NATIONAL UNIVERSITY OF SCIENCE & TECHNOLOGY

FACULTY OF INDUSTRIAL TECHNOLOGY

DEPARTMENT OF INDUSTRIAL & MANUFACTURING ENGINEERING

MANUFACTURING SYSTEMS TIE 6111_SECTION A

2005

M. Mutingi and S. Mhlanga

i
 TABLE OF CONTENTS

1.0 CLASSIFICATION OF MANUFACTURING SYSTEMS....................................................................................1

1.1 The Manufacturing function................................................................................................................................1
1.2 Business implications of process choice (Hill 1993)...........................................................................................3
1.3 Hybrid Processes .................................................................................................................................................3
1.3.1 Batch Related Hybrids ................................................................................................................................4
1.4 The Role of Numerical Control in Hybrid Processes ..........................................................................................6
, 2.0 FACILITY LAYOUT AND DESIGN ....................................................................................................................8
2.1 Classes of Plant Layout Problems .......................................................................................................................8
2.2 Objectives Of A Good Plant Layout ...................................................................................................................8
2.3 Layout Types.......................................................................................................................................................8
2.3.1 Process Layout ............................................................................................................................................8
2.3.2 Product Layout ............................................................................................................................................9
2.3.3 Cellular Layout............................................................................................................................................9
2.3.4 Fixed Position Layout .................................................................................................................................9
2.3.5 Hybrid Layout .............................................................................................................................................9
2.4 Flow Analysis In Layout Problems ...................................................................................................................10
2.4.1 From-To Chart ..........................................................................................................................................10
2.4.2 Systematic layout planning (SLP) .............................................................................................................10
2.5 Potential Criteria for Evaluation of Layouts .....................................................................................................14
2.6 Simple Facility Location ...................................................................................................................................14
2.7 Pairwise Exchange Method for Layout Improvement.......................................................................................16
2.8 Computer Assistance in Facility Layout ...........................................................................................................18
2.8.1 Practical Session Using Computer Layout Design II (COMLAD II)........................................................18
3.0 FLOW LINE DESIGN AND ANALYSIS............................................................................................................20
3.1 Automated flow lines ........................................................................................................................................20
3.1 Configuration of automated flow lines..............................................................................................................21
3.1.1 In Line Type ..............................................................................................................................................21
3.1.2 Rotary Type...............................................................................................................................................22
3.1.3 Selection Of Configuration .......................................................................................................................22
3.2 Methods Of Workpart Transport .......................................................................................................................22
3.2.1 Continous Transfer ....................................................................................................................................23
3.2.2 Intermittent Transfer .................................................................................................................................23
3.2.3 Asynchronous Transfer .............................................................................................................................23
3.3 Analysis Of Automated Flow Lines..................................................................................................................23
3.4 Analysis Of Transfer Lines Without Storage ....................................................................................................25
3.4.1 The Upper Bound Approach .....................................................................................................................25
3.4.2 The Lower Bound Approach .....................................................................................................................25
3.5 Partial Automation ............................................................................................................................................26
3.6 Buffer Storage ...................................................................................................................................................28
3.7 Line Balancing ..................................................................................................................................................28
3.7.1 Terminology..............................................................................................................................................28
3.7.2 Methods Of Line Balancing ......................................................................................................................29
3.7.2.1 Largest candidate rule method ......................................................................................................30
3.7.2.2 Kilbridge & westers’ method........................................................................................................31
3.7.2.3 Ranked positional weights method (rpw) ..................................................................................32
3.7.3 Computerized Line Balancing Methods ....................................................................................................33
3.7.3.1 Comsoal ...........................................................................................................................................33
3.7.3.2 Bryton .....................................................................................................................................................34
3.7.4 Other Ways to Improve Line Balance .......................................................................................................36
3.8 Multistation Assembly Machines......................................................................................................................36
3.8.1 Analysis of Multi-station Assembly Machines .........................................................................................37
3.8.2 Measures of Performance..........................................................................................................................38
3.9 Assembly lines ..................................................................................................................................................39
3.9.1 Multi-model lines ......................................................................................................................................39
3.9.2 Mixed-model lines.....................................................................................................................................40
3.9.2.1 Line balancing of mixed-model lines ..........................................................................................40
3.9.2.2 Model Sequencing problem of mixed model lines: Method 1 .................................................42
3.9.2.3 Model Sequencing problem of mixed model lines: Method 2 .................................................42

ii
 3.10 Transfer Lines and General Serial Systems...................................................................................................44
3.10.1 Introduction ...............................................................................................................................................44
3.10.2 Paced lines without buffers ...........................................................................................................................45
3.10.2.1 Operation - Dependent failures .............................................................................................................45

iii
1.0 CLASSIFICATION OF MANUFACTURING SYSTEMS

1.1 The Manufacturing function

The principal function of the manufacturing process is to take inputs (materials, labour and energy) and
convert them into products. To do this a business must choose between different modes of manufacturing.
The chosen process is the one best able to support the company competitively in the market place. Each
choice of process has certain implication for the business in terms of response to its market, manufacturing
capabilities, level of investment required, unit cost involved and the style of management that is appropriate.

There are 5 generic types of manufacturing process and these are:

1: Project
2: Jobbing
3. Batch
4 Line
5 Continuous Processing

1. Project
Companies that produce one off, large scale, complex products will normally provide these on a
project basis. A project process concerns the provision of a unique product requiring the co-
ordination of large-scale inputs to achieve a customer’s requirement. *The resource inputs will
normally be taken to where the product is to be bought, since it isn’t feasible to move it once it is
completed. All activities, including all necessary support functions, will usually be controlled by a
total system for the duration the project.

2. Jobbing
A jobbing process is used to meet the one off order requirements of customers. The product
involved will be of an individual nature. This requires that the suppliers interpret the customers’
design and specifications and then apply relatively high skill levels in the conversion process. One
example is a tool room. The one off characteristic means that the product will not be required again
in its exact form, or if it is the demand will tend to be irregular with long time periods between
orders. For this reason, investment in the manufacturing process (jigs, fixtures and specialist
equipment) will not normally be warranted.

3. Batch
A Company decides to manufacture using batch processes because it is providing similar items on a
repeat basis usually in larger volumes than those associated in the jobbing process. This type of
process covers a wide range of volumes compared to other processes. At the low volume end, repeat
orders will be shown and infrequent. At the high volume end, the order quantities may involve many
hours, shifts or even weeks of work for the same product.

Each order quantity is manufactured by setting up that step of the process necessary to complete the
first operation for a particular product. The whole order quantity is completed at this stage. Then
the next operation in the process is made maybe and the total order quantity is completed, until all
stages required to make a product are completed. Meanwhile, the process used to complete the first
operation for the product is then reset to complete an operation for another product and so on.
Examples

1
 4. Line
With further increases in volumes, investment is made to provide a process dedicated to the needs of
a single product or a small range of products. The product range is determined at the time of
investment.

In a line process, products are passed through the same sequence of operations. The standard nature of the
products allows for this.

5. Continuous
With continuous processing, a basis material is passed through successive stages or operations and
refined or processed into one or more products. E.g. Petrochemicals. The choice of process is based
on two features, the first is very high volume demand and the second is that the materials involved
can be moved easily from one part of the process to another e.g. Fluids, gases and foods.

The high volume nature of the demand justifies the very high investment involved. The processes
are designed to run all day and everyday in minimum shut downs, due to the high costs of starting up
and closing down. Normally the product range is quire narrow and often the products offered are
purposely restricted in order to enhance volumes.

Another feature in continuos processing is the nature of the materials being processed.

Whereas in line there are manual inputs into the manufacture of the products as they pass along, in
continuous processing, the materials will be transferred automatically from one part of the process to
the next, with the process monitoring and self adjusting flow and quality. The labour tasks in these
situations predominantly involve checking the system and typically do not provide manual inputs
into the process as they would on a line.

Process choice
wide

Project civil engineering

one-off, purpose-built
Jobbing
Range

Engineering
Batch

Line motor vehicles

narrow

Continuos petrochemicals
process
Low Volume High

Figure 1: Types of manufacturing Processes

2
1.2 Business implications of process choice (Hill 1993)

JOBBING BATCH- LINE CONTINOUS

ASPECTS PROJECT PROCESSING
Products & Markets
Product type Special Special Ÿ Standard Standard
Product range very wide Wide Ÿ Narrow Very narrow
Customer order size Small Small Ÿ Large Very large
New product Introductions High High Ÿ Low Very low
What is the company selling? Capability Capability Ÿ Product Product
How are orders won? (Order Delivery speed/ Delivery speedŸ Price Price
winners) unique design / unique design
capability capability
How are orders won? (Order Price/ delivery Price/ delivery ŸQuality Quality/delivery
qualifiers) reliability/quality reliability/quality reliability
Manufacturing
Process Technology General Purpose Universal Ÿ Dedicated Highly dedicated
Process flexibility High High Ÿ Low Inflexible
Ÿ
Production volumes Low Very High
Low High
Delivery Ÿ Production
Key Manufacturing Task Delivery Production
Investment & Cost
Capital Investment Low/High Low Ÿ High Very high
Inventory Levels
Raw materials As required As required =>Medium =>Planned Planned with stock
Work In Progress (wip) High High Very High/Low Low
Finished goods Low Low Ÿ High High
%age of total costs
Direct labour Low High Ÿ Low Very Low
Direct Materials High Low Ÿ High Very High
Infrastructure
Appropriate Organizational Control Decentralized/ Decentralized Centralized Centralized
Centralized
Appropriate Organizational Style Entrepreneurial Entrepreneurial Bureaucratic Bureaucratic
Level of specialist support to High Low High Very High
manufacturing

1.3 Hybrid Processes

Many companies have developed hybrid processes to better enable them to support the characteristics of the
markets. Some comprise a mix of 2 of the 5 generic processes; others are developments within control (NC)
machines.
For each hybrid these is still a root process.

3
 Process choice

wide
Project

Jobbing
NC machines

Range
dedicated use of a general
machining centres purpose process
Batch
Group technology
Flexible manufacturing
Line systems (FMS) Transfer line

narrow
Mix mode assembly
Continuos
process
Low Volume High

Figure 2: Types of Manufacturing Processes and the hybrids

1.3.1 Batch Related Hybrids

(A) Numerical Control (NC) Machines

An NC system is a process that automatically performs the required operations according to a
detailed set of code instructions. Application includes metal cutting processes such as milling
boring, turning as well as bending and shearing. In reality, an NC machine is a development of a
batch process. It is batch because the machine stops at the end of one process and is reset for a new
job or new program being loaded. It is also low volume in nature.

(B) Machining Centre

Machine Centres combine NC operations previously provided by different machines into one
machining centre. With tool changing automatically controlled and carousels holding up to 150
tools or more, the intentions is to maximize the combination of operations completed at a single
location. *An MC completes many operations in sequence without removing items from the
process.

(C) Flexible Manufacturing Systems

Whereas an MC is best suited to low volumes an FMS is appropriated to mid volume requirements.
This is also designed to complete a given number of operations on an item before it leaves the
system. However, rather than the item being contained in a single centre, in an FMS the work-piece
is transferred from one process to the next automatically. The hybrid nature of FMS is based on
similar logic to that for MCs, i.e. maximizing the combination of operations completed at a single
location. The root process is still batch. FMS are designed around families of parts.

Look at the Companies on the Zimbabwe Stock Exchange for each company identify
the manufacturing system or process which predominates. Include service sector
companies.

These families can be categorized by

(1) Assembly - grouping parts together that will be required to make a single assembly e.g. a car
engine or any engine.
This system is designed to allow the user to order against an assembly requirement rather than
working within a complex scheduling environment where parts are produced in functionally laid out
processes.
(II) Type - categorizing parts by type for a range of similar products means higher volume production
and this reduces the number of changeovers required. This aggregate family demand justifies the capital
investment, with the inherent flexibility within the FMS allowing a wide range of products to be
manufactured.

4
D) Group technology
Group technology gains for batch processes some of the advantages in high volume line situations.
It does this by changing the process or functional layout associated to batch manufacturing into the
product layout associated to line.

E) Dedicated use of general purpose process

Where the volume of a specific part is sufficient, manufacturing can justify the allocation of a
process to its sole use. This dedication isn’ t in the plant itself but in the use of a general-purpose
process. Thus the potential flexibility of a GP process are retained, and they will be reclaimed when
volumes reduce. The process is still batch and in general purpose in nature. What becomes
dedicated is the use of the process, which reflects the volume requirements for the product in
question.
BATCH

Lathe Milling Grinding

LINE

Group technology

no changeover time (minimal)

Figure 3: Batch, Line and Group Technology Layouts

F) Mix mode assembly

A line process that can cope in a broad range of products without stopping is known as a Mix Mode
assembly line. The term mix-mode is used to reflect processes in which companies have made
systematic and purposeful investments to increase the product range involved. Typically, this entails
programming the line, to make small quantities of different products in a determined sequence.
What makes the root here a line process is that it does not have to be stopped and reset to
accommodate the next product.

Mix-mode assembly isn’ t technically a hybrid, it is simply a line process that can accommodate the
requirements of a wider range of products that can typically be made on a classic line process.

G) Transfer lines
A transfer line is a hybrid between line and continuous processes. However its root process is still
line because it can be stopped without major cost. The high demand associated in this process
justifies investment designed to reduce the manual inputs associated in a line process and to move
more towards a process that automatically transfers a part from one station to the next positions it,
completes the task and checks quality. Furthermore, deviations from the specified tolerances will be
registered and automatic tooling adjustments and replacements will often be part of the procedure.
To achieve this, the process is numerically controlled in part or in full.

5
 1.4 The Role of Numerical Control in Hybrid Processes
Of the hybrid processes mentioned above four are based on the concept of numerical control (NC). The
basis of choice between one NC hybrid process and another is volume.

The NC base of these processes brings in it a level of flexibility that is for greater than that inherent in non-
NC alternatives. This means therefore that the process is more able to cope with a wider range of products
and to handle product mix changes over time.

D
Hi ecre
gh a
er sin
Hybrid NC pr g p
oc ro
es ce
s i ss
Process Choice nv f
es lex
tm ibi
en lity
ta a
Machine Centre nd nd
low pro
er duc
Stardard co t r
st an
FMS m ge
an v
Dedicated uf ar
ac iet
tu y
rin
Mix Mode Assembly LIne g

Low Volume High

Figure 4: Hybrid NC process choice related to volume

o
pp
Process choice o
rt
wide

u n
Project civil engineering ity
co
one-off, purpose-built st
s
Range

Jobbing
O Engineering
Batch po ut o
ck f
et
co
narrow

Line st motor vehicles

Continuos petrochemicals
process
Low Volume High

Figure 9:
The area below the diagonal is characterized by out of pocket expenses. This area is where the rigidity and
capital intensity of the process are relatively further advanced than the acceptance of the process’ s products.
A company here places significant pressure on its marketing people to drum up the sales volumes needed to
sustain it relatively further advanced process characteristics. The Japanese have successfully employed this
strategy.
The area above the diagonal is characterized by, opportunity costs. Here, the processes characteristics are
relatively less advanced than its product acceptance.
The Company’ s profits suffer because process advances could manufacture the product for less i.e. for goes
the opportunity to earn more.
By positioning above the diagonal, the company places pressure on its manufacturing people to lower costs.

Tutorial and Self Asessment

1. Find one company in the country that represent each of the manufacturing generic group and justify the
information given in the notes.
2. Which companies in the country fall within the hybrid regions of classification? Justify your answers.
3. Choose two companies that you would recommend that they change from the current way of operation to
another. What would be business implications to your recommendation?

6
 References
Hill T (1993) Manufacturing Strategy: The strategic management of the Manufacturing function (2nd Ed.) pp. 105-
MacMillan. ISBN 0-333-57648-9
Groover M P (1985), Automation, Production systems, and Computer-Aided manufacturing
Askin R G and Standridge C R (1993) Modeling and Analysis of Manufacturing Systems, John Wiley & Sons

7
 2.0 FACILITY LAYOUT AND DESIGN
Layout is the physical organization or geography found at a facility/location.
Layout problems are stimulated by one of the following
x Product design change
x Introduction of new products
x Changes in volume of demand
x Facilities becoming obsolete
x Poor worker environment
x Change in location or market
x Cost reduction

2.1 Classes of Plant Layout Problems

Plant layout problems fall into the following four categories.
1. Minor changes in present layout
2. Existing layout rearrangement
3. Relocating into existing facilities
4. Building a new plant

The last 2 occur very infrequently whilst the first two are common in industry.

2.2 Objectives of A Good Plant Layout

An optimum plant layout is one that provides maximum satisfaction to employees, management and
stockholders. The major objectives of a good plant layout are:
1. minimize cost of materials handling
2. provide effective space utilization
3. minimize work in progress
4. provide for worker convenience
5. Promote job satisfaction and safety
6. Avoid unnecessary capital investment
7. Stimulate effective labour utilization

2.3 Layout Types

Basic layout types include process (functional), product, cellular and fixed position.

2.3.1 Process Layout

This is characterized by grouping like facilities or functions together for example grouping lathes in the
turning section, milling machines in the milling section etc.

Figure 5: Process Layout

8
 Job shop environments, for either goods or services, have tended towards process layout. The disadvantage
is that it puts distance between provider and customer

2.3.2 Product Layout

Here facilities are arranged along the flow of goods or services. The customer (next process) is adjacent to
or very near the provider, e.g. assembly lines. In manufacturing, product layouts can require substantial
investments and may not be ideal if output volumes are projected to be low.

Figure 6: Product Layout

2.3.3 Cellular Layout

The idea here is to arrange workstations into cells that process families of goods or services that follow
similar flow paths. Items are designed to have as many common features as possible and cell operators are
usually cross trained.

2.3.4 Fixed Position Layout

Here it is the product whose position is fixed and resources (man, machines and materials) must come to it
e.g. civil construction, ship-building and hospital patients.

Two or more layout types in a single facility are very common. This is termed a mixed layout e.g. in the Sun
hotel the food can be brought to you whilst you are seated or you can choose your own food by moving in a
plate and picking the food-staffs you desire.

2.3.5 Hybrid Layout

One feature of the product layout and cellular layout is the arrangement of stations into a U-shape. A U-
shaped layout has the following advantages:
1. Staff Flexibility and Balance
The U-shape allows operators to tend to several work-centres near or across the U, as the distance is short.
Balancing work among operators is easier.

Figure 7: U-shaped Layout

(2) Teamwork - Getting all staff into a cluster enhances teamwork and joint problem solving

9
 (3) Rework
When a mistake occurs a product is normally sent back to the department or machine or workstation
where it occurred or to a separate rework line. In a U-shaped layout, the distance to return a mistake
is short, making it easier to follow the (Total Quality Management) TQM principle of correcting
mistakes at source.

(4) Work and Tool Distribution

Since all stations in a U are immediately accessible from the centre it is easier to distribute materials, parts,
instruction sheets etc.

2.4 Flow Analysis In Layout Problems

In a complex layout situation, one layout principle widely used is to arrange work areas in order of dominant
flow. Flow analysis is important in layout design. Two examples of flow analysis techniques are:
1. Flow process chart (See handout)
2. From To Chart

2.4.1 From-To Chart

From-to charts provide information concerning the number of material handling trips made between two
centres of activity and the total material handling distance. They are a convenient means of reducing a large
volume of data into a workable form. By inspecting the data displayed in the “form-to” chart a layout
analyst can identify the departments having large volumes of item movement and can develop a layout
design in which these departments are located close to each other. An e.g. of a “from-to” chart is shown
below

*total distance covered in a given period (Distance between machine * number of trips done in a period)

TO
FROM SAW LATHE DRILL MILL INSPECTION STATION

SAW 120* 576

LATHE 160 528
DRILL 96 80 288
MILL 128
Inspection Station 160

“From-to” charts are useful in:

x justifying a proposed layout
x analyzing material movement
x developing detailed layout arrangements.
x Evaluating layout alternatives.
x Showing the interrelationship of product lines.

2.4.2 Systematic layout planning (SLP)

This is a widely used approach that consists of the following steps.

10
STEPS POSSIBLE TOOLS
Analyse Product flows From-to chart

Identify and include non-flow factors Activity-relationship (REL) chart

Asses data and arrange work areas Activity arrangement diagram

Determine space arrangement plan space relationship diagram

Fit space arrangement into available space Floor plan Detailed layout modes

The SLP procedure is best illustrated by means of an example

Example

The five departments of a warehouse within approximate space requirement and activity relationships are as
follows.

2 ACTIVITY RELATIONSHIPS
Department AREA (m )

Material Scheduling 1000

E
4 A
Packaging and Crating 1500 1
O A
E 2 I
Materials Supervisor 500
I I
O
Shipping and Receiving 3000
A
6000 1
Storage
Total 12000

Closeness Rating
A- Absolutely necessary
E- Especially important
I- Important
O- Ordinary closeness
U- Unimportant
X- Not desirable

Reasons Code
1- Work flow
2- Supervision
3- Safety
4- Communication

(a) Develop an activity arrangement diagram based on the REL chart data
(b) Develop a space relationship diagram for the 5 departmental areas.
(c) Fit the 5 departments into a 100m x 150m building and try to maintain 10m aides between the
department

ANSWERS
(a) Activity Arrangement Diagram
This diagram shows the arrangement of all activities but without indicating requirements for space, utilities,
aisles etc.

11
 Rating lines are introduced with 4 lines corresponding to A rating, 3 lines to an E rating etc.

Figure 10: Activity arrangement diagram

(b) Space Relationship Diagram

This shows the arrangement of activities with space data included. The diagram is generally
rectangular in shape. It may be regarded as a rough layout inactivity blocks showing relative size

Figure 11: Space Relationship Diagram

(c) Layout Plan

12
 Figure 12: Layout Plan

Figure 13: Diagrammatic Representation of the SLP Procedure

13
2.5 Potential Criteria for Evaluation of Layouts
These include:
(1) Ease of future expansion or contraction
(2) Adaptability & versatility
(3) Layout Flexibility
(4) Flow or movement effectiveness
(5) Material handling effectiveness
(6) Storage effectiveness
(7) Space utilization
(8) Safety and housekeeping
(9) Utilization of natural conditions

Possible layout performance measurement ratios

DLMHD
1. DLMH
DL
Direct labour material handling ratio = Direct labour material handling( dollars)
Direct labour total (people)

2. SSV = OSS
TSS
Storage Space utilization = Storage Space Occupied by materials or occupied storage space
Total storage space

3.
ASP= AS
TS

Aisle space proportion = Aisle Space (Space)

Total Space

4.
MO = M
O

Move Operations ratio = Moves (total)

Operations (total)

5.
ADM = TD
M

Average distance per move = Total distance

Moves (total)

2.6 Simple Facility Location

A special case of the facility layout - positioning problem involves what is termed simple facility location.
This problem arises where there are existing facilities at known distinct point and a new facility must be
located at some point within the layout. In real situations, there may be only a handful of practical positions
to locate the new facility.

14
 The following are EXAMPLES
(i) A new pump in a chemical operation (factory)
(ii) A new warehouse relative to facilities and customer
(iii) A component in an electrical network
(iv) A new appliance in a kitchen
(v) A photocopier in a library

A general formulation of the problem is as follows:

Existing facilities are located at known distinct points P1,……,Pm; A new facility is to be located at point X;
costs of a transportation nature are incurred that are directly proportional to an appropriately determined
distance between the new facility and an existing i
letting d(X, Pi) represent the distance travelled per trip between points X & Pi, and letting wi represent the
product of cost per unit distance travelled and non of trips made per year between the new facility and i.e.
total annual cost due to travel between the new facility and all existing facilities is given by

¦
w d ( X
m
f ( x) i i Pi ) .
i 1
Where the wi terms are sometimes referred to as weights.

The one facility location problem is to determine the location of the new facility that minimises f(x),the total
annual transportation cost.

The appropriately determined distance is typically a straight line distance or rectangular distance.

Straight line or Euclidean distance is given by

d  X , Pi  x  a    y  bi 
2 2
i

Where the coordinates for the new facility are x and y and for the existing facility i, are ai and bi so that

X= (x, y)

Pi = (ai, bi)

In movement practical location problems, travel only occurs along a aisle arranged in a rectangular pattern
parallel to the walls of the building. This is referred to as the rectilinear or rectangular distance and is given

d  X , Pi 
by

x  a i  y  bi
In a typical factory situation, transportation cost is assumed to be a simple function of distance. In addition,
there is an associated load/unload cost per trip at each facility.

The total cost is therefore given by summing up the load/unload cost and the transportation cost. If Lk is the
load/unload cost index and Cx is the transportation cost index, the problem is formulated as follows:

f X ¦  L R 
 Ck Ri d  X , Pi 
m

k i
i 1

Where Ri = Number of trips per period to facility i.

Example
With reference to the table below customers 1 to 6 can be saved from either location A (30,30) or B(80,60).
Given that the load/unload cost per trip Lk = $1.00 and the cost per unit distance Ck = $0.001/m for all
customers 1 to 6 which is the best location.

15
 Customer Co-ordinates No of trips/period Measurement type
x y
1 40 60 185 Rectilinear
2 50 20 300 Rectilinear
3 60 40 150 Rectilinear
4 80 40 205 Straight line
5 40 40 200 Straight line
6 60 60 175 Straight line

Solution

From location A(30,30)

Ck = $0.001 Lk = $1.00
Transportation Load/Unload
Customer Distance No of trips Total distance Cost Cost
1 40 185 7 400 7.4 185
2 30 300 9000 9.0 300
3 40 150 6000 6.0 150
4 51 205 10455 10.4 205
5 14 200 2800 2.8 200
6 42 175 7350 7.35 175

$43.005 $1 215

Total Cost A = $ 1258,005

From location B(80,60)

Customer Distance No of trips Total Distance Transport Cost Load/Unload
Cost
1 40 185 7400 7.4 185
2 70 300 21000 21.0 300
3 40 150 6000 6.0 150
4 20 205 4100 4.1 205
5 45 200 9000 9.0 200
6 20 175 3500 3.5 175
$51,00 $1215

Total Cost B = $ 1266

Choose location A

2.7 Pair-wise Exchange Method for Layout Improvement

Majority of layout problems involves the redesign of an existing facility with the problem of coming up with
an improvement. This heuristic method is on layout improvement based on minimizing the total cost of
transporting materials among all departments in a facility. Assuming that the distance between departments
is rectilinear and is measured from the department centroids.

Example
Consider four depts of equal sizes. The material flows between departments are given in Table 2.7.1 and the
existing layout is shown in Figure 2.7.1. a distance matrix can be obtained based on the existing layout as
given in Table 2.7.2.

16
 The total cost for the existing layout is computed as follows:
TC1234= 10(1) + 15(2) + 20(3) + 10(1) + 5(2) + 5(1) =125
The pairwise exchange method simply states that for each iteration, all feasible exchanges in the locations of
the dept pairs are evaluated and the pair that results in the largest reduction in total cost is selected. Since all
depts areas are assumed to be equal size, the feasible exchanges are 1-2, 1-3, 1-4, 2-3, 2-4 and 3-4.

Table 2.7.1: Material Flow Matrix

To dept
1 2 3 4
1 - 10 15 20
From Dept 2 - 10 5
3 - 5
4 -

(a) Iteration 0 1 2 3 4

(b) Iteration 1 3 2 1 4

(c) Iteration 2 2 3 1 4

Figure 2.7.1: Layouts corresponding to each iteration

Table 2.7.2: Distance Matrix Based on Existing Layout

To dept
1 2 3 4
1 - 10 15 20
From Dept 2 - 10 5
3 - 5
4 -

The distance matrix is recomputed each time an exchange is performed. The total costs resulting from these
exchanges are:
TC2134(1-2) = 10(1) + 15 (1) + 20(2) + 10(2) + 5(3) + 5(1) =105
TC3214(1-3) = 95
TC4231(1-4) = 120
TC1324(2-3) = 120
TC1432(2-4) = 105
TC4213(3-4) = 105

Thus the pair 1-3 is selected and performs the exchange in the layout as shown in Figure 2.7.1(b)

For the next iteration all feasible exchanges are considered which consists of the same set as in iteration 1.
The resulting total costs are:
TC3124(1-2) = 10(1) + 15 (1) + 20(2) + 10(1) + 5(1) + 5(3) =95
TC1234(1-3) = 125
TC3241(1-4) = 110
TC2314(2-3) = 90
TC3412(2-4) = 105
TC4213(3-4) = 105

The pair 2-3 is selected with a total cost value of 90. Figure 2.7.1 (c) shows the resulting layout after 2
iterations. Continuing on to the third iteration calculations are:
TC3214(1-2) = 10(1) + 15 (2) + 20(1) + 10(1) + 5(1) + 5(2) =95
TC1324(1-3) = 120
TC3421(1-4) = 125

17
 TC2134(2-3) = 90
TC3412(2-4) = 105
TC4123(3-4) = 95

Since the lowest total cost for this iteration, 95, which is worse than the cost value of 90 in the second
iteration, then the procedure is terminated. Thus the final layout arrangement is 2-3-1-4 as shown in Figure
2.7.2 (c).

The pairwise exchange procedure described above is not guaranteed to yield the optimal layout solution
because the final outcome is dependent on the initial layout. Thus we can claim local optimality.

2.8 Computer Assistance in Facility Layout

Some of the drudgery effort can be avoided by using computer software in the search phase of layout
planning. One program CORELAP (computerized relationship layout planning), uses closeness ratings from
the chart as inputs. It produces a single layout of rectangular shaped departments.

It can be used for either office or factory layouts. A program called CRAFT (computerized relative
allocation of facilities technique) requires an existing layout as input. Its purpose is to improve the layout. It
uses flow data only (from-to) with the objective being to minimize material handling costs.

2.8.1 Practical Session Using Computer Layout Design II (COMLAD II)

Handouts will be given and diskette copy of program.

Tutorial

1. Collolog Coop, is building a new plant. 8 departments are involved. As part of a plant layout
analysis, the activity relationships and the area needs for the departments are shown on the REL Chart
below.

Department 1 2 3 4 5 6 7 8 Area
1. Shipping and A I O E U U O
Receiving 3 3 1,2 600
2. Stores A E O O U O
3 3 3 3 1,2 1500
3. Fabrication E U E O O
3,4 1,3 3 1,2 800
4. Assembly E O O I
3 1,3 3 1,2 700
5. Painting U X U
5 500
6. Tool Room O U
3 300
7. Canteen O
3 600
8. Offices
1200

18
Closeness Ratings Code Reasons
A Absolutely necessary 1 Personal Contact
E Especially important 2 Paperwork Contact
I Important 3 Product/Resource Flow
O Ordinary 4 Use same equipment/tools
U Unimportant 5 Possible Fumes
X Not desirable

(a) Develop an Activity Arrangement diagram based on the REL chart date.
(b) Develop a space relationship diagram for the 8 departments
(c) Fit the 8 departments into a 100m x 80m building in as close to an optimal layout as you can include aisle
between departments in the layout.
Allowing for an aisle of 5m wide.

2. With reference to the table below, customers 1-6 can be served from either location A(x = 20, y=50)
or Location B (x = 80, y = 70) or C(x=40, y=60) given that the load/unload cost, Lk = $0.75 per trip
and the cost per unit distance, Ck = $0,002/m for customers 1,2 and 3 and Lk = $0,50 per trip and Ck
= $0,003/m for customers 4,5 and 6. Which is the best location

Customer Co-ordinate No of trips movement

x y per period type
1 40 60 250 Straight line
2 60 70 165 Straight line
3 80 70 201 Straight line
4 40 30 104 Rectangular
5 10 80 306 Rectangular
6 10 60 55 Rectangular

Reference/ Bibliography
1. Groover M P (1985), Automation, Production systems, and Computer-Aided manufacturing
2. Askin R G and Standridge C R (1993) Modeling and Analysis of Manufacturing Systems, John Wiley &
Sons,
3. Sule D R, (1994), Manufacturing Facilities: Location, planning and design, PWS Publishing Company
ISBN 0-534-93435-8
4. Tompkins, J A, White, J A, Bozer, Y A, Tanchoo, J M A and Trevino, J, (1996), Facilities Planning,
John Wiley & Sons

19
 3.0 FLOW LINE DESIGN AND ANALYSIS
In flowline items are processed as they pass through a series of work stations along a line. Generally, raw
materials, items or components are fed in the beginning of and at certain points along the line, and goods are
delivered from the end of the line.
Assembly lines: flow lines which are engaged essentially, in product assembly.
Transfer lines: Flow lines which use automatic material transfer between the automatic machining stations.
Multi-model : processing of two or more products in separate batches
Mixed -model: processing of two or more products simultaneously on the line.
Table 3.1 summarises the classes and varieties of Flow-line.
Table 3.1: Classes and varieties of flow line
Flowline description Number of Product Flow of items Setting of
products changes equipment and
allocation of work
Class Variety
Single 1 No changes required
model None Regular
Transfer Multi- >1 Batch Regular Changes of equipment
line model changes batches setting and/or work
allocation required on
change of batch
Single >1 none Irregular (a) No change required
model
Assembly Multi- >1 Batch Irregular (a) changes of
Line model changes equipment setting
and/or work
allocation required
or batch change
mixed >1 Continualc Irregular changes of
model equipment setting
and/or work
allocation normally
requiredb
a
Because of variable work station times-characteristic of manual flow lines
b
Alternatively as in group technology; tasks and equipment might be permanently allocated a specific group
of components.
c
At any time the line contains a mixture of product types.

3.1 Automated flow lines

An automated flow line consists of several machines or workstations. These are linked together by work,
handling devices that transfer parts between the stations. The transfer of work parts occurs automatically and
workstations carry out their specialised functions automatically. Transfer lines/machines are examples of
automated flow lines.

Manual assembly lines are another example of flow lines. They consist of multiple workstations in which
the assembly work is accomplished as the part or sub-assembly is passed from station to station along the
line. At each workstation one or more human workers perform a portion of the total assembly work on the
part. When the part comes off the final station the work has been completed.

Another variation is the automated assembly system, which makes use of automated methods of the
workstations rather than human beings.

Automated flow lines can be represented with symbols as follows.

20
 Symbol Component
Workhead x Workstation
XXXX XXXX:
YYY PROC = Processing Station
ASBY = Assembly Station
INSP = Inspection Station
SORT = Sortation Station
YYY:
AUT = Automated
MAN = Manual
x Work part
Raw Work part
Partially Processed Part
Finished Part

Storage buffer

Automated flow lines are generally the most appropriate means of production in cases of relatively stable
product life; high product demand and therefore high rates of production; and where the alternative method
of manufacture would involve a large labour content. The objective of the use of flow lines automation are:
1. to reduce labour costs
2. to increase production rates
3. to reduce work in process or work in progress
4. to minimise distances between operations or distances moved between operations
5. to achieve specialisation of operations

3.1 Configuration of automated flow lines

The 2 general forms that a flow line can take are inline and rotary

3.1.1 In Line Type

The in-line configuration consists of a sequence of workstations in a straight-line arrangement. However, the
flow of work can take a few 90° turns either for workpiece orientation or due to factory layout limitations. A
common pattern is the rectangular shape, which allows the same operator to load the raw workpiece and
remove the finished workpiece.
An example is:

PROC PROC PROC

AUT AUT AUT

AUT
PROC
AUT
PROC

PROC
AUT
AUT
PROC

PROC
AUT

Figure 14: Configuration of Automated Flow in Line Type

21
A rectangular shape can still qualify as an in-line.

3.1.2 Rotary Type

In the rotary configuration the workparts are indexed around a circular table or dial. Workstations are
stationary and located around a periphery of a dial. The parts or workparts ride on the rotating table and are
registered or positioned at each station for processing. (This type of arrangement is usually called dial or
indexing machine)

AUT
PROC

PROC
AUT

AU OC
T
Figure 15: Configuration of Automated Flow Rotary Type PR

3.1.3 Selection Of Configuration

The choice between 2 types depends on the application. The rotary type is commonly limited to smaller
workpieces and to few stations. Typically, buffer storage cannot be catered for. The in-line design is
preferable for larger workpieces and can accommodate a larger number of workstations. There is usually
built in storage capability to smooth out the effect of work stoppage at individual stations and other
irregularities.

3.2 Methods Of Workpart Transport

The general methods of transporting work-parts or workpieces on flow lines can be classified into the
following 3 categories.

(1) Continuous flow/transfer

(2) Intermittent or Synchronous Transfer
(3) Asynchronous or Power-and- free transfer

These 3 categories are distinguished by the type of motion that is imported to the work-part/workpiece by the
transfer mechanism. The most appropriate transport system for a given application depends on features such
as:
I. The type of operation to be performed
II. The number of stations on the line
III. The weight and size of the work-part
IV. The existence of manual stations within the line
V. Production rate requirements
VI. Balancing the various process times on the line

22
 3.2.1 Continous Transfer
Here work-parts are moved continuously at a constant speed. This requires the workheads
to move during processing to maintain continuous registration in the work-part i.e.
continuous contact with the work-part. Examples can be formed in beverage bottling
operations and manual assembly operations where the operator can move in a moving flow
line.

3.2.2 Intermittent Transfer

Here workparts are transported in an intermitted or discontinuous motion. The workstations are in a
fixed position and the parts are moved between stations and then registered at the proper locations
for processing.

3.2.3 Asynchronous Transfer

Here each part moves independently of other parts. Some parts can be processed on the line at the
same time that others are being transported between stations. Asynchronous transfer systems are
more flexible than the other two but their cycle rates or production rates are generally slower.

3.3 Analysis Of Automated Flow Lines

Assuming a synchronous transfer system automated flow lines analysis can be approached as follows:
Parts are introduced into the 1st workstation and are processed and transported at regular intervals to
succeeding stations. This interval defines the ideal or theoretical cycle time of the flow line. This is equal to
the time required for parts to transfer plus the processing time at the longest workstation. The components of
Tc these are shown on the diagram below:

Figure 16: Production time

Because of breakdowns on the line, the actual average production time Tp will be longer than the ideal cycle
time. When a breakdown occurs at any station we assume that the entire line is shut down. Let Td represent
the average down time to diagnose the problem and make repairs when a breakdown occurs due to a
particular reason j. The frequency with which line stops per cycle occur for reason j is devoted by Fj. The
term FjTdj gives the mean time per cycle that the machine will be down for reason j. If there is only one
reason why the machine may be down, the average production time Tp is given by:

Tp Tc  FTd (1)

If there are several reasons why the line is down the average production time becomes

Tp Tc  ¦ FjTdj (2)
j

The average production rate Rp is based on the average production time Tp and is given by
1
Rp (3)
Tp

23
 Because of the possibility of work-parts being scrapped during processing the theoretical production rate is
rarely achieved in practice and is given by

1
Rc (4)
Tc
The time efficiency E is the proportion of time that a line is up and operating and is given by:

Tc Tc
Tc  FTd
E (5)
Tp

Alternatively the proportion down time is given by:

FTd FTd
Tc  FTd
D (6)
Tp

NB: ED 1 (7)

Another measure of flow time performance is the cost per item produced. If Cm represents the cost of raw
materials per product, CL-the cost per minute to operate the line and Ct-the cost of disposable tooling per
workpiece, then the total cost per workpiece Cpc is given by:

C pc Cm  C L Tp  Ct (8)

Example 1
Suppose a 10 station transfer machine is being considered for production of a component used in a pump.
The item is currently produced by more conventional means that demand cannot be met. The production
department has estimated that the ideal cycle time Tc is 1 minute. From similar transfer times, it is estimated
that breakdown of all types will occur with a frequency F = 0.10 breakdowns per cycle and that the average
downtime per line stop will be 6 minutes (Td = 6 mins)
The scrap rate for the current processing method is 5% and this is considered a good estimate for the transfer
times. The starting costing for the component costs $1.50 each and it will cost $60 an hour ($1 a minute) to
operate the transfer line. Cutting tools are estimated to cost 15c per work-part.

Find
(a) the production rate
(b) the number of hrs required to meet a demand of 1500 units per week
(c) Line efficiency
(d) cost per unit produced

Solution
(a) average production time Tp = Tc + FTd
Tc = 1 min
Tp = 1+ 0.1(6) = 1.6 minutes

Therefore average production rate

Rp = 1/Tp = 1/1.6 = 0.625 pieces per minute
Correcting for scrap rate of 5% the actual production rate of good production rate of good products:
Rp = 0.95 * 0.625 = 0.59375 pieces/min
= 35.625 pieces/hr

(b) Assuming that we want 1500 good units plus scrap. The Number of hrs required are:

24
 hours = 1500 =42.1 hrs
35.625

(c) The line efficiency

E= Tc = 1 =0.625
Tp 1.6

(d) The cost per unit produced Cpc

Cpc = 1/0.95(Cm + CLTp + Ct)

= 1/0.95( 1.5 + 1.6*1 + 0.15)
= $3.42 per good unit produced

3.4 Analysis of transfer lines without storage

In the previous example, a breakdown frequency was estimated. In considering the analysis of continuous
and intermittent transfer machines without internal buffer storage capacity we must consider further what
happens when a workstation breaks down.
There are 2 possibilities and their analyses are referred to as the upper bound and lower bound approach.
With the upper bound approach, we assume that the work-part is not removed from the workstation when
breakdown occurs at that station. With the lower approach the work-part is taken out of the station and the
station breaks down.

3.4.1 The Upper Bound Approach

This approach provides an estimate of an upper limit on the frequency of line stops per cycle. It is assumed
here that when a break down occurs at a station it is not necessary for the part to be removed from that
station. In this case it, is possible that there would be more than one line stop associated with a particular
work-part. Let Pi represent the probability that a part would jam at a particular station where i = 1, 2, … , n.
Since the parts are not removed when a jam occurs, it is possible (though not probable) that the part would
jam at every station. The expected number of line stops a part is given by

¦ P
n
F i (9)
i 1

If the probabilities Pi are all equal

P1= P2 = … = Pn = P then
F nP (10)

Example
In a 10 station transfer mine the probability that a station breaks down will occur for a given work-part is
0.01. The probability is the same for all workstations. Determine the frequency of line stops per cycle on
this flow-line using the upper bound approach.

Solution
F = nP
= 10 * 0.01
= 0.1

3.4.2 The Lower Bound Approach

This approach gives an estimate of the lower-limit on the expected number of line stops per cycle. In this
approach we assume that the station breakdown results in damage or destruction e.g. A drill breaking off in a
part during processing. The broken drill must be replaced and the work-part must be removed from the line
for the subsequent or the scrap. Let Pi be the probability that the part would join at station 1.

25
 The term (1-Pi) is the probability that the part would not join at station 1 and thus be available from
subsequent processing. The quantity P2(1-P1) is the probability that the given part would jam at station 2. In
general the quantity Pi(1-Pi-i)(1-Pi-2) … (1-P2)(1-P1) for i = 1, 2, … , n. Is the probability that the given part
would jam at station. Summing all these probabilities from i= 1 through to n would give the probability of
line stops per cycle.

Alternatively; the frequency may be determined as follows: The probability that a given part will pass

 1  P 
through all n stations without a line stop is given by:

–
n

i
i 1

Therefore the frequency of line stops per cycle is given by:

1  – 1  Pi 
n
F (11)

i 1

1  1  P 
If the probabilities, Pi that a part would jam at a particular station are equal P1 = P2 = … , Pn = P then
n
F (12)
gives the frequency of time stops per cycle.

With the lower bound approach a number of work-parts coming off the line would be less than the number
starting on the line. Therefore the production rate formula becomes:
1 F
Rp
Tp
Where F represent the frequency of line stops per cycle.
Therefore the term (1-F) represents the yield of the transfer machine. Tp is the average cycle time.

Example
Compute the value of F, using the lower bound approach for the data in the previous example.

F = 1-(1-0.01)10 = 0.0956

To calculate the production rate.

Rp = (1-0,0956)/ Tp

Tp = Tc + FTd = 1+ 0.0956*6 = 1.5736

Rp = (1-0.0956)/1.5736 = 34.5 pieces/hr

The difference between the UBA and the LBA is usually small but grows in high p and n.

Observations
1. Line efficiency decreases substantially as the number of stations increases.
2. The UBA leads to lower efficiency values but higher production rates.

3.5 Partial Automation

There are many examples of flow lines in industry that contain both automated and manually operated
workstations. These cases occur because
1. Mechanization of a manually operated flow line is often introduced gradually.
2. Some operations are difficult to automate and it may be uneconomical to do so.

26
 To analyze performance, we initially assume an automated flow line without buffer. The ideal cycle time
Tc is determined by the slowest machine which we assume to be a manual station. We also assume that no
breakdown occurs at a manual station. Let P be probability of station breakdown and n be the total number
of workstation, na be number of automated station, no = number of manually operated stations.

na  no n

Co is operator cost per manual station ($ /min)

Cas is the cost per automated workstation ($/min)
Cat is the cost per minute of the automated transfer mechanism per line.

Combining the above costs gives

CL no Co  na Cas  Cat (13)

The average production time is given by

Tp Tc  na PTd (14)

The cost per unit produced Cpc is then given by

C pc Cm  no Co  na Cas  Cat  T  na PTd   Ct (15)

Example
A proposal has been made to replace one of the current manual stations with an automatic workhead on a 10
station transfer line. The current system has 6 automatic workheads and 4 manual stations. The proposed
automatic station would allow the cycle time to be reduced to 24s. The new station is cost at $0.25/min.
Other cost data for the existing line are: Co = $0.15/min
Cas = $0.10/min
Cat = $0.10/min
Breakdowns occur at each of the 6 automatic workstations to a probability p= 0.01. The average down time
per breakdown is 3 mins. It is estimated that the value of P for the new automatic station is P = 0.02. The
average down time for the line will be unaffected. Material for the cost 50c per unit (Cm = $0.50/Unit).
Tooling costs can be neglected (Ct=0). It is desired to compare the new automated station with the current
manual station on the basis of cost per unit.

Solution
The current line

Tp = Tc + naPTd = 0.5 + 6*0.01*3 - 0.68 mins/unit

CL = noCo + naCas+ Cat = 4*0.15 +6*0.10+0.1

= $1.30/min
Cpc = Cm+CL(Tc+naPTd) + Ct
= 0,5+1,30(0,68) + 0
= $1.38/unit

For a new line

Tp = Tc + naPTd
= O.4 + (6*0.01*3) + 1*0.02*3
= 0.64 min/unit

CL = 3*0.15 + 6 + 6*0.10 + 0.25 + 0.1

= $1.40/min

27
 Cpc = Cm + CLTp + Ct
= 0.5 + 1.40*0.64 + 0
= $1.396/unit

Therefore you do not need to replace the manual station.

3.6 Buffer Storage

The preceding analysis has assumed that there is no buffer storage between stations on the line. When the
automated portion of the line breaks down, the manual station must also stop working because of lack of
work-parts. It will be beneficial to the line operation to build up an inventory, of parts before and after each
manual station, i this manner, these stations could continue to produce while the automated it portion of the
line is down.

Figure 17: Workstations with Buffer Storage

3.7 Line Balancing

Assembly involves the joining together of 2 or more separate parts to form a new entity. The new entity is
called a sub-assembly or as assembly. In flow line production, the sequence of processing or assembly steps
is restricted in terms of the order in which the operation can be carried out. These restrictions are called
precedence constraints.

The line balancing problem is to arrange the individual processing and assembling tasks at the work stations
so that the total time required at each station is approximately the same. If the work elements can be grouped
so that all the station times are exactly equal, there is perfect balance on the line. In practice perfect balance
is unachievable and the slowest station determines the production rate of the line.

3.7.1 Terminology
In order to spread the job done on a line among its station, the job must be divided into its component tasks.
The minimum rational work elements are the smallest practical indivisible tasks into which the job can be
divided.

Let Tj be the minimum rational work element, where j is used to identify the element out of the elements that
make up the total work or job. Then the total work content, Twc, is given by

¦ T
n
Twc j (1)
j 1

Workstation Process Time: A workstation is a location along the flow line where work is performed either
manually or by some automatic device. The work performed at the station consists of one or more of the
individual work elements and the time required is the sum of the times of the work elements done at the
station.

Tsi is used to indicate the process time at station i of an n - station line.

¦ Tsi ¦ T
n n
(2)

i 1 j 1
j

28
 The cycle time Tc must meet the following condition

Tc d
E
(3)
Rp

Where E is the line efficiency and Rp is the required production rate

The balance delay is a measure of the line efficiency, which is a result of idle time due to imperfect
allocation of work among stations. It is given by
nTc  Twc
d
nTc
Example
Considering the work elements shown in the table below.

Work elements

No. Element description Tej (min) Preceded by

1 Place frame as work-holder and clamp 0.2 -
2 assemble plug 0.4 -
3 assemble brackets to frame 0.7 1
4 wire power code to motor 0.1 1,2
5 wire power cord to switch 0.3 2
6 assemble plate to bracket 0.11 3
7 assemble blade to bracket 0.32 3
8 assemble motor to brackets 0.6 3,4
9 align blade and attach to motor 0.27 6,7,8
10 assemble switch to motor bracket 0.38 5,8
11 attach cover, inspect, test 0.5 9,10
12 pack 0.12 11

Given that table above then

¦ T
n
Twc j = 4.0 min
j 1

Assume Tc = 1 min and assuming perfect balance,

then the balance delay d = (4*1- 4)/(4*1.0) = 0
Practically it does not happen
Assume n = 5 and Tc = 1min then
d = (5*1.0 - 4.0)/ (5*1.0) = 0.20 or 20%
There would be a balance delay of 20%.

E.g. 1 2 3 4 5
n=5 0.6 1 0.5 1 0.9

Tc = maximum of station process time Tci

This is not a perfectly balanced line.

3.7.2 Methods Of Line Balancing

There are several methods for solving the line-balancing problem by hand. These methods are based on
logic and common sense rather than on mathematical proof. They don’ t guarantee an optional solution but
are likely to result in good solutions, which approach the true optimum. Manual methods considered are:

1. Larger candidate rule method

2. Kilbridge and Westerns method
3. RPW - ranked positional weights method

29
3.7.2.1 Largest candidate rule method

This is the easiest method. The work elements are selected for assignment to stations simply on the basis of
the size of their Tej values. The steps used in solving the line balancing problem are:
(a) List all elements in descending order of Tej value, in the largest Tej at the top of the list.
(b) To assign elements to the 1st workstation, start at the top of the list and work down, selecting the 1st
feasible element is one that satisfies the precedence arrangements and does not cause the sum of the
Tej values at the station to exceed the cycle time Tc.
(c) Continue the process of assigning work elements to the station as in step (b) until no further elements
can be added wont exceeding Tc.
(d) Repeat steps (b) + (c) for the other stations on the line until all the elements have been assigned.

Example continued

Work Element Tej (min) Preceded by

3 0.7 1
8 0.6 3,4
11 0.5 9,10
2 0.4 -
10 0.38 5,8
7 0.32 3
5 0.3 2
9 0.27 6,7,8
1 0.2 -
12 0.12 11
6 0.11 3
4 0.1 1,2

Precedence diagram

0.11
6
0.7 0.27
3 7 9
0.2 0.5 0.12
0.32 4
1 11 12
0.1 0.6
4 8
0.4 0.3
2 5 10
0.38

(b) Tc = 1 min

30
 Station Elements Tej(min) ¦Tej at Station
1 2 0.4 0.4
5 0.3 0.7
1 0.2 0.9
4 0.1 1.0
2 3 0.7 0.7
6 0.11 0.81
3 8 0.6 0.6
10 0.38 0.98
4 7 0.32 0.32
9 0.27 0.59
5 11 0.5 0.5
12 0.12 0.62

Minimum n is the smallest integer t Twc / Tc

n = Twc / Tc = 4.0 / 1.0 = 4

3.7.2.2 Kilbridge & westers’ method

This is heuristic procedure (uses complex rules) which selects work elements for assignment to stations
according to their position in the precedence diagram. The elements at the front of the diagram are selected
first into the solution. This technique has been applied to several complicated line-balancing problems in
good success.

The steps in the procedure are:

(1) Construct the precedence diagram so that nodes representing work elements of identical precedence
are arranged vertically in columns.
(2) List the elements in order of their columns in column 1 at the top of the list. If an element can be
located in more than 1 column, list all the columns by the element to show transferability of the
element.
(3) Assign elements to workstations, starting in the column 1 elements. Continue the assignment
procedure in order of column number. Until the cycle time is reached.

Step 1 Precedence diagram

0.11
6
0.7 0.27
3 7 9
0.2 0.32 0.5 0.12
1 11 12
0.1 0.6
4 8
0.4
0.3
2 5 10
0.38

I II III IV V VI

31
Step 2.

Work Element Column Tej (min) Sum of Column Tejs

1 I 0.2 0.6
2 I 0.4
3 II 0.7
4 II 0.1
5 II, III 0.3 1.1
6 III 0.11
7 III 0.32
8 III 0.60 1.03
9 IV 0.27
10 IV 0.38 0.65
11 V 0.5 0.5
12 VI 0.2 0.12

Tc = 1

Station Column Tej (min) Sum of Column Tejs

1 1 0.2 0.2
2 0.4 0.6
4 0.1 0.7
5 0.3 1
2 3 0.7 0.7
6 0.11 0.81
3 7 0.32 0.32
8 0.6 0.92
4 9 0.27 0.27
10 0.38 0.65
5 11 0.5 0.5
12 0.12 0.62

1 0.81 0.92 0.65 0.62

1, 2, 4, 5 3, 6 7, 8 9, 10 11, 12

3.7.2.3 Ranked positional weights method (rpw)

This method combines the strategies of the largest candidate rule and Kilbridge and Westers method. An
RPW value is computed for each element. The RPW takes account of both the Tej value of the element and
its position in the precedence diagram. The elements are then assigned to workstations in the general order
of their PRW values.

Steps in the procedure are:

Step 1 Calculate the RPW for each element by summing the elements’ the Tej together with the Tej values
for all the elements that follow in the arrow chain of the precedence diagram.
Step 2 List the elements in the order of the RPW, in the largest RPW at the top of the list. for convenience,
include the Tej value and immediate predecessors for each element.
Step 3 Assign elements to stations according to RPW, avoiding precedence constraint and time cycle
violations

32
 Work Element RPW Tej (min) Immediate Predecessor
1 3.3
2 2.67
3 3
4 1.97
5 1.3
6 1.0
7 1.21
8 1.87
9 0.89
10 1.0
11 0.62
12 0.2

Station Work Element RPW Tej(min) ¦Tej Immediate Predecessor

1 1 3.3 0.2 0.2 -
3 3.0 0.7 0.9 1
2 2 2.67 0.4 0.4 -
4 1.97 0.1 0.5 1,2
3 8 1.87 0.6 0.6 3,4
5 1.3 0.3 0.9 2
4 7 1.21 0.32 0.32 3
6 1.0 0.11 0.43 3
10 1.0 0.38 0.81 5,8
5 9 0.89 0.27 0.27 6,7,8
11 0.62 0.5 0.77 9,10
12 0.12 0.12 0.89 11

0.9 0.5 0.9 0.81 0.89

1, 3 2, 4 8, 5 7, 6, 10 9, 11, 12
1 2 3 4 5

Comment on the best method

The ranked positional weights method (RPW) is the best method since the bottleneck station (one with the
largest value of workstation process time Tsi) has the smallest value for the 3 methods.

Actual Solution

0.9 0.91 0.92 0.65 0.62

1, 3 2,4,5,6 8, 7 10, 9 11, 12

3.7.3 Computerized Line Balancing Methods

The 3 methods described in the preceding sections are generally carried out manually. Computer programs
have been developed based on several heuristic approaches to yield optimal solutions. Use of computing
power (speed) allows a more complete enumeration of the possible solutions to a line balancing problem
then is practical in a manual method. An example of a computer based technique for solving large scale line
balancing problems is COMSOAL (computer method of sequencing operations for assembly lines).

3.7.3.1 Comsoal
It involves selecting a seed work element from a predefined list and then iterating through a sequence of
attendance solutions and then choosing the best one.

33
 To minimise total delay time ** with four workstations
Step 1 Construct list A with work elements in one column and the total number of elements that immediately
precede each element in a adjacent column.
Step 2 Construct list B showing all elements from list A that have no immediate predecessors
Step 3 Select at random one of the elements from list B The only constraint is that the element selected must
not cause cycle time Tc, to be exceeded
Step 4 Eliminate the element selected in step 3 from list A and B and update both lists
Step 5 Again select one of the elements from list B which is feasible for cycle time
Step 6 Repeat steps 4 and 5 until all elements have been allocated to stations within the cycle time
constraints
Step 7 Retain the current solution and repeat steps 1 to 6 in an attempt to determine an improved solution. If
an improved solution is found it should be retained

3.7.3.2 Bryton
There are 2 types of line balancing scenarios:
x Fixed production Rate to optimise number of stations
x FIXED work stations to minimise cycle time
The Bryton procedure will find a local optimum, given a set number of workstations. The procedure
minimises overall idle time whilst also improving the production rate

Solution Procedure
The procedure here is a convergence procedure. It gets closer and closer to the solution in each step The
steps are:
1) Establish required data (element times, precedence requirements and the production volume
required for a given time interval)
2) Make preliminary calculations of cycle time Tc the minimum number of stations, nmin and the
cycle time associated with the minimum number of stations Tcmin
3) Devise a feasible arrangement
4) Select maximum and minimum Tsi
5) Interchange the elements
6) Re-examine all Tsi values after interchange
7) Repeat interchanging procedure
8) Repeat steps 6 and 7 until total delay appears constant

Example
Element times
Element Tej/ minutes per piece Precedence
Requirements
1 1.0
2 0.6 1
3 2.7 2
4 1.2 2
5 3.4 4
6 0.9 3
7 1.6
8 1.1 5, 6, 7
Total 12.5

The assembly line operates 2 shifts a day ( Assume shift is 8 hrs and a week is 5 days) Management would
like 1300 units per week. Our task is to find the optimum assembly line balance

Solution
1. Establish all the data required
2. Tc=8*2*5*60/1300=3.7 min/cycle
nmin=12.5/3.7=3.4 therefore we work with 4 stations
Tcmin=12.5/4=3.125

34
 Deriving a precedence matrix from a precedence graph
||P|| =Pij where I represents row position and j column position
Pij =1 if i precedes j
Pij =0 if I and j are unordered
Pij =2 if I follows j
Elements j
1 2 3 4 5 6 7 8
1 0 1 1 1 1 1 0 1
2 2 0 1 1 1 1 0 1
3 2 2 0 0 0 1 0 1
4 2 2 0 0 1 0 0 1
Element i 5 2 2 0 2 0 0 0 1
6 2 2 2 0 0 0 0 1
7 0 0 0 0 0 0 0 1
8 2 2 2 2 2 2 2 0

3 You can use any method e.g. LCR (largest candidate rule)
(b) Repeat step 5 (a) until all stations have been examined with respect to maximum Tsi. If no interchange
has occurred it is now a local optimum. If the local optimum seems to be the best solution to the problem,
select the next rated maximum and compare it with the minimum Tsi etc. (as above). But if this local
optimum is just a satisfactory solution go to step 5 (d)
(d) If the local optimum is not satisfactory, a large element may be shifted out of the maximum Tsi or the
largest element in the whole group may be moved. Let this element be assigned permanently to the new
station and start the procedure again with step 4

n maxTsi   ¦Tsi
Stations

¦ d
1 2 3 4 Max ds n n

i 1
s

i 1

Te1=1.0 Te4=1.2 Te1=0.9 Te8=1.1

Initial Te2=0.6 Te5=3.4 Te1=1.6
Solution Te3=2.7
4.3 4.6 2.5 1.1 3.5 5.9

Interchange -Te5 +Te5 4*4.5-

(4.3+1.2+2.5+4.5)
1 4.3 1.2 2.5 4.5 3.3 5.5

etc.

Computer Precedence Check

For computer purposes a solution may be stated in the form of a solution matrix
e.g. The initial feasible solution
||S||
1 2 3 4 5 6 7 8
1 0 1 1 1 1 1 0 1
2 2 0 1 1 1 1 0 1
3 2 2 0 0 0 1 0 1
4 2 2 0 0 1 0 0 1
Element i 5 2 2 0 2 0 0 0 1
6 2 2 2 0 0 0 0 1
7 0 0 0 0 0 0 0 1
8 2 2 2 2 2 2 2 0

35
To check precedence restrictions, the computer adds the row of the interchanged element from the proposed
solution matrix ||S||. If a 3 appears in the sum then the precedence restrictions have been violated.

3.7.4 Other Ways to Improve Line Balance

The techniques described above rep. Strict and precise procedures for allocating work elements to stations
according to a specified cycle time. There are, however, other possible ways of improving the op of the line.
1. Devising work elements-Where possible technologically it may be beneficial to sub: further a minimum
rational work element. This then provides more flexibility in allocating work elements to stations.
2. Changing work-head speeds at automatic stations-It may be possible to reduce the process time at a
bottleneck station by increasing its speed/feed rate. This may however result in shorter tool life and more
frequent line stops for tool replacement. At now bottleneck stations, however, where there is idle time,
the workhead speed and feed could be reduced to prolong tool life and reduce frequency of down-time.
3. Methods Analysis- It is concerned in the study of human work activity for possible improvements. It is
applicable on manual flow lines. Waste analysis on automated flow lines may also lead to improvements
in work-flow and balance.
4. Preassembly Of Components-To reduce the total amount of work done on the regular assembly line,
certain subassemblies can be prepared off-line, either by another assembly cell in the plant or by
purchasing them from an outside vendor that specializes in the type of processes required
5. Inventory Buffers Between Stations- Storage buffers on both automated and manual lines serve to smooth
out disruptions due to variations in process times, breakdowns and small line imbalances.
6. Parallel Stations-It may be beneficial for some stations to be arranged in parallel rather than sequentially.
Considering our previous example, the total work content time Twc = 4,0min and the minimum no. Of
stations is 4. However, none of the methods used so far were able to achieve a solution with less than 5
stations. By considering 11 stations it is possible to achieve a perfectly balanced solution in 4 stations as
follows:

Station Elements Te (min) ¦Te at station

1,2 in parallel 1 0.2
2 0.4
3 0.7
4 0.1
8 0.6 2
5 0.3
6 0.11
7 0.32
9 0.27 1.0
4 10 0.38
11 0.5
12 0.12 1.0

1
1,2,3,4,8 3 4
2 5,6,7,9 10,11
1,2,3,4,8

There is no formalized procedure for developing a solution that utilizes 11 stations. The use of 11 stations is
easier to implement on manual lines because of the inherent flexibility of the human operator.

3.8 Multistation Assembly Machines

Here we assume that components are added or joined to an existing assembly. The existing assembly
consists of a base part plus the component assembled to it at previous stations.

36
Base part Completed assembly
As by As by As by
Aut Aut Aut

3.8.1 Analysis of Multi-station Assembly Machines

Defects occur with a certain faction rate, q, which can also be considered as the probability that the next
component on an assembly workstation is defective. If m = probability that a defect will result in
malfunction and stoppage of a workstation then qi and mi relate to stations i where i =1, 2,… , n.

At any station i, there are 3 possible events that might occur when the feed mechanism attempts to feed the
component and the assembly device attempts to join it to the existing assembly.

1. The component is defective and causes a station jam

2. The component is defective but does not cause a station jam
3. The component is not defective

The probability of the 1st event is given by

Pi mi qi ........................................................................................................................... (1)

Where Pi is probability that a part will jam at station i

q is the probability that a part is defective

The 2nd possible event, when the component is defective but does not cause a station jam, has a probability
given by (1  mi ) qi .

With this outcome a bad part is joined to the existing assembly.

The 3rd possibility, that the component is not defective, is given by (1-qi)

The probabilities of the 3 possible events must equal unity

mi q i  1  mi q i  1  qi  1 ........................................................................................ (2)

mq  1  mi q  1  q 
For the special case where qi = q + mi = m for all i, then
1 ............................................................................................. (3)

To determine the complete distribution, of possible outcomes that can occur on an n-station assembly line,

> m q  1  m q  1  q  @
the terms in equation 2 are multiplied together for all n stations to give:

–
n

i i i i i 1 ............................................................................ (4)
i 1

for the special case where all mi and qi are equal:

>mq  1  mq  1  q @
n
1 ......................................................................................... (5)

Expansion of equations 4 and 5 will reveal the probabilities for all possible sequences of events which can
take place on the assembly line.

37
3.8.2 Measures of Performance
We are not required to calculate every term to make use of the concept of assembly machine operation. Get

1  q  mi qi  .................................................................................................. (6)
the proportion of acceptable product coming if the line, Pap, we use.

–
n
Pap i
i 1

The above is the proportion of assemblies with no defective components, the proportion of assemblies that
contain least one defect is given by

1  – 1  qi  mi qi  ............................................................................................. (7)
n
Pqp

i 1

1  q  mq  n .......................................................................................................... (8)
The case where mi = qi the above 2 equations become

1  1  q  mq  .................................................................................................... (9)
Pap
n
Pqp

The proportions given by equations 6 & 8 give the yield of assembly machine.
Calculate production rate. We must first determine the frequecy of down time occurrences per cycle, F,
which given by:

¦ P ¦ m q
n n
F i i i ........................................................................................................ (10)
i 1 i 1

all Pi and mi are equal, their

F nmq .......................................................................................................................... (11)

The average production time per assembly is given by:

Tc  ¦ mi qi Td ......................................................................................................... (12)
n
Tp
i 1
(From Tp =Tc + FTd)

or Tp Tc  nmqTd .......................................................................................................... (13)

When all mi + qi are equal.

 1  q  mi qi 
Therefore the rate of production of acceptable product, Rap is given by

–
n

i 1
i Pap
Rap ...................................................................................... (14)
Tp Tp

Where the production rate has been corrected to give the rate of output of assemblies that contain no defects.
When all mi and q i are equal the equation then becomes

1  q  mq  n
Rap ......................................................................................................... (15)
Tp

38
The line efficiency is given by

Tc Tc
Tc  nmqTd
E ..................................................................................................... (16)
Tp

The proportion downtime

nmqTd nmqTd
Tc  nqmTd
D ............................................................................................. (17)
Tp

The cost per assembly produced, which is corrected for yield is given by

Cm  C L T p  Ct Cm  C L Tp  Ct
1  q  mq 
C pc n ..................................................................... (18)
Pap

Example

A 10 section in-line assembly line/machine has a 6s ideal cycle time Tc = 6s. The base part is automatically
headed prior to the 1st station, and components are added at each of the stations. The fraction defect rate at
each of the 10 stations is q = 0.01 and the probability that a defect will jam is m = 0.5.

When a jam occurs the average downtime, is 2 mins. Determine the average production rate, the yield of
good assemblies and the up time efficiency of the assembly machine.

3.9 Assembly lines

3.9.1 Multi-model lines

Multi-model line-approach requires the following major decisions to be made:
1) How will the line be balanced?
2) What will be the batch sizes of the models?
3) In what order will the batches be processed (The batch sequencing problem.)

The optimum sequence for the batches of different models is influenced by the cost of setting up the line.
The total cost of setting up the line comprises the cost of tool and machine changeover, tool and machine
resetting, machine and labour idle time etc., and is clearly influenced by the nature of the preceding and
succeeding models. The problem therefore is to determine the sequence order of the model batches to
minimize the total setting-up cost over a given period of time.

Example
Applying Hungarian method in allocation considering the costs.
Cost Succeeding Model
A B C D A B C D
A 0 100 150 80 A 1000 100 150 80
B 50 0 100 75 B 50 1000 100 75
C 80 40 0 110 C 80 40 1000 110
D 115 100 60 0 D 115 100 60 1000

39
 A B C D ?Sequence = {Do C o B o A o D}
A 920 20 70 0 (80) Ao D
B 0 950 50 25 (50) BoA
C 40 0 960 70 (40) CoB
D 55 40 0 940 (60) Do C
{A o D o C o B o A o D o C}

3.9.2 Mixed-model lines

Mixed-model line approach requires the following major decisions:
1) How will the line be balanced
2) In what order will the models be launched into the line (the model searching problem).

This has the advantage that a steady flow is produced in order to meet demand requirements, theoretically
without the need of large stocks of finished goods. The major disadvantage arise from the differing work
contents of the models, resulting in the uneven flow of work and consequent station idle time and/or
congestion of semi-finished products.

3.9.2.1 Line balancing of mixed-model lines

It can be considered merely as several single-model balancing problems i.e. each model could be considered
separately and the total work content divided as equally as possible between the workstations. If model A
and B are similar it is possible to balance for A and then for B minimising balancing loss on condition that
the processing of each model involves similar work elements to be undertaken in a similar order or when the
processing of all models merely involves the repetition of similar work elements.

In a case of dissimilar models independent balancing will result in dissimilar work elements being allocated
to each station. Thus balancing should be undertaken in a way to ensure the similar work elements are
allocated to the same work stations or groups of stations irrespective of which model is being processed.
This can be achieved by assigning elements to stations on a total time rather than a cycle time basis.

Consider two dissimilar models A and B for which the precedence relationship of the elements are shown
below.
11 12 13 7

0 2 5 6 8 9 10

17
1 3 4
14 15 16 18
Model A (200/week) {0,1,2,3,4,5,6,7,8,9,10}
Model B (100/week) {1,3,4,7,8,9,11,12,13,14,15,16,17,18}

40
Table : Data for two model line balancing
Element Element Total time/wk Positional weight
No Duration (h) A B Total/wk (hr) (using total times)
0 0.32 200 0 200 64 411
1 0.1 200 100 300 30 463
2 0.2 200 0 200 40 347
3 0.05 200 100 300 15 292
4 0.1 200 100 300 30 277
5 0.23 200 0 200 46 307
6 0.2 200 0 200 40 247
7 0.05 200 100 300 15 125
8 0.32 200 100 300 96 206
9 0.1 200 100 300 30 110
10 0.3 200 0 200 60 60
11 0.1 0 100 100 10 167
12 0.15 0 100 100 15 157
13 0.17 0 100 100 17 142
14 0.08 0 100 100 8 61
15 0.07 0 100 100 7 53
16 0.13 0 100 100 13 46
17 0.2 0 100 100 20 20
18 0.13 0 100 100 13 13

Positional weights are calculated by summing total times.

Applying Ranked Positional weight technique the results in table below are obtained assuming available time
per station per week is 40 hrs (five-eight hour days)

Table: Two model assembly line balance.

Station No: Element No. Total time per week for Time remaining from 40-
element (hr) hour(hr)
1 1 30 10
11 10 0
2,3 0 64 16
3 15 1
4 2 40 0
5,6 5 46 34
4 30 4
7 6 40 0
8,9,10 8 96 24
12 15 9
14 8 1
11 13 17 23
7 15 8
15 7 1
12 9 30 10
13,14 10 60 20
16 13 7
15 17 20 20
18 13 7

It has been necessary to combine several work stations because the total time for elements (0,5,8 and 10) is
greater than the available 40 hours.
Balancing loss = 5.1% which is a respectable line balance for the week’ s production. However this method
of line balancing is justified only when production is truly mixed-model. If there is a tendency to send
models through the line in batches rather than individually a quite unsatisfactory situation would result.
When model-utilised and incur a great deal of balancing loss and while the batch of model B is being
processed stations 2,3,4,5,6,7,13 and 14 will be under-utilised.

41
Thus this method of ‘combined’ line balancing for a shift’ s or week’ s production of all models is beneficial
where:
a) models are to be processed concurrently on the line and not in batches
b) dissimilar work elements are involved and it is desirable to ensure that work of a similar nature is
allocated to separate stations or groups of stations

3.9.2.2 Model Sequencing problem of mixed model lines: Method 1

This problem is concerned both with time interval between the “ launching” or starting of models onto the
line and also with the order in which models are launched onto and flow along the line. The objective of
such sequencing is to provide for the best utilization of the line which is characterised by minimum station
idle time and minimum congestion of work along the line.

Example
Three models A, B&C of a particular product are assembled concurrently on an assembly line.
The quantities required over a given period and the model cycle times are as follows:

Model No of unitsRequired Model cycle time

j N Cj
A 60 0.5
B 110 0.6
C 55 0.8

Calculate the fixed interval at which nuts must be launched onto the line, and show how the sequence of
models might be determined in order to avoid station idle time:

¦N C 30  66  44
¦N
D
60  110  55
j j
0.62

Units must be launched onto the line in such a way that multiples  of the launching interval are less than, but
¦
as nearly equal as possible to, the sum of the model cycle time,  j C  
h
h 1

Table: Fixed-interval launching of units onto an assembly line.

¨ ¦ Ch ¨ ¦ Ch  i
Deviation Model § j · § j ·
Units i i Model cycle time Cj ©h 1
 ¹̧ ©h 1
  ¹̧
A B C
1 0.62 -.12 -.02 .18 C 0.8 0.8 0.18
2 1.24 .06 .16 .36 A 0.5 1.3 0.06
3 1.86 B 0.6 1.9 0.04
4 2.48 B 0.6 2.5 0.02
5 3.10 B 0.6 3.1 0.00
6 3.72 C 0.8 3.9 0.18
7 4.34 A 0.5 4.4 0.06
8 4.96 B 0.6 5.0 0.04
9 5.58 B 0.6 5.6 0.02
10 6.20 B 0.6 6.2 0.00

3.9.2.3 Model Sequencing problem of mixed model lines: Method 2

Assuming that the desired product mix is known. Let qj be the proportion of product type j, j=1, ---, P to be
produced. First step is to develop an assembly line balance for the weighted average product. Let tij be the
time to perform task i on product type j and Sk the set of tasks assigned to workstation k.

42
 Average feasibility condition can be stated as:

¦
 ¦
q j t ij d C
p
k=1,- - -,K
i Sk j 1

This condition states that averaged across all items produced in the long term, no workstation is overloaded.

¦
The feasibility condition indicates that there is need only solve one single product assembly-line balancing
p
problem. In solving this problem task times of t i j 1
q j t ij are used.

Sequencing products on the line now becomes the issue. For each item j we must produce Qj items this
period (shift). Let r be the greatest common denominator of all Qj. There is need to construct a repeating

¦
cycle comprised of Nj = Qj/r units of product type j, j=1, ----, P. The cycle will be repeated r times to satisfy
p
period demand. N j 1
N j items are produced each cycle. A cycle that both smoothes the production
rate of each item and prevent excessive idle (delay) time at workstations is the aim because of mix-induced
starving of workstations. A workstation is starved if on completion of all its tasks there is no item available
for it to work on because the next item has not yet been completed at the prior station. Maintaining a

¦
constant flow is most important at the bottleneck station. Define the relative work load for station k as
Ck  t . The bottle neck station kb is the station with maximum total work (or equivalent average
i S i

work per cycle) i.e. k bc

k

arg max k Ck

To determine the dispatch sequence, Let Xjn be 1 if item type j is placed in the nth position and 0 otherwise.
Likewise j(n) will denote the type of item placed nth. The approach becomes selecting the nth item to be
entered to the line to optimise the following problem.

¦" ¦
! t , j (n)  nC
n

 
Minimize max
1 n N
i kb
j 1 i Skb

¦# X
N
subject to jn Nj j = 1,-------P (1)
n 1

 S1 d ¦ X jh d  S1
nN j n nN j
n = 1, ---- N, j = 1, ----- P (2)
N $
h 1 N

¦& ¦& ¦ X jh d n  s2 Ck
n P
t n=1, ----N; k=1, ---,K; Xjn = 0 or 1 (3)
%
h 1 j 1 i Sk
ij

The objective function minimizes the maximum deviation from assigning average work load to the
bottleneck station at any point during the period. Constraints (1) insure that all items are produced during the
cycle. Constraints (2) restrict the production rate of each product to be within Si of its average rate at all
times. Constraints (3) limit maximum over-utilization at all times. These constraints attempt to restrict
unplanned station idle time due to starving.

Sequencing Heuristic

Step 0. Initialization. Create a list of all products to be assigned during the cycle. Call this list A.
Step 1: Assign a product for n=1,---, N from list A, create a list B of all product types that could be assigned
without violating any constraint. From list B select the product type (j*) that minimizes;

max ¦
# ¦i
t , j (n)  nCkb
n

1 n N
j 1 i Skb

Add product type j* to the nth position. Remove a product type j* from A and if nN goto 1

43
Example
It would be foolish to market just one toy car. Every child wants to “ collect the whole set.” Suppose the
times used to balance the line are actually averages over all the models in this product line. The estimate
sales by model are:
Table: Estimated weekly sales by model type.
Model Sales Percentage Station 2 Time
Red Z 250 16.7 1/6 72
Blue Q 250 16.7 1/6 68
Black R 500 33.3 2/6 68
RWB American 500 33.3 2/6 66

Solution
Required production of each model in a cycle is 1 Red, 1 Blue, 2 Black, and 2 RWB per cycle (NZ=1, NQ=
1, NR = 2, NRWB = 2). Thus N=6.

Stage Cum Red Z Blue Q Black R RWB Choice Cum Assigned

time Assigned time
1 68 1/6, 4 1/6, 0 1/3, 0 1/3, 2 Black R 68
2 136 1/3, 4 1/3, 0 -1/3, 0 2/3, 2 Blue Q 136
3 204 ½, 4 -1/2, 0 0, 0 1, 2 RWB 202
4 272 2/3, 2 -1/3, 2 1/3, 2 1/3, 4 Red Z 274
5 340 -1/6, 6 -1/6, 2 2/3, 2 2/3, 0 RWB 340
6 408 0, 4 0, 0 1, 0 0,0 Black R 408
E(units produced )-units competed prior to stage i;Cum slack if model is selected
Cum slack = total time – total assigned time

3.10 Transfer Lines and General Serial Systems

3.10.1 Introduction
Assembly lines previously discussed assume 100% reliability (workstation never break down) and
processing times are deterministic. In this case flow-line systems are subject to breakdown and processing
time variability. A breakdown can be cause by fractured tool or jamming of the transport mechanism in a
transfer line. Or it may be due to temporary unavailability of worker on an assembly line or blockage on
upstream station.

Transfer line may be defined as a set of serial, automatic machine and or inspection stations linked by a
common material transfer and control system.

Transfer lines and automated flexible flow lines are capital intensive and must be kept running to be
justifiable. Thus the objective is to determine the effectiveness of a line given both buffer capacities and
failure and repair rates for each workstation. Effectiveness is measured by the net production rate.

1 1 2 2 3 3 4

# workstation # # Buffer #

Figure 3.1: Schematic of four-stage line with intermediate buffers

Station may be “ down” for one of four reasons:

i) Station Failure - a fractured tool, quality out of control, missing/defective part program or jammed
mechanism. In this case the station may stop while the other stations continue.

44
ii) Total line failure - exists in all stations are in operative- power outrage or error in central line controller
iii) Station blocked - when if on completion of a cycle it is unable to pass the part to station it 1. This may
be due to a failure in handling system, failure of a down stream station prior to the next buffer or failure
of a down stream station with the intermediate buffer between these stations currently being full.
iv) Station starved - if an upstream failure has halted the flow of parts into station I - Thus station will be
idle.

Failure may be also classified as time or operation (cycle) dependent failures.

i) Time-dependant failures occur with a chronological frequency independent of the number of successful
cycles since the last failure. Hence the span of time between time dependent failures is measured in units
such as hours. Daily maintenance would be an example.
ii) Operation-dependent failures such as tool wear, occur only while the system is running. Time between
operation failures is measured in cycles. It has been estimated that approximately 80% of transfer line
failures are operation-dependent station failures.

Performance of a line configuration is measured by its effectiveness E (availability)

q (t )
t of Q(t )
E lim
where q(t) - actual output over time t
Q(t) - theoretical capacity over t in the absence of work stoppage.
Thus t is estimated by comparing productive cycle (time) to total cycles (time)

E (uptime)
E (uptime  downtime)
E

E(.) -refers to the expected value, uptime - interval during which production is occurring and downtime -
period during which finished product is not leaving the line

3.10.2 Paced lines without buffers

3.10.2.1 Operation - Dependent failures

Considering an M-stage transfer line and assuming:
1) The number of cycles until failure of station i is assumed to be a geometric random variable t with failure
rate 7 with failure rate D i. Hence the density function for the number of cycles to failure is fi(t) = D i.
(1- D i. )t-1. Mean cycles to failures (MCTF) is accordingly D i. -1
2) The number of cycles for repair at station i is geometric with mean bi-1cycles.
3) Normally assume bi=b, i=1, ---,M
4) All uptime and downtime random variables are independent
5) Idle stations do not fail
6) Failures occur at the end of the cycle; failure do not destroy the product
7) At most one station can fail on any cycle

Let Qr be the event that all m stations survived r cycles. The probability that the line first fails at the end of
the tth cycle ( t good products are produced but the tth cannot be passed on) is given by

P(T=t) = Prob(Qt-1) *[1-Prob(Qt/Qt-1)]

(
= ®3 (1  D i ¾ «¬ 3 1  D i »
­M ½ ª º

t 1 M

¯ i' 1 ¿ ¼
1
i' 1

*
P(T=t)= «1  3(1  D i )»®1  >1 @ 3(1  D i )]¾
ª M º­ ½
M t 1

¬ i1 ¼¯ ¿
(3.1)
) )
i 1

45
 E = ª1  3+ (1  D
? º
« )»
M
(3.2)
¬ ¼
i
i 1

P(7 t) E (1  E ) t , 1

This is a geometric distribution with E as parameter. Thus M station line behaves like a single station but
with failure parameter E replacing Di

Expanding terms in E

§ ·
E = 1 - ¨1 
©
¦ D  ¦- D D
i
i
i j
i j .......
-.¦
-/- D
i j ... m
j D j  D m
¹̧
Since Di is small, higher order terms approach zero

E| ¦0 D
m

i (3.3)
i 1

Line effectiveness can now be calculated by the ratio of expected productive cycles between failures
(uptime) divided by expected total cycles between failures (uptime and downtime). Using equation 3.3 and
repair the distribution assumptions.

E 1
1 1
E 1 b 1  Eb
E= 1 1 1 1 1
(3.4)

Example
Consider a two-station line where the first station fails on the average every 10 cycles and the second station
averages a failure every 15 cycles. Average repair time is two cycles. Find line availability.

Solution:
1 1 1
Di D2 b
10 15 2

1 1
1  Eb 1  (  115)2
E= 2 1 1
0.75
10

3.10.2.2. Time-dependent failures

Failures occur continues by through time with the exception that a failed station cannot fail again until it is
repaired. Thus time to failure is measured as clock time since the station was last repaired. Completion of a
station repair initialises a renewal of the failure process. Assuming that on repair completion, time to failure
for station i is exponential with rate parameter Di. Mean time to failure is 1/DI. Repair times are exponential
with parameter bi. Total line failures have the same effect as station failures, thus using the pseudo
workstation 0 to represent the line. The expected number of failures during an operating time of t is Di t for
station i. Expected repair time is Dit/bi
Thus for station i alone, its effectiveness is

E (uptime) t 1
t of E (uptime)  E (downtime)
Ei = lim lim
D t D
t i 1 i
bi bi
In cycle-dependent failures, time was suspended when any station failed. Thus is not the case with time-
dependent failures. As stations are independent;

46
 41
¦3
ª Di º
«1  »
M

¬ bi ¼
E= 3.5
i o

Example
Consider a two-station line where the first station fails on the average every 10 cycles and the second station
averages a failure every 15 cycles. Average repair time is two cycles. Assuming that the line is subjected to
time-dependent failures. How is the effectiveness affected?

Solution
1 1 1
Di D2 and b
10 15 2
51 61
§ 2· § 2·
E = ¨1  ¨1 
© 10 ¹̧ © 15¹̧
= 0.735

Effectiveness has reduced from 0.75 to 0.735 since station 1 and 2 can now continue the ageing process
while idle as a result of a failure at the other station.

3.10.3. Two-Stage Paced Lines with Buffer

The buffer reduces the dependence between station i.e. if the buffer contains parts when station 1 fails, then
station 2 continues operating. Likewise, if station 2 fails, 1 continues operating until the buffer is full. Using
operation-dependent failures let Z be the buffer capacity.

3.10.3.1 - Operation-Dependent failures

Using a Markov chain where the states are (S1, S2, Z). Si is the status of station i and Z is the number of
items in the buffer status for station I is denoted by W for working (operational) condition and R for repair
required. Note that if station 1 is “ working” but 2 is under repair, station 1 may still be idle because of a full
buffer. If 1 is under repair and 2 is working, an empty buffer will starve station 2.

1-b1

RR0 RW1

RW0 WW1

WW0 WR1

WR0 RR1

47
Table: Transitions for Two-Stage line with Buffer
Initial State Station 1 Station 2 Probability Resultant State
Up Up (1-D1)(1-D2 ) | 1-D1-D2 WWx
WWx,0”x•Z Up Fail (1-D1)D2 | D2 WRx
fail Up D1(1-D2) | D1 RWx
RW0 Repaired Idle b1 WW0
Down Idle 1-b1 RW0
Fixed Up b1 (1-D2) | b1 WWx-1
RWx,0<x”Z Down Up (1- b1)(1-D2) | 1- b1-D2 RW x-1
Down Fail (1-b1) D2 | D2 RR x-1
Up Down (1-D1)(1- b2) | 1-D1 -b2 WR x+1
WRx, 0”x<Z Up Repaired (1-D1)b2 | b2 WW x+1
Fail Down D1(1- b2) | D1 RRx+1
WRZ Idle Down 1- b2 WRZ
Idle Repaired b2 WWZ
Down Down (1- b1)(1-b2) | 1-b1 -b2 RRx
RRx, 0”x”Z Down Fixed (1- b1) b2 | b2 RWx
Fixed Down b1(1-b2) | b1 WRx

if S is the set of states of the system the stead-state balance equations can be defined by applying the
Chapman-Kolomogorov result as

P(S1) = 6P(s)p(S,S1) (3.6)

sS

where P(s) is the probability of being in state S and p(u,v) is the transition probability for ending in state v
given that we began the cycle in state u.

To get the different results of state by applying equation 3.6 we get

P(WW0) = (1-D1-D2)P(WW0) +b1 P(RW0)+ b1P(RW1)

P(WWx) = (1-D1-D2)P(WWx) +bP(RW,x+1)+ b2P(WR, x-1) o<x<z

P(WWZ) = (1-D1-D2)P(WWZ) +b2P(RW,Z-1)+ b2P(WRZ)
P(RW0) = D1P(WW0) +(1-b1)P(RW0)+(1- b1- b2P(RW1)+ b2P(RR0)
P(RWx) = D1P(WWx)+(1- b1- D2)P(RW, x+1)+ b2P(RRx) o<x<z
P(RWz) = D1P(WWZ)+ b2P(RRZ)
P(RWz) = D1P(WWZ)+ b2P(RRZ)
P(WR0) = D2P(WW0)+ b1P(RR0)
P(WRx) = D2P(WWx)+ (1- D1 - b2 )P(WR,x-1)+ b1P(RRx) o<x<z
P(WRZ)= D2P(WWZ)+(1- D1 - b2 )P(WR, Z-1)+ (1- b2)P(WRZ)+ b1P(RRZ)
P(RR0) = D2P(RW1)+(1- b1- b2)P( RR0)
P(RRx)= D2P(RW,x+1)+ D1P(WR, x-1)+(1- b1 - b2 )P(RRx) 0<x<z
P(RRZ)= D1P(WR, z-1)+ (1- b1 - b2 )P(RRZ) ------ (3.7)

System effectiveness for a buffer of maximum size Z, z can be measured:

¦ P(WWx )  ¦7 P( RWx )
z z
Ez (3.8)
8
x o x 1

Expression (3.8) could be used regardless of whether the low-probability exerts such as two station failures
on the same cycle are excluded from the Markov chain model. However for the case of their exclusion.
Buzacott presents a closed-form expression for the effectiveness of this model

48
 Di
Let xi be the ratio of average repair time to uptime.
bi
x2 D2 (D1  D 2 )(b1  b2 )  D 1b2 (D1  D 2  b1  b2 )
(D1  D 2 )(b1  b2 )  D 2b1 (D1  D 2  b1  b2 )
Then s r and C =
x1 D1

Thus Effectiveness with a buffer of size Z is:

1  sC z
1  x1  (1  x2 ) sC z
Ez = when Sz1

1  r  b2 (1  x)  Zb2 (1  x)
(1  2 x)[1  r  b2 (1  x)]  Zb2 (1  x) 2
or when S= 1 (3.9)

The gain due to the buffer is given by G z =Ez - E0

Limiting effectiveness is given by:

1
Z of 1  x*
lim E where x* = max (x1,x2)

Meaning as the buffer capacity is increased asymptotic effectiveness approaches the capacity of the least
effective station.

Example
1 1 1
Given D 1 D2 b
10 15 2
Find the availability of the line if a buffer of four spaces is placed between the work stations. Also, find the
availability if unlimited buffer space is available between the workstations.

Solution
First compute the model parameters
D1
1
.
01 15
x1 = 0.2 x2 0133
.
b1 0.5 0.5

since x 2  x1 , station 2 has greater natural availability

since S z 1
x2 2
S=
x1 3

(D 1  D 2 )(b1  b2 )  D 1b2 (D 1  D 2  b1  b2 ) 0108335

.
(D 1  D 2 )(b1  b2 )  D 2 b1 (D 1  D 2  b1  b2 ) 0127778
C= = 0.847828
.

1  SC 4
1  x1  (1  x 2 ) SC 4
E4 0.81

Thus the addition of buffer space for four units has increased availability from 0.75 to 0.81. In every 100
cycles an average of six additional good nuts are produced maximum possible production rate is:

1 1
E9
1  max(0.2,01333
0.833
. ) 12
.

49
3.10.3.2 Deterministic Failures and Repairs
Assuming identical stations thus failures will alternate between stations, when station 1 fails, station 2 will
continue to operate from buffer until 1 is repaired or buffer empties. If Z t b-1, the cycles to repair then
station 1 worked through station 2’ s previous repair time at least enough to ensure buffer of b-1 parts.
for Z t b-1
1
1 x
Thus E Z

5
4




7LPH
1 2 2 1 1 2 2 1 1
UHSDLUHGIDLOVUHSDLUHGIDLOVUHSDLUHGIDLOVUHSDLUHGIDLOVUHSDLUHG

Figure: Buffer levels for deterministic up-and downtimes

From this example a general rule that buffers should normally be large enough to accommodate at least
average repair time production.

3.10.3.3 Systems Reduction

In normal case systems have more than two workstations so to include this a set of stations that must be
jointly active or idle can be aggregated into a single station provided they have common repair rate. This
logic holds whether the stations are all in series or act as feeder stations for main station. The aggravated
failure rate is obtained by summing the individual failure rates. Result is summarised by the following rule
Rule 1: Any set of stations I =1---m with no intermittent buffers, with failure rates D i and repair rates bi=b
can be replaced with the single station j with failure rate

D cj ¦: D
m

i and bj=b provided that all stations must stop if any individual station fails
i 1

Di D2 D3 D4 D5

D1 + D2 = D 2+ D4 + D5

a) Single line

50
 D6

D1 D2 D3 D4 D5

Di + D 2 + D3+ D6 D4+ D5

b) Feeder line (unbuffered)

D1

D3 D4 = D1
D2
D2 + D3 + D4
c) Feeder line (buffered)
Figure: Equivalent transfer lines for analysis

Example
An engineer is to decide where to place a buffer in a four-stage production process. Each stage has the same
fixed cycle time. Uptimes are unpredictable but the mean cycle time between failures is 250 for each station.
Repair time averages 10 cycles. The buffer will have a station. Repair time averages 10 cycles. The buffer
will have a capacity of 20 units since most repairs can be completed in 20 cycles

Solution
There are three possible location - Repair time parameter b=0.1 while the individual workstations have Di =
0.004.
Table: Possible Buffer location
Buffer after WS D‘1 D0 E20 E;
1 0.004 0.012 0.885 0.893
2 0.008 0.008 0.895 0.926
3 0.012 0.004 0.885 0.893
Therefore the best location to place buffer is between station 2 and 3.

Rule 2: Medium Buffer location if only one buffer is to be inserted, it should be placed in the middle of the
line.
Reference/ Bibliography
1) Groover M P (1985), Automation, Production systems, and Computer-Aided manufacturing
2) Askin R G and Standridge C R (1993) Modeling and Analysis of Manufacturing Systems, John Wiley &
Sons,
3) Sule D R, (1994), Manufacturing Facilities: Location, planning and design, PWS Publishing Company
ISBN 0-534-93435-8
4) Tompkins, J A, White, J A, Bozer, Y A, Tanchoo, J M A and Trevino, J, (1996), Facilities Planning,
John Wiley & Sons
5) McMahon C and Browne J, (1993), CADCAM: form principle to practice, Addison-Wesley

51
 APPENDIX A: FACILITY PLANNING
FACILITY PLANNING

ƒ Facility planning determines how an activity’ s tangible fixed assets best support achieving the activity’ s
objective.

ƒ In manufacturing, facility planning involves the determination of how the manufacturing facility best supports
production.

Facilities planning hierarchy

Facility Location - placement and orientation with respect to customers, suppliers, and other facilities with which it
interfaces.
Facility design components- the facility systems, the layout, & the handling system.
9 Manufacturing facility systems – the structure and envelope systems, power, light, air conditioning, water,
sewage
9 Facility Layout - the production areas, production-support areas, personnel areas within the building
9 Handling system – the materials, personnel, information equipment handling systems for supporting
production
Facilities location-the determination of how the location of the facility should support the facility’ s objective. It
addresses macro issues.
Facilities design - determination of how the design components should support achieving the facility’ s objective. It
addresses micro issues.

Plant
location

Manufacturing
facilities Facility
planning system

Plant Plant layout

design

Materials
handling

Discuss….

So, which comes first, the material handling system or the facility layout ?
Appropriate answer is, "Both!"
ƒ The layout and the handling system should be designed simultaneously.
ƒ The complexity of the design problem generally requires that a sequential process be used. For this
reason, it is recommended that a number of alternative handling systems be developed and the
appropriate layout be designed for each.

SIGNIFICANCE OF FACILITY PLANNING

Significant portion of investment goes into new facilities. Effective facilities planning therefore offers opportunities for
improvement. Consider the following:
ƒ The impact of facilities planning on handling and maintenance costs

52
 ƒ The impact of facilities planning on employee moral, and the impact of morale on operating costs
ƒ How convertible are the lots of capital investments in facilities
ƒ The impact of facilities planning on the management of the facility
ƒ The impact of facilities planning on a facility’ s capability to adapt to change and satisfy future requirements

Objectives of Facility planning:

Facility design must be central to the facility planning. Typical facilities design objectives
ƒ Support the organization' s vision through improved material handling, material control, and good
housekeeping.
ƒ Effectively utilize people, equipment, space, and energy.
ƒ Minimize capital investment.
ƒ Be adaptable and promote ease of maintenance.
ƒ Provide for employee safety and job satisfaction.

Facility planning cycle

Define/ redefine objective of the

facility

Specify/ update primary & related

activities to accomplish objectives

Determine space requirements for

all activities

Maintain, improve N ew operation No Determine

continuously or facility on facility location
existing site?

Yes
Develop alternative plans
and evaluate

Select facilities plan

Implement plan

Facilities planning process

ƒ Define (or redefine) the objective of the facility- Consider products to be produced and/or service (s) in
specific quantifiable terms
ƒ Specify primary and support activities to be performed in accomplishing the objective -specified in terms
of the operations, equipment, personnel, and material flows involved.
ƒ Determine the interrelationships among all activities-how activities interact or support one another
ƒ Determine the space requirements for all activities – for equipment, material, personnel
ƒ Generate alternative facilities plans:- alternative facilities locations and layout designs, structural designs,
and material handling system designs.
ƒ Evaluate alternative facilities plans:-Rank the plans specified.
ƒ Select a facilities plan.:- determine which plan is most acceptable
ƒ Implement the facilities plan.
ƒ Maintain and adapt the facilities plan.

53
 ƒ Redefine the objective of the facility :-As in Step 1, identify products (services) to be produced (provided)
specific quantifiable terms. Any potential modifications/expansion?

Model of Success
x presents a clear direction of where a business is headed
x -model used to best achieve the set objectives of p
ƒ The vision:-a description of where you are headed
ƒ The mission:-how to accomplish the vision
ƒ The requirements of success
ƒ The guiding principles
ƒ The evidence of success

STRATEGIC FACILITIES PLANNING

DEVELOPING FACILITIES PLANNING STRATEGIES
Facilities planning is a strategic process; an integral part of the overall corporate strategy.

In developing a facilities planning strategy, it is important to note that;

„ Team-based implementation of company objectives will ensure that all members of the organization are
involved in its achievement.
„ Strategies are needed for all functions-marketing, manufacturing, distribution, purchasing, facilities, material
handling, information systems, among others. Each functional strategy is multidimensional; each must
contribute to the strategic plan for the entire organization.
„ Effectively translate objectives into action using the model of success. It is effective due to its lateral approach
as opposed to hierarchical.

Dimensions for improving facilities planning

‰ Focusing on physical aspects of facilities planning, e.g., buildings, equipment, and people.
‰ Focusing on control aspects of facilities planning-space standards, materials control, etc
‰ Focusing on time planning -Sufficient lead time is needed to do it right!

Physical aspects

Control

Time

Impact of internal functional areas

„ Marketing decisions affect the location of facilities and the handling system design- Materials handling
affected by decisions on unit volume, product mix, packaging, delivery times.

54
 „ Product design and development decisions affect processing and materials requirements-affect layout
and material handling
„ Manufacturing decisions will have an impact on both facilities location and facilities design, e.g.,
decisions concerning the degree of vertical integration, types and levels of automation

„ Discuss impact of other functions.....human resources, finance, etc

ISSUES HAVING LONG-RANGE IMPACT ON STRATEGIC FACILITY PLAN

„ Number, location, and size of warehouse and/or distribution centres

„ Centralised/decentralized storage:-supplies, raw materials, products

„ Acquisition of existing facilities versus design of modern factories

„ Flexibility required due to market and technological uncertainties

„ Storage and manufacturing interface

„ Level of vertical integration , and ‘subcontract versus manufacture’

„ Control systems-material control, equipment control

„ Movement of material between buildings and sites

„ Changes in customers’ and suppliers’ technology
„ Design-to-cost goals for facilities

INADEQUATE PLANNING EXAMPLES

„ A textile firm installed a large high-rise AS/RS for one of its divisions. Requirements subsequently changed in
terms of the amount and size of the product to be stored. Other changes in technology were projected. The
system became obsolete before it was operational.

„ An electronics manufacturer was faced with rapid growth. Management received proposals that required
approximately equivalent funding for large warehouses at two sites having essentially the same storage and
throughput requirements. Management questioned the rationale for one "solution" being a high-rise AS/RS and
the other being a low-rise warehouse with computer-controlled industrial trucks.

Facility Design and Layout

Facility layout: Arrangement of machines, storage areas, and/or work areas usually within the confines of a
physical structure, such as a retail store, an office, a warehouse, or a manufacturing facility.

Factors that influence layout

ƒ Volume, weight of items to be produced.
ƒ Nature of the service to be provided.
ƒ Cost of the building to house the operation.
ƒ The product mix that must have a facility.
ƒ The fragility of the product or component.

55
 Flow-Line Layout

Lathe Drill Grind Drill

Press Bend Drill

Mill Drill

Lathe Lathe Drill

• Applicable to both manufacturing and non manufacturing operations.

• Arrange machines and/or workers in accordance with the sequence of operations for a given product or service.
• Advantages of flow-line layout
• Reduces materials handling.
• Accommodates small amounts of work in process.
• Reduces transit times.
• Simplifies production planning and control systems.
• Simplifies tasks, enabling unskilled workers to learn task quickly.

Flow-Line Layout disadvantages

• Lack of process flexibility.

• Lack of flexibility in timing: the product can not flow through the line faster than the slowest task can
be accomplished unless that task is performed at several stations.

• Large investments: special-purpose equipment and duplication is required to offset lack of flexibility
in timing.

• Dependence of the whole on each part: a breakdown of one machine or absence of enough operators
to staff all work stations may stop the entire line.

• Worker fatigue: workers may become bored by the endless repetition of simple tasks.

56
Process Layout
Lathe Lathe Drill Weld Weld

Lathe Lathe Drill Paint Paint

Mill Mill Grind Assemb

Mill Mill Grind Assemb

Process Layout
Grouping together of machines and/or workers doing similar tasks.
Applicable to both manufacturing and non manufacturing operations.
Advantages
– Flexibility: equipment and personnel can be used where they are needed.
– Smaller investment in equipment: duplication is not necessary unless volume is large.
– Expertise: supervisors for each department become highly. knowledgeable about their functions
– Diversity of tasks: changing work assignments make work more satisfying for people who prefer
variety.

Disadvantages
– Lack of process efficiency: backtracking and long movements may occur in the handling of materials.
– Lack of efficiency in timing: workers must wait between tasks.
– Complication of production planning and control.
– Cost: workers must have broad skills and must be paid higher wages than assembly line workers.
– Lowered productivity: because each job is different it requires different setups and operator training.

Fixed Position Layout

• Manufacturing and non-manufacturing operations of bulky or fragile products, e.g., ships and planes.

• Move machines and/or workers to the site; products normally remains in one location for its entire
manufacturing period.

• Advantages of fixed position layout

– Reduces movement of work items; minimizes damage or cost of moving.
– More continuity of the assigned work force (since the item does not go from one department to
another). This reduces the problems of re-planning and instructing people each time a new type of
activity is to begin.

• Disadvantages of fixed position layout

– Since the same workers are involved in more operations, skilled and versatile workers are required.
The necessary combination of skills may be difficult to find and high pay levels may be necessary.
– Movement of people and equipment to and from the work site may be expensive.
– Equipment utilization may be low because the equipment may be left at a location where it will be
needed again in a few days rather than moved to another location where it would be productive.

57
Group Technology Layout-GT
¾ Definition of Group Technology
“ Group technology is the technique of identifying and bringing together related or similar parts in a production
process in order to utilize the inherent economy of flow production methods.”

V. B. Solaja,Institute of Machine Tools, Belgrade, Yugoslavia

¾ Group Technology layout is also called manufacturing cell layout.

¾ Example:
z A plant producing 10,000 part numbers may be able to group the parts into 50 or 60 families. Each
family would possess similar design and manufacturing characteristics.
z Hence, the processing of each member of a given family would be similar, and this results in
manufacturing efficiencies in the form of:
• Reduced set-up,
• Lower in-process inventories,
• Better scheduling,
• Improved tool control,
• Standard process plan.

Concept-GT
¾ Many problems are similar, by grouping similar problems, a single solution can be found to a set of problems,
thus saving time and effort.
¾ A manufacturing philosophy in which similar parts are identified and grouped together to take advantage of
their similarities in design and manufacturing.
¾ A technique for identifying and bringing together related or similar components in order to take advantage of
their similarities in the production process.

Concept-GT
Concept D esign Manufacturing
Many problems are Similar Shapes Sim ilar
similar Manufacturing
Processes
G roup Similar D esign Families Production Family
Problems
Single Solution of O ne standard design O ne standard
Problems plus m inor process plan to a
m odification fam ily and
m odification and
extension

Manufacturing Cell (MC)

¾ Cellular manufacturing is the physical division of the manufacturing facilities into production cells.
z Each cell is designed to produce a part family. A part family is a set of parts that require similar
machinery, tooling, machine operations, and/or jigs and fixtures.
z The parts within the family normally go from raw material to finished parts within a single cell.
¾ Justification
z Batch production
• Only about 5% of the lead time in producing a part is direct working time; 95% of the
time, the part is waiting.

58
 • The 5% direct working time includes 30% actual processing and 70% for positioning,
chucking, gauging, etc.
• Hence, only 1.5% is accounted for actual machining.
z Cellular manufacturing directs its effort towards the remaining 98.5% by organizing the plant
layout according to work cell, rather than functions. A work cell is a unit that includes all of the
machines required to produce a family of parts.

¾ Advantages
z Reduced material handling; reduced interdepartmental roundabout movements
z Reduced set-up time, production time
z Reduced tooling, pallets, jigs & fixtures
z Reduced in-process inventory
z Easy identification of bottlenecks
z Increase operator expertise
z Improved human relations, team-work spirit, & morale.
z Implied reduction of necessary control
z Better output quality, reduced scrap, increased accountability of operators and supervisors

¾ Disadvantages
z Reduced shop flexibility (compare with job shop)
z Possible reduced machine utilization
z Possible extended job flow times
z Possible increased job tardiness.
¾ Implementation Issues
z Reorganization - machine layout need reorganization every so often.
z Work cell supervision - supervisors must be expert in several field (e.g., milling, turning, grinding,
etc.) represented in the cell.
z Shop floor control / production planning - cell concept might lead to unbalanced workload on
machines.

Discussion-current layout
¾ Discuss basic layout types in your company

¾ Disadvantages/problems faced with the layout

¾ Any improvements/re-layout/suggestions for the layout ?

59