented during the design phase, and the
results therefore show no effect on cost of owner-requested Changes. Implementing IPD
principles during the design phase of the project may reveal additional useful links.
NASFA et al. (2010) stated that the influence of IPD principles comes early in projects in
the form of their procurement and contracting process. It was noticed that the
absence of links between co-location of teams and the rest of the KPIs can be attributed to
the poor communication lines that exist within poorly managed co-located teams.
Well-managed strategies need to be planned to have an effective co-location of teams. It
was noticed that the absence of any links between IPD principles and KPIs related to
owner-requested changes can be attributed to the fact that owners try to use unspent
contingency money at the end of the project to satisfy their desires to add extra work or to
change the standards of work. If the construction project has no room for enhancements or
additions, owners try to create new projects to spend the contingency money under the
existing contract since that money has been allocated to be spent on construction
activities only.
The IPD principle related to integrated and collaborative teams was rarely implemented
in the data set projects, and therefore its effect could not be determined on any of the KPIs.
This explains why integrated and collaborative teams is commonly neglected by public
owners. According to Baiden and Price (2011), integrated and collaborative teams is a very
important tool in improved project delivery. It requires a spirit of cooperation to overcome
the traditional adversarial attitudes and barriers.
The focus group comprising the owners, engineers and contractors of the control water
and wastewater construction projects validated the individual links between the IPD
principles and the KPIs displayed in Figure 6 by attributing the KPI improvements to the
implemented IPD principles. This relation is further validated by the fact that the IPD
principles average scores of the control projects are 6.87 and 6.25, compared to the average
score of 2.07 for the data set. Aside from CO cost, all the KPI scores of the control projects
are higher than all the average KPI scores of the data set.
Due to the strong effects of the above mentioned IPD principles on their corresponding KPIs,
it is recommended to implement those IPD principles on all projects. It is important to note that
the implementation of open communication and collaborative decision making require a simple
effort and planning compared to co-location of teams. Co-location of teams will be efficient only
on large size projects due to the preplanning and large costs associated with it.
A guideline metric was developed as a tool to assist owners and contractors wishing to
improve particular KPIs of their water and wastewater construction projects. This tool
(shown in Figure 9) allows the user to choose which IPD principle to implement in a project
and what level of implementation is needed in order to attain a pre-desired percentage
of improvement in the score of a particular KPI that the user desire to improve. The level of
improvement designated in this metric is measured with respect to the average KPI score
of the data set. The description for each implementation level (L1–L4) for the three IPD
principles used in the guideline metric are shown in Tables AI–AIII. The use of this
guideline metric is illustrated in the example next.
An owner suffering from chronic cost overrun on previous projects and aiming to improve
this KPI in future projects, will use the guideline metric as follows. First, the user will enter
the “Project KPI” column of the guideline and locate “Cost Overrun” in the first row.
 IPD
Project KPI IPD IPD levela Improvement
principle in KPI
implementation
framework

L1 0%

L2 26%
629
Open
Cost overrun
communication L3 44%

L4 43%

L1 4%

RFI response L2 18%

Co-location of
time per
teams L3
unit price 18%

L4 76%

L1 35%

L2 61%
Total cost of Collaborative
field rework decision making L3 100%

L4 100%

Notes: aIPD implementation level L1 corresponds to a measuring scale of 2.5; IPD

implementation level L2 corresponds to a measuring scale 5; IPD implementation level L3 Figure 9.
IPD implementation
corresponds to a measuring scale 7.5; IPD implementation level L4 corresponds to a measuring guideline metric
scale 10 (as defined by Azhar, 2014)

Second, the user will move to the right within the selected row to identify the IPD principle
that has the most significant effect on the results of the KPI – in this case
“Open Communication.” The user will then proceed to the third column of the guideline
metric to examine the various levels of IPD implementations and the corresponding
improvement in the KPI (last column). In the case of cost overrun, level L1 implementation will
result in no improvement in the KPI, while levels L2, L3 and L4 will result in 26, 44 and
43 percent improvement, respectively. According to Azhar (2014), level L3 for
“Open Communication” requires that communication flow between project participants in
informally open and that meetings frequency between all project participants is high. While
level L4 would result in similar improvement in the KPI (43 percent decrease in cost overrun),
ECAM it is not the optimal implementation level for this KPI. This level would provide no additional
27,3 improvement in the KPI over level L3 while requiring additional meetings and communication
channels between the project participants.
It is clear from Figure 9 that the optimal combination of implementation levels
for open communication, co-location of teams and collaborative decision making IPD
principles is shown to be L3, L4 and L3, respectively. This combination will achieve
630 the most cost-efficient improvement in cost overrun, RFI response time and cost of field
rework KPIs.

6. Conclusions
Although the construction industry has a general positive perception of IPD, the correlation
between IPD and various project KPIs has so far been overlooked. This is the first study to
investigate the correlation links between various IPD principles and performance indicators
of a project through a statistical regression analysis of a rather large data set representing
actual water-related public projects. The results demonstrated that while significant
correlation does exist between some IPD principles and certain project KPIs, many widely-
accepted IPD principles provide little or no improvement in KPIs of water and wastewater
projects. This study thus provides owners and contractors of water and wastewater
construction project with a guideline framework for assessing the type and level of IPD
implementation needed to improve the success of a project, relative to database of past
projects delivered using the traditional DBB delivery method. This practical tool allows
utility owners, engineers and contractors to identify which IPD tools to implement in their
project delivery in order to improve certain aspects of the project performance. When
compared to the normally-accepted approach of implementing the full gamut of IPD
principles in a project, this selective implementation will save projects money as well as
managerial time.
While benchmarking using a single-number index is a powerful tool for measuring the
performance of a project and for identifying IPD principles for performance improvements,
it is not without limitations. The performance index offers very little indication of other
factors that can affect the performance of a project. Factors such as the project’s complexity
level, project manager’s managerial and leadership skills, and workers competency and
training have not been incorporated in this performance scoring model.
Further research is needed to better understand the relationships among IPD principles
and KPIs in public and private projects. Research including the implementation of IPD
principles during the design phase is essential to reveal additional useful links with the
KPIs. This will be worthwhile for owners to know the important explanatory variables that
they must pay close attention to in order that their projects can be completed to their
acceptable level of quality and satisfaction. Owners will be able to decide to what level they
should use IPD procurement method in order to obtain the desired results.

References
Abi Shdid, C. and Andary, E. (2016), “Improving the delivery process of water and wastewater
treatment plant public projects through the use of IPD principles: a case study”, 2016 AWWA
Utility Management Conference (UMC 2016), San Diego, CA, February 24–27.
Abi Shdid, C., Andary, E., Chowdhury, A. and Ahmad, I. (2019), “Project performance rating model for
water and wastewater treatment plant public projects”, Journal of Management in Engineering,
Vol. 35 No. 2, p. 04018064.
AIA (2007), “Integrated project delivery: a working definition”, American Institute of Architects
California Council, available at: http://ipdca.net/images/Integrated%20Project%20Delivery%20
Definition.pdf (accessed March 5, 2019).
AIA (2010), “Integrated project delivery: case studies”, American Institute of Architects California IPD
Council, available at: http://hga.com/sites/default/files/downloads/resources/ipd_casestudies_ implementation
aiacc_final_010410_0.pdf (accessed March 5, 2019).
framework
Alarcon, L.F. and Ashley, D.B. (1996), “Modeling project performance for decision making”, Journal of
Construction Engineering and Management, Vol. 122 No. 3, pp. 265-273.
Azhar, N. (2014), “Integrated construction project delivery system in the US public sector: an
information modeling framework”, Doctoral Dissertation, Florida International University 631
Archives, Miami, FL.
Azhar, N., Kang, Y. and Ahmad, I. (2014), “Factors influencing integrated project delivery in publicly
owned construction projects: an information modeling perspective”, Procedia Engineering,
Vol. 77, pp. 213-221.
Baiden, B. and Price, A. (2011), “The effect of integration on project delivery team effectiveness”,
International Journal of Project Management, Vol. 29 No. 2, pp. 129-136.
Chan, A. and Chan, A. (2004), “Key performance indicators for measuring construction success”,
Benchmarking: An International Journal, Vol. 11 No. 2, pp. 203-221.
Chan, A., Scott, D. and Chan, A. (2004), “Factors affecting the success of a construction project”, Journal
of Construction Engineering and Management, Vol. 130 No. 1, pp. 153-155.
CURT (2004a), “Collaboration, integrated information and the project lifecycle in building design,
construction and operation”, Construction Users Roundtable, WP-1202, Cincinnati, OH.
CURT (2004b), “Collaboration, integrated information, and the project lifecycle in building design,
construction and operation”, Construction Users Roundtable, available at: www.curt.org/
Training-WP1202-Collaboration-Integrated-Information-and-the-Project-Life-Cycle-in-Building-
Design.aspx
Darrington, J. (2011), “Using a design-build contract for lean integrated project delivery”,
Lean Construction Journal, pp. 85-91.
El Asmar, M. and Hanna, A.S. (2013), “Quantifying performance for the integrated project delivery
(IPD) system as compared to established delivery systems”, Journal of Construction Engineering
and Management, Vol. 139 No. 11, p. 04013012.
Ernstrom, B., Hanson, D., Hill, D., Clark, J.J., Holder, M.K., Turner, D.N., Sundt, D.R., Barton, L.S.I. and
Barton, T.W. (2006), The Contractors’ Guide to BIM, Associated General Contractors of America,
Arlington, VA.
Forbes, L.H. and Syed, M.A. (2011), Modern Construction: Lean Project Delivery and Integrated
Practices, CRC Press, Boca Raton, FL.
Kent, D.C. and Becerik-Gerber, B. (2010), “Understanding construction industry experience and
attitudes toward integrated project delivery”, Journal of Construction Engineering and
Management, Vol. 136 No. 8, pp. 815-825.
Kim, Y. and Dossick, C.S. (2011), “What makes delivery of a project integrated? A case study of
Children’s Hospital, Belleveue, WA”, Lean Construction Journal, pp. 53-55.
Kreyszig, E. (1979), Advanced Engineering Mathematics, Wiley, New York, NY.
Lam, E.W., Chan, A.P. and Chan, D.W. (2007), “Benchmarking the performance of design-build projects:
development of project success index”, Benchmarking: An International Journal, Vol. 14 No. 5,
pp. 624-638.
Ling, F.Y.Y., Chan, S.L., Chong, E. and Ee, L.P. (2004), “Predicting performance of design-build and
design-bid-build projects”, Journal of Construction Engineering and Management, Vol. 130 No. 1,
pp. 75-83.
Liu, J., Love, P., Davis, P., Smith, J. and Regan, M. (2014), “Conceptual framework for the performance
measurement of public-private partnerships”, Journal of Infrastructure Systems, Vol. 21 No. 1,
p. 04014023.
Matthews, O. and Howell, G.A. (2005), “Integrated project delivery an example of relational
contracting”, Lean Construction Journal, Vol. 2 No. 1, pp. 46-61.
ECAM Molenaar, K., Bogus, S. and Priestley, J. (2004), “Design/build for water/wastewater facilities: state of
27,3 the industry survey and three case studies”, Journal of Management in Engineering, Vol. 20
No. 1, pp. 16-24.
Mossman, A., Ballard, G. and Pasquire, C. (2010), “Integrated project delivery – innovation in integrated
design and delivery”, Draft for Architectural Engineering and Design Management.
NASFA, COAA, APPA, AGC and AIA (2010), “Integrated project delivery for public and private
owner”, available at: www.aia.org/aiaucmp/groups/aia/documents/pdf/aiab085586.pdf (accessed
632 February 14, 2019).
Newbold, P. (1991), Statistics for Business and Economics, 3rd ed., Prentice-Hall, Englewood Cliffs, NJ.
Pinto, M.B. and Pinto, J.K. (1991), “Determinants of cross-functional cooperation in the project
implementation process”, Project Management Journal, Vol. 22 No. 2, pp. 13-20.
Pocock, J.B., Hyun, C.T., Liu, L.Y. and Kim, M.K. (1996), “Relationship between project interaction and
performance indicator”, Journal of Construction Engineering and Management, Vol. 122 No. 2,
pp. 165-176.
Post, N. (2011), “Pioneers push paradigm shift”, Engineering News Record, McGrawHill.
Shane, J., Bogus, S. and Molenaar, K. (2013), “Municipal water/wastewater project delivery performance
comparison”, Journal of Management in Engineering, Vol. 29 No. 3, pp. 251-258.
Singleton, M.S. and Hamzeh, F. (2011), “Implementing integrated project delivery on department of the
navy construction projects”, Lean Construction Journal, Integrated Project Delivery Special
Issue, pp. 17-31.
Thomsen, C., Darrington, J., Dunne, D. and Lichtig, W. (2009), “Managing integrated project delivery”,
Construction Management Association of America, available at: http://cmaa.com/files/shared/
IPD_White_Paper_1.pdf (accessed February 10, 2019).
Water Design-Build Council (2013), “Design-build and design-bid-build project compared”, available at:
www.waterdesignbuild.com (accessed January 23, 2019).

Appendix. IPD scoring mechanisms (Azhar, 2014)

Score for open communication

10 7.5 5 2.5 0

Communication Communication Communication Communication flow is Communication flow is

flow is formally flow is flow is formally restrictive and routes restrictive and routes
Table AI. open and direct, informally open, open and direct, through long through long
Scoring mechanism frequency of frequency of frequency of transmission chain, transmission chain,
for open meetings is high meetings is high meetings is low frequency of meetings is frequency of meetings
communication high

Score for co-location of teams

10 7.5 5 2.5 0

When all key When all key When owner, When owner, When contractor is
participants are participants are engineer and engineer and located onsite and is
Table AII. located onsite and located onsite and contractor are contractor are dedicated full time
Scoring mechanism also all key also all key located onsite and located onsite and while owner or
for co-location of participants are participants are not are dedicated full are not dedicated engineer are not
teams dedicated full time dedicated full time time full time located onsite
 IPD
implementation
Score for collaborative decision making
10 7.5 5 2.5 0
framework
When input is taken When all key When input is taken When input is When major
from all key participants provide from owners, A/E taken from decisions are made
participants and also input while final and contractor and owners, A/E and on sole discretion of 633
all participants are decision makers are also same contractor while either owners/A/E or
involved in decision owners, A/E and participants are decision makers contractor, without Table AIII.
making contractor involved in decision are either only input from other Scoring mechanism
making owner/A/E or project participants for collaborative
contractor decision making

Corresponding author
Caesar Abi Shdid can be contacted at: [email protected]

For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: [email protected]