a .. .
3.3%

_.. 7.4.

it w .
. kw , 28m...

émnfi

r
_

. . .. , mu...
. ‘lm. . . . . . WWW. ..«WmW. to.
"Q; hr . . 1.5%};
it v
.. 1“...”

4h“...

hmfu/i
x 4 .. 2
1530...?

i... am

$1....
:5”...
i.

a.

$.22;

.. .e?...»............
‘05. a I

.4...
15*.

:1. 1

OX.
.3.
I... 31....- ..
gang}... _
.ia. X lid A;
“Firm .“W

«.1. “an. .3

.rlnnnvlnfiamk: xx.
m‘ ’ Agar-IQ-
.fi 5 ‘0 ‘.
31“}
.321...va
fig“...
19."
....
s.)
, .2» .
x. ..
.. xufiwmu
.5»?
I. .r‘.m.h:t=t .35 J k .
...r.......:. 12.... I

 

 

EM -
#555315
5- ..

w

1. .

. a u}

_ a . . g“: r
.

my...
D I: . . . . .

v

1.

 

'NHP°6

1.
2,003

J'L??‘%.36'7 2

This is to certify that the

thesis entitled

"Performance Assessment of Planning Processes During
Manufactured Housing Production Operations Using Lean
Production Principles."

presented by

Vijaykrupal Reddy Chitla

has been accepted towards fulfillment
of the requirements for

M.S. degree in Building Construction
Management

/%.C/.//.M

/ Major professor

 

 

WW8

 

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

. LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDuest-p. 1 5

 

PERFORMANCE ASSESSMENT OF PLANNING PROCESSES DURING
MANUFACTURED HOUSING PRODUCTION OPERATIONS USING LEAN
PRODUCTION PRINCIPLES
By

Vijaykrupal Reddy Chitla

A Thesis
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Construction Management

2003

ABSTRACT
PERFORMANCE ASSESSMENT OF PLANNING PROCESSES DURING
MANUFACTURED HOUSING PRODUCTION OPERATIONS USING LEAN
PRODUCTION PRINCIPLES
By
Vijaykrupal Reddy Chitla
From the time of trailers of pre-World War 11, used as recreational vehicles, to today’s
urban alternative for affordable housing, Manufactured Homes have come a long way.
Built completely on an assembly line in a factory, 21.4 million Americans have chosen
these Manufactured Homes as a housing option, in 1999. With a similar production style
like the car manufacturing, Manufactured Housing industry has been building homes,
which have improved in quality, size, and price. MH industry should undertake a critical

examination of its production planning processes to identify opportunities for decreasing

production costs while improving performance.

The goal of this research is to evaluate the plant production planning process in the
manufacturing housing industry and to identify opportunities for improvement. Lean
construction techniques, inspired and adopted from Lean Manufacturing principles of
automobile industry are utilized to achieve the intended goal. Using Last Planner’s
Percent Plan Complete, a production control tool of Lean Construction, and Labor
Utilization Factor results, using productivity ratings, 3 method to evaluate the
performance of production planning process in a Manufactured Housing plant is
developed. Techniques like Pareto Analysis and Fish Bone diagrams were used to

identify process improvement opportunities, and provide guidelines to the industry.

Dedicated to My Wonderful Parents

iii

ACKNOWLEDGEMENTS

I am thankful to Dr. Tariq Abdelhamid for his support and advice during the last
two years. I am also thanka to Dr. Matt Syal and Dr. Shawnee Vickery for their help

during this research.

I am very thankful to my wonderful parents, my dear sister and brother, for all

their love and support and for being there. I also thank the support from the personnel of

the plants used in this research.

Finally, I am thankful to my dear friend Namita Mehrotra, for her valuable

support and help during the last two years.

iv

TABLE OF CONTENTS

LIST OF TABLES ................................................................................. x
LIST OF FIGURES ................................................................................ xi
CHAPTER I — INTRODUCTION .............................................................. 1
1.1 Motivation. .............................................................................. 2

1.2. Need Statement ......................................................................... 4
1.3 Goal and Objectives. .................................................................... 7

1.4 Report Overview .......................................................... 8
CHAPTER II - LITERATURE REVIEW ..................................................... 9
2.1 Introduction ............................................................................ 10
2.2 Manufactured Housing ............................................................... 10

2.2.1 Previous Research ......................................................... 11

2.2.2 Definitions ................................................................... 12

2.2.1.1 Factory-Built Housing .......................................... 13

2.2.1.2 Site-Built Housing ............................................. 13

2.2.1.3 Manufactured House ........................................... 13

2.2.1.4 Single-Section House .......................................... 13

2.2.1.5 Multi-Section House ........................................... 14

2.2.1.6 Modular House .................................................. 14

2.2.1.7 Panelized House ................................................ 14

2.2.1.8 Pre-cut House ................................................... 14

2.2.1.9 Mobile House ................................................... 15

2.2.1.10HUDCode ...................................................... 15

2.2.3 Production Process ......................................................... 15

2.3 Background Of Lean Production ................................................... 18
2.3.1 Origin Of Lean Production ............................................... 18

2.3.2 Types Of Production Processes .......................................... 19

2.3.3 Construction From A Lean View ........................................ 22

2.3.4 Lean Construction ......................................................... 23

2.3.4.1 Lean Project Delivery System ................................. 26

2.3.4.2 Lean Construction Principles ................................. 29

2.3.5 Research In Lean Construction .......................................... 30

2.3.6 Traditional Project Control ............................................... 31

2.3.6.1 Production Control ............................................. 33

2.3.7 Last Planner ................................................................ 36

2.4 Summary ............................................................................... 40
CHAPTER III — METHODS ............................................................................. 41
3.1 Introduction ............................................................................ 42
3.2 Methods And Tools For Objectives ................................................ 43
3.3 Objective 1 ............................................................................. 43
3.3.1 Understanding Production Process in Manufactured Housing ...... 44

3.3.2 Documenting Production Planning process ............................ 46

3.3.2.1 Data on Production Planning Process ........................ 46

3.4 OBJECTIVE 2 ........................................................................ 48

vi

3.4.1 Last Planner ................................................................ 48

3.4.2 Work Sampling ............................................................ 50

3.4.3 Productivity Ratings ....................................................... 50

3.4.3.1 Sample Size ...................................................... 51

3.4.3.2 Classification of work .......................................... 53

3.4.3.3 Labor-Utilization Factor ....................................... 55

3.5 Objective 3 ............................................................................. 57
3.5.1 Constraint Analysis ........................................................ 57

3.5.2 Pareto Analysis ............................................................ 58

3.5.3 Cause-and-Effect Diagrams .............................................. 59

3.6 Production Planning Process In Manufactured Housing Production Plants ...61

3.6.1 Brief About The Production Plants ...................................... 61

3.6.2 Production Process ......................................................... 62

3.6.2.1 Macro level planning .......................................... 62

3.6.2.2 Problems in the production process planning ............... 64

3.7 Summary ............................................................................... 65
CHAPTER IV - DATA COLLECTION AND ANALYSIS .............................. 66
4.1 Introduction ............................................................................ 67
4.2 Data Collection ......................................................................... 67
4.3 Production Planning Process .......................................................... 68
4.3.1 Manufactured Housing Production Plant ................................ 68

4.3.1.1 Manufacturing Codes and Standards ........................ 69

4.3.1.2 Size and Capacity ................................................ 71

vii

4.3.1.3 Manufactured Housing Plant Assembly Line and
Stations ..................................................................... 72

4.3.2 Organizational Chart in a Manufactured Housing Plant

Production .......................................................................... 73
4.3.2.1 Hierarchical Planning .............. 75
4.3.2.2 Production Planning System and Strategy .................. 76
4.4 Production Planning Operations at a Manufactured Housing Production
Plant .......................................................................................... 79
4.4.1 Cluster Operations ......................................................... 81
4.4.2 Daily and Weekly Operations ............................................. 82
4.5 Summary ................................................................................ 84
4.6 Percent Plan Complete ................................................................ 85
4.6.1 Production Planning ........................................................ 85
4.6.2 Production Planning at Two Plants ....................................... 86
4.6.3 Percent Plan Complete Calculations ..................................... 88
4.6.4 PPC Data and Calculations ................................................ 90
4.7 Summary of PPC ...................................................................... 99
4.8 Labor Utilization Factor ............................................................. 101
4.8.1 Labor Force ................................................................ 102
4.8.2 Production Rating Study ................................................. 103
4.8.3 Labor Utilization Factor Data ........................................... 105
4.9 Relation between LUF and PPC ................................................... 107
4.10 Coefficient Determination ......................................................... 109
4.11 Summary ............................................................................. 113

viii

CHAPTER V — DESCRIPTION AND CONCLUSION ................................. 114

5.1 Introduction ........................................................................... 115

5.2 Constraint Analysis .................................................................. 115

5.2.1 Pareto Analysis ............................................................ 118

5.2.2 Fish Bone Technique ..................................................... 119

5.3 Results And Contributions ......................................................... 123

5.4 Recommendations and Guidelines for the Industry .............................. 124

5.5 Limitations and Future Areas Of Research ....................................... 126

5.6 Summary .............................................................................. 127
APPENDICES .................................................................................... 128
REFERENCES .................................................................................... 138

ix

Table 1.1

Table 2.1

Table 3.2

Table 3.3

Table 4.1

Table 4.2

Table 4.3

Table 4.4

Table 4.5

Table 4.6

& PPC ......

Table 5.1

Table 5.2

LIST OF TABLES

Manufactured Houses Sales Price ................................................ 3
Theory of production ............................................................. 25
Sample format of PPC data collection .......................................... 49
Productivity ratings for several construction trades .......................... 54
Manufactured housing stations and activities at each station ............... 90
PPC of roof truss building assignment/station crew .......................... 93
PPC of Internal walls assignment/station crew ................................ 95
PPC for External Walls assignment/station crew ............................. 96

AM and PM LUF of individual crewmembers at the observed stations. 106

PPC & LUF data of a single day for observing Correlation between LUF

.......................................................................................... 109
Percent of non—completed assignments ....................................... 1 16
Problems in production operations of a MH production plant ............ 118

Figure 1.1
Figure 1.2
Figure 2.1
Figure 2.2
Figure 2.3
Figure 2.4
Figure 2.5
Figure 2.6
Figure 3.1
Figure 3.2
Figure 3.3
Figure 3.4
Figure 3.5
Figure 3.6
Figure 4.1

Figure 4.2

Plant .........

Figure 4.3
Figure 4.4
Figure 4.5

Figure 4.6

LIST OF FIGURES

Recent Home Buyers by Age ...................................................... 5
Expected demand for Manufactured Houses in different age groups ........ 6

Production Process Layout of a Typical Manufactured Housing Plant. . ..17

Methodologies Based on Concepts and Principles ........................... 24
Lean Project Delivery System Model .......................................... 26
Traditional Planning and Control System. .................................... 35

Last Planner Planning process ................................................... 36
Construction Planning vs. Last Planner ........................................ 38
Example of production operations in MH ..................................... 44
Activity Definition Model for Roof Insulation activity ...................... 45
Comparison of Percent Plan Complete vs. Labor-utilization factor ....... 56
General Format for a Pareto Chart .............................................. 59
Cause-and-Effect Diagram (F ishbone Diagram) .............................. 60
Organization Chart for Plant A .................................................. 63
Manufactured Housing Production Plant Setup ............................... 69

Organization Hierarchies in a Manufactured Housing Production

.......................................................................................... 74
Hierarchical Planning Process ................................................... 76
Required inputs to the Production Planning System ......................... 77
ADM diagram representation for Roof Shingles Station .................... 83
Production Planning Strategies at Plant A and B ............................. 87

xi

Figure 4.7
Figure 4.8
Figure 4.9
Figure 4.10
Figure 4.11

Figure 4.12

CI’CWS .........

Figure 4.13
Figure 4.14
Figure 4.15
Figure 4.16

Figure 5.1

Percent plan complete for roof truss station crew ............................. 94
PPC for Internal walls station ................................................... 95
Percent plan complete of exterior boards assignment/station crew. . . . . ....97
Percent plan complete of doors and windows installation station crew. . .97
PPC for exterior siding and external finish assignment/station crews. . ....98

PPC for paint ceiling, roof board, and shingles assignment/station

......................................................................................... 99
Average crew PPC of all observed stations .................................. 100
Average LUF of all observed station crews .................................. 107
LUF Vs. PPC for all the observed stations ................................... 108
Scattered Plot of LUF vs. PPC for all the Observed stations ............... 110

Pareto analysis chart depicting the factors responsible for problems. . ...1 19

xii

Chapter One

Introduction

I. INTRODUCTION

1.1 Motivation

Every man’s basic necessities are food, shelter, and clothing. Shelter apart, both food and
clothing are very much affordable and man can acquire these with some effort and work.
The third basic need of having a place to live has come a long way from caves to modern
luxury houses. Not every one can afford such houses, but to fulfill a common man’s

dream, the industry has come up with manufactured houses.

Manufactured houses have come a long way from its earlier form of trailer
coaches of pre-world war 11. These trailer coaches were built to serve the American
traveler. In the 19308 and 19403, mobile homes were viewed as recreational housing and
as such very highly mobile. Though the original intention of manufacturers was to
provide temporary and recreational housing, many units were used as permanent houses.
Housing shortages after World War 11 increased the use of mobile homes as permanent

housing.

Mobile homes had a very high demand in the 19605 and 19705. These mobile
homes then transformed into manufactured houses in 1974, after Congress passed the
National Mobile Home Construction and Safety Standards Act, also known as the HUD

Code.

The HUD Code came into effect in 1976 and it was designed to more clearly

define mobile homes as buildings, rather than vehicles. The houses were being

manufactured in bigger spans, growing from 8 to 16 feet. Soon double wide and multi-
section houses were being manufactured along with single section houses, and the
manufactured housing industry has grown to a booming industry. Sales have increased
along with the size and production of manufactured houses. According to the
Manufactured Housing Institute, “In the year 1999, 21.4 million Americans (about 7.6

percent of the US population) lived full time in 8.9 million manufactured homes” (MHI

2000).

 

Manf.

1993 1994 1995 1996 1997 1998 1999
Houses

 

Avg.

Sales $30,500 $32,900 $35,400 $37,400 $40,000 $41,900 $43,600
Price

 

Avg. Sq.

1,295 1,335 1,360 1,385 1,420 1,455 1,480
Footage

 

Cost/Sq
foot
Site
Built 1993 1994 1995 1996 1997 1998 1999
Houses
Avg.
Sales $147,700 $154,100 $158,700 $166,400 $176,200 $181,900 $195,800
Price
Land
Price

$23.55 $24.64 $26.03 $27.00 $28.17 $28.80 $29.46

 

 

 

$36,925 $38,525 $34,574 $35,250 $37,750 $39,775 $42,375

 

Price of

Structure $110,775 $115,575 $124,125 $131,150 $138,450 $142,125 $153,425

 

Avg. Sq.
Footage

Cost/Sq
foot

2,095 2,1 15 2,050 2,090 2,140 2,170 2,230

 

$52.88 $54.65 $60.55 $62.75 $64.70 $65.50 $68.80

 

 

 

 

 

 

 

 

 

 

Table 1.1: Manufactured Houses Sales Price
(Source: MHI Fast Facts, 1999-2001)

With the increase in demand for housing over the years and the cost of site-built

housing relatively rising, manufactured houses have come out to be an affordable option

to site-built housing. The average cost per square foot of manufactured houses was
$29.46 based on data from 1999, as shown in Table 1.1. In comparison, the average cost
per square foot of a site built house, excluding land cost, is $68.80 (Manufactured
Housing Institute, 2000). Hence, manufactured housing (MH) costs 57% less than site-
built housing. Information from Table 1.1 also indicates an increase of about $6 (25%) in
the cost per square foot from the year 1993 to 1999 for manufactured houses as compared

to approximately $16 (30%) increase for site-built over the same period.

So not only the Site-built houses are more expensive as compared to manufactured
houses; the increase in the overall cost of this form of housing is higher. Data in these

figures cast little doubt that MH are more affordable compared to site-built.

Due to its affordability, manufactured housing is now one of the nation’s fastest
growing housing segments (Perkins 1999). With the past trends and future predicted
trends, it is clear that manufactured housing is an affordable option for the households in

America.

1.2. Need Statement

Increasing household rates in America help in growth of the housing market, including
manufactured housing. The household growth is expected to continue steadily and will
increase more rapidly by the year 2010 (Vermeer and Louie 1997). The baby-boom
generation, which is expected to move into prime household formation years, will be a

prime market for manufactured housing.

Over the years, manufactured houses have improved both in quality, size and
price. Figure 1.1 shows that manufactured housing is a desirable and attractive product to
homebuyers. Figure 1.2 shows the expected demand for Manufactured Housing by the
year 2010 in different age groups. From Figure 1.2, it can be seen that the age group of
35-64 years would be a primary market for the manufactured housing industry. There is
also an increasing demand in other age groups.

A;ManufaCtured Houses H
I Other Units) I i

l
’.k_J‘

A? "i‘ 1

 

No. of Units

    

<25 25-34 35-44 45-54 55-64 65-74
I Age

Figure 1.1 Recent Home Buyers by Age
(Source: Vermeer, and Louie. Jan 1997)

j
1
l
1
1
1
1
1

 

 

 

 

  

   

   

 

1

} 25 ——1 1
1 l
. 7,: ___ _ ‘
1: E 20 5115551
1; E15 3I1995H
1 g- [-1 1D200011
12 £10 ~——~ — 151200511
1 732910;?
1 5 .

 

 

 

 

'15-24 25-34 '7 3544’ 45-54T 55-64
1 Age Group

Figure 1.2: Expected demand for Manufactured Houses in different age groups
(Source: Vermeer, and Louie. Jan 1997)

In general the demand for manufactured houses is expected to increase in most of
the age groups. If these future projections are accurate, the manufactured housing
industry, with its current production capacity, may be able to meet the increasing
demands. However, actual future demands may fall below or above current projections.
In the case of over projections, manufactured housing plants production capacity will
exceed demands resulting in cost increase to offset the imbalance. In the case of under-
projections, manufactured housing plants will not be able to meet the demand. In either
scenario, the manufactured housing production companies stand to loose. Clearly, the
MH industry should undertake a critical examination of its production planning
processes to identify opportunities for decreasing production costs while improving

performance.

The cost for a Manufactured Housing plant consists of both fixed and variable
costs. Fixed cost is the cost of the production plant, administrative offices, and material
warehouses, whereas variable cost is the cost of production of a unit of manufactured
house, which includes labor, material, and equipment. This research focuses on ways to

reduce the variable cost and improve production and planning performance.

1.3 Goal and Objectives

Previous research efforts on manufactured housing plants production operations have
focused on removing inefficiencies and variability in performance using process mapping
and operations simulation (Senghore 2001). Another way to investigate manufactured
housing plants production operations is to determine the effectiveness with which

operations are executed.

The goal of this research is to evaluate the plant production planning process in
the manufacturing housing industry and to identify opportunities for improvement. To

reach this goal, the following objectives are undertaken:

1. Documenting the currently utilized production planning process in a
Manufactured Housing plant.

2. Assessing and quantifying the production planning process performance using
lean construction principles.

3. Identifying causes of off-target performance through constraint analysis
techniques and suggesting opportunities for production planning process

improvement.

1.4 Report Overview

This thesis report is divided into five chapters. In Chapter I, an overview of the evolution
of manufactured houses was introduced. Concerns to the manufactured housing industry,
stemming mainly from increased demands, were discussed and a problem statement

formulated. The research goal and objectives were also presented.

In Chapter II, a background of lean production, production processes, and lean
construction will be discussed. This chapter will also look at traditional project and
production control and concludes with a description of the last planner technique.
Chapter III will outline the methods and tools that will be used to achieve the set goal in
this chapter. Detailed methodology and approach for each objective will be outlined.
Chapter IV reports on the data collection process and analysis using the methods and
tools described in chapter 111. Chapter V presents the findings and conclusions of the

research.

CHAPTER TWO

LITERATURE REVIEW

2.1 INTRODUCTION

This chapter details the related literature used for this research. The literature review is
used to brief the reader about the manufactured housing industry, lean production, and
lean construction. The manufactured housing industry terms and production procedures
are explained first. Lean production and lean construction concepts, and the last planner

technique are explained next.

2.2 MANUFACTURED HOUSING

Manufactured housing is a relatively new industry in terms of its presence in the market
and in the world of academic research. Evolving from a trailer industry, manufactured
houses have not been recognized in the construction industry until the 1970s. The
creation of the HUD code in 1976 has catapulted the mobile homes industry and its
image in the eyes of the public and soon the name Manufactured Housing emerged. To
this date manufactured housing struggles to make its own mark and place in the

construction industry because of its unconventional method of building houses.

The academic world is making an effort to play its role in helping the
manufactured housing industry both in terms of production improvement and as well as

overall improvement of the industry.

10

2.2.1 PREVIOUS RESEARCH

Manufactured housing has been explored by a handful of researchers. Previous research
efforts in the manufactured housing industry have focused on finding ways to improve

productivity, zoning and regulatory issues, and other aspects of the production process.

One important source on the subject is a master’s thesis submitted by Senghore
(2001) in the Construction Management program at Michigan State University. His
research titled “Production and Material Flow Process Model for Manufactured
Housing” outlines a detailed study of the production process in a manufactured housing
plant. The research developed a production process model for the plant and extensive
data collection on productivity in the production process. This data was used to simulate
the production process at certain stations along the assembly line. The goal of the
research was to develop a tool to improve the production process by identifying the
bottlenecks in the process. The work focuses on improving the production process based
on the simulation results and by recommending scenarios to remove the bottlenecks. This
work serves as a reference for manufactured housing literature on production process and

production flow.

The data from Senghore’s (2001) work will be used in later chapters for a
comparative analysis with the results from data collected in this research. To do this, data
in this research work will be collected from the same stations as in Senghore’s (2001)

work.

11

 

Another important research work in the area of manufactured housing is that by
Mehrotra (2002) at Michigan State University. Her work titled “Facilities Design Process
of a Manufactured Housing Production Plant” looks at the spatial aspects of the
assembly layout in a manufactured housing plant. The work considered the production
process details to define the spatial and proximity-based requirements between stations,
and also among the subassembly stations and feeder stations. Factory PLAN, plant layout
software was used to create different layout patterns and to arrive at alterative layouts

based on the requirements and specific production process.

No prior research could be found that considers the planning of operations in a
manufactured housing plant. Moreover, being closer to a manufacturing facility than a
construction site, it is apt to look at the production planning aspects from a lean
production perspective, as lean production has evolved from the manufacturing of

vehicles.

The following sections present brief literature on definitions used in industry and
in the academic literature of manufactured housing, followed by a short description of the

production process.

2.2.2 DEFINITIONS

A good source of manufactured housing terminology is the Manufactured Housing
Institute (MHI). MHI is a nonprofit national trade association representing all segments
of the manufactured housing industry, including manufactured home producers; retailers;

developers; community owners and managers; suppliers; insurers and financial service

12

providers. The following definitions are introduced by MHI to help potential customers

have a better understanding of terms used in the industry.

2.2.2.1 Factory-Built Housing

A house built partially or totally in a controlled environment like a factory. Many types of
structures are built in the factory and designed for long-term residential use. There are
different types of factory-built housing namely, manufactured houses, modular houses,

panelized houses, pre-cut houses, and mobile houses.

2.2.2.2 Site-Built Housing

Site-built housing is the traditional method of constructing a house where the entire house

is built on the home site.

2.2.2.3 Manufactured House

“These are homes built entirely in the factory under a federal building code administered
by the US. Department of Housing and Urban Development (HUD). The Federal
Manufactured Home Construction and Safety Standards (commonly known as the HUD
Code) went into effect June 15, 1976. Manufactured homes may be single or multi-
section and are transported to the site and installed. On—site additions, such as garages,
decks and porches, often add to the attractiveness of manufactured homes and must be

built to local, state or regional building codes” (Manufactured Housing Institute).

2.2.2.4 Single-Section House

13

A manufactured house built on a permanent chassis in single section. The width of the
section is regulated by the width of the road on which it is to be transported. The usual
width of the single-section house ranges from 8 feet to 16 feet. Average area of single-

section homes is 1,130 square feet.

2.2.2.5 Multi-Section House

A manufactured house built and delivered to the site in two or three single sections. The
average square footage is 1,640 square feet, with houses manufactured up to 2,400 square
feet. Additional attachments like garages are built on the site where installation will take

place.

2.2.2.6 Modular House

Modular houses are built in a factory under ideal conditions, structurally engineered to be
built in sections (modules). Modules are transported to the site and installed. These are

built to the state, local or regional code where the home will be located.

2.2.2.7 Panelized House

These are factory-built homes in which panels - a whole wall with windows, doors,
wiring and outside siding - are transported to the site and assembled. The homes must

meet state or local building codes where they are sited.

2.2.2.8 Pre-cut House

14

This is the name for factory-built housing in which building materials are factory-cut to
design specifications, transported to the site and assembled. Pre-cut homes include kit,
log and dome homes. These homes must meet local, state or regional building codes

(MHI).

2.2.2.9 Mobile House

This is the term used for factory-built homes produced prior to June 15, 1976, when the
HUD Code went into effect. By 1970, these homes were built to voluntary industry

standards that were eventually enforced by 45 of the 48 contiguous states (MHI).

2.2.2.10 HUD Code

Federal construction standards for manufactured housing, enforced by the Department of
Housing and Urban Development (HUD). Commonly known as HUD code, it regulates
the design and construction, and sets stringent performance standards for the utilities of a
manufactured house. The federal standards regulate manufactured housing design and
construction, strength and durability, transportability, fire resistance, energy efficiency
and quality. The HUD Code also sets performance standards for the heating, plumbing,
air-conditioning, thermal and electrical systems. It is the only federally regulated national

building code.

2.2.3 PRODUCTION PROCESS

Manufactured housing is an exceptional sector of the construction industry. It is a

different kind of manufacturing and less sophisticated in comparison to other

15

manufacturing industries. Though large numbers of sections are produced each day in a
manufactured housing factory, some customization is required depending on customer

needs.

The manufactured housing industry as mentioned earlier is a different kind of
‘construction’ or rather manufacturing of houses. Unlike site-built construction, the
manufactured house is built in a factory on an assembly line, as shown in Figure 2.1,

where house sections move from one station to another with work done on each station.

A typical manufactured housing plant assembly line will have 12-18 main stations
depending upon the complexity of houses the plant manufactures. Most of the production
takes place at these stations. All manufactured houses are built on a steel base frame

called chassis.

Some parts of the house are pre-assembled at sub-assembly stations located along
the main assembly line and adjacent to the main stations. The sub-assembly stations
fabricate and manufacture major parts of the house, like the roof truss assembly, interior
walls, cabinets, etc. Workers are assigned to each station to work on a specific job.
Feeder stations either along the assembly line or at some fixed positions supply the

necessary material for the work to the main stations and the sub-assembly stations.

An example of a main station is the roof truss station, which has a fixed crew,
feeder stations, and subassembly station in its proximity. The major activities at this main

station are placing the pre-assembled roof truss, roof insulation, and at the same time,

16

work takes place in the lower part of the house such as installing doors and windows,

exterior boards, and siding.

    
 

  
   

2 3a Exterior &
F100" Interior DOUBLE Marriage
Decking Walls STATION Walls, Blee-

& Mech.

Chassis
Frame
Sub floor

     

Frame &
Material

Supply

  

 

 

 

 

 

5 Roof
DOUBLE Assembly
STATION Installation

      
 
     

5a
Feeder
Station
Roof
Assembly
Prep

 

6 Roof
DOUBLE Shingles
STATION Placement

 

 

 

    
 
  

    
   
     
    
   

   
 
  

7
DOUBLE
STATION

Exterior
Finishes
Including
Doors &

 

 

 

  
  
    

    
    
   
 

    
  
  

    
    
 

    

l 1
Testing &
Inspection

10
Fixtures
Appliance

9
Interior
Finish

8
_DOUBLE
STATION

.1...-‘...‘.‘. . I 2 ~ ,_ ' I ..1.....~..-i...t.-.-..'.( 4.42.”... 1:33;,
i- . .' ., -1 .. . _ ' . I . 1‘5? '1 4‘ ‘ 1.1.35.4,
35.11.. . I 1 .. -. . 1:... -a .1.;-at»... Liam

Figure 2.1 Production Process Layout of a Typical Manufactured Housing Plant
(Modified from: Senghore, 2001)

Electrical
Mech.

    
 

Finished
Product

    

 

 

 

 

Completed units are shipped to Site for installation on previously prepared site

foundations. The final assembly of the units and additional activities like landscaping,

17

fencing and garage units take place at the site. This form of manufacturing houses

provides an affordable Option with a growing market.

The previous sections provide a brief overview of manufactured housing, terms
and definitions used, and plant production processes. More details can be found in
Senghore (2001) and Mehrotra (2002). The remaining sections of Chapter II will present
lean production, lean construction, and then give a background on project control in

construction and manufactured housing.
2.3 BACKGROUND OF LEAN PRODUCTION

The second part of this literature review introduces the lean production concept in detail,
starting with its origin, different types of production processes, lean construction,
principles of lean construction, traditional project control, lean production control, and
finally the last planner system of production control as conceived under lean

construction.
2.3.1 Origin of Lean Production

From the early days of manufacturing, dating back to the early 19008, there has been a
persistent effort aimed at improving production. The industrial revolution in the late 19th
century coupled with new techniques of manufacturing fueled these efforts. More
recently, a number of new approaches to production management have emerged. Just In
Time (JIT), Total Quality Management (TQM), value based management, process
reengineering, world-class manufacturing, and concurrent engineering are some of them

(Koskela, 1993).

18

When analyzed closely, all of the above management approaches seem to have a
common core, but perspectives vary. For instance, JIT stresses the elimination of wait
times, whereas TQM aims at the removal of non-value adding activities; however both

try to improve the flow of work, material and information (Koskela, 1993).

Generalizing these approaches led to a new production philosophy called lean
production, which is modeled after the successful Japanese automobile manufacturer
Toyota. The concept of Lean was developed by Toyota, led by engineer Ohno, to cut

waste and improve efficiency.

The three objectives observed by Toyota engineer Ohno for a lean production

system are (Kaufman Consulting Group, LLC, 2000):

1. Delivery of product that meets the requirement of customer
2. Produce with zero waste

3. Maintain minimum inventory

American researchers studying the production practice of Toyota coined the term
“lean”. Lean identifies customer value, understanding how it is delivered, applying pull
systems (systems release material or information into a system based on the state of the
system), and assures system transparency and zero waste. Reorganizing work to cut costs

is at the heart of lean production.

19

2.3.2 Types Of Production Processes

The world of manufacturing/production has seen many types of production processes
over the past two centuries. Two of the successful kinds of production techniques were
craft and mass production. The 19th century and early 20th century have witnessed craft

production and then came Henry Ford’s Mass Production after the industrial revolution.

Lean production is the newest type of production. Lean production has also
evolved in the car manufacturing industry and lean applications have shown tremendous
results in the manufacturing sector, revolutionizing the production systems and forcing
manufacturers to reassess their current production systems. This innovative production
system is better understood in contrast with the other two kinds of productions (Craft and

Mass production) that human beings have devised to make things.

Craft production uses highly skilled workers and simple but flexible tools to make
exactly what the customer ask for—one item at a time. For example exotic sports cars
and custom fumiture provide current day examples. Anyone would like the idea of craft
production, but the problem with it is obvious: Goods produced by the craft method cost
too much for most of us to afford. So, mass production was developed at the beginning of

the twentieth century as an alternative (Womack 1990).

Mass-production uses narrowly skilled professionals to design products to be
mentioned using expensive single-purpose machines. Due to the high costs of machinery,

mass production designs are standardized such that higher volumes are produced at lower

20

costs but at the expense of variety. This means of producing goods is monotonous and

uninspiring to workers.

Lean production combines the best features of craft production (high-quality,
individualized, custom-made products) and mass production (manufacttuing at great
quantities to satisfy broad consumer needs at lower prices). Lean uses multi-skilled
workers at all levels of the organization. Flexible and increasingly automated machines

are used to produce volumes of products with reasonable variety.

Lean principles have been applied successfully worldwide in the automobile
industry. Manufacturers like Toyota have strived to work towards the ideal, which is
100% value added work with zero (or minimum) waste. These lean principles are being
increasingly employed in many other industrial sectors with a lot of success. The best

results can be obtained when used in a repetitive or continuous production environment.

A general characteristic of most industries is the drive to improve performance for
competitive advantage (Porter 1985). Performance can be measured against value, i.e.,
meeting the customer’s needs. Improving performance has two key components: doing it
more effectively and efficiently (Horman & Kenley 1995). Effectiveness refers to
maximizing value of the output, whereas efficiency refers to minimizing or eliminating
non value-adding items in production. Performance is therefore described in terms of
attaining value effectively and efficiently. Lean principles strive for the same kind of
performance. Using lean production techniques most industries would be able to tailor

their processes to achieve this performance and meet the unique customer requirements.

21

 

As mentioned above, best results of lean production process can be obtained when
used in a repetitive or continuous production environment like manufactured housing
production. Principles of lean production have been applied in site-built construction and
have shown good results. These principles when applied in a controlled environment like
the manufactured housing plant would give tremendous results and improve the

performance.

In recent years, a handful of construction companies have embarked on a lean
conversion path. Guided by research efforts to tailor concepts of lean production to fit the
construction industry, these companies are observing good returns on their investments,

specifically in the areas of waste reduction in on-site production activities.

2.3.3 Construction from a Lean View

The construction industry is unique for its on-site production environment, and temporary
multi-organization for each project. Projects in the construction industry vary
considerably in terms of sector they serve. In addition, projects can also be characterized
as slow and quick, simple and complex, and certain and uncertain (Ballard and Howell,

1998).

For the small, simple and certain kind of projects a manufacturing strategy is
appropriate; i.e., making construction more like manufacturing through initiatives such as
standardization. However, for dynamic and uncertain projects, a manufacturing strategy

is insufficient. The uncertain, complex, and quick nature of construction projects must be

22

handled within the characteristic construction conditions of on-site production, and

temporary organization (Ballard and Howell, 1998).

Site-built construction is a combination of fabrication and assembly.
Industrialization should help to simplify site construction to final assembly and testing.
This would help shifi much of the work to the shop conditions where it can be done more
efficiently. The final assembly, which is always done on site, can vary with the stage of

development and the facility being assembled.

The following section looks at the research efforts in the construction industry

involving lean applications.

2.3.4 Lean Construction

Lean construction is in its formative stages as a new project management philosophy to
handle construction projects. According to Koskela, and general consensus in the
literature, methods/tools for designing, planning, controlling, and analyzing are based on
principles that are in turn based on concepts. Koskela used Figure 2.2 to depict the

relation between concepts, principles, and methodologies/tools.

Koskela states that the production concepts used in various industries are of three

main types (Koskela 2000):

1. Transformation view — Concept of transforming inputs to outputs.

2. Flow view — Materials and information flow in a production process.

23

3. Value generation view — Process where the value for customer is created

through fulfillment of his/her requirements.

Each of these three views has introduced various principles, methods, and tools.
Koskela believes the integration of these three partial theories will create a unified theory
of production in construction, namely the TF V theory. Table 2.1 gives an overview of the

integrated TFV theory (Koskela et al 2000).

Concept

 

Principles

 

Methodologies/Tools

 

Figure 2.2 Methodologies based on concepts and principles
(Source: Koskela, 2000)

In this research, the focus will be on Last Planner system shown in Table 2.1 as a
method of production control under the flow view. The Last Planner focuses
management’s control functions on production processes versus on overall activities. A
main feature of Last planner is to allow the last person in the authority line to plan and
decide the assignments to be done. To assess the quality of assignments made, a simple

metric is used. This metric is termed as Percent Plan Complete (PPC).

24

 

Transformation view

Flow view

Value generation view

 

Concept of

production

Transformation of

inputs to outputs

Flow of material
and information,
composed of
transformation,

inspection, moving

Process where value for
the customer is created
through fulfillment of

his/her requirements

 

 

 

and waiting
Main Getting production Elimination of Elimination of value loss
principle realized efficiently waste i.e., non- (achieved value in
value adding relation to best possible
activities value)
Methods WBS, Material Continuous flow, Method for requirement
and Requirements Planning Pull techniques, capture, Quality Function
practices (MRP), Organizational Production Control Deployment
responsibility chart (Last Planner

 

 

system) Continuous

Improvement, .1 IT

 

 

Table 2.1 Theory of production
(Modified from Koskela et al 2000)

PPC is calculated by expressing the number of completed assignments as a ratio

of the total number of assignments planned for a particular production unit along the

assembly line in a given week or any other period over which assignments are typically

made. PPC can take on values from O to 100%, with the latter being the best case. If PPC

is less than 100%, there is a failure in the production planning process. Understanding the

reasons for the failure will enable future improvement of the planning process. By

improving the quality of assignments, PPC will increase reflecting a more reliable

planning system.

25

 

2.3.4.1 Lean Project Delivery System

The Lean Project Delivery System is “a set of interdependent functions, rules of decision-
making, procedures for execution of functions, and as implementation aids and tools,
including sofiware when appropriate” (Ballard 2000). The intersection of projects and
production systems forms the domain for LPDS, which is called the project-based

production system (Ballard 2000).

 

 

 

 

 

 

' — 1
V

‘ Fabrication l

& Logistics

Detailed I

Engineering ' Installation

 
     
 

 
   

Criteria

 

 

 

 

 

 

 

 

 

 

 

Project Definition Lean Design Lean Supply Lean Assembly Use

 

 

 

Production Control

 

 

 

 

Work Structuring

 

 

 

 

 

 

[1 . \
\Learnmg Loops/

Figure 2.3 Lean Project Delivery System Model
(Modified from LPDS, Ballard, 2000)

Essential features of LPDS are as follows (Ballard 2000):

0 Structure and management of a project is aimed at creating value.

26

Cross-functional teams, involved in front end planning and design, include
members from all the areas of a production process.

Project control would be a tool executed throughout the project as opposed to
reliance on after-the-fact variance detection.

Optimization efforts are focused on making work flow reliable and not to focus
on improving productivity.

Pull techniques are used to govern the flow of materials and information.
Capacity and inventory buffers are used to absorb variability in the production
process.

Feedback loops incorporated at every level, are aimed at a rapid system

adjustment and learning.

Work Structuring

Work structuring is a fundamental level of process design in the Lean Project Delivery

System. The purpose of work structuring is to make work flow more reliable and quick

while delivering value to the customer.

Production Control

Production control governs execution of plans and extends throughout the project. Last

Planner is the system of production control in Lean Construction. It consists of workflow

control and production unit control. Workflow control is accomplished primarily through

the lookahead process, which is explained later in this chapter, while production unit

control is accomplished primarily through weekly work planning. Production control

27

extends through all phases of LPDS. The different phases of Lean Project Delivery

System are Project definition, Lean design, Lean supply, Lean assembly and Use.

Project Definition

This phase, managed by the project manager responsible for the project, includes both
designing and building of the project. Traditional sources are used as inputs and are
integrated with post-occupancy evaluation. Design criteria for both product and process

will be produced in this phase.

Lean Design

This phase develops the conceptual design from Project definition into Product and

Process design, consistent with the design criteria produced in Project definition.

Lean Supply

Lean supply phase consists of detailed engineering of the product design produced in
Lean Design, then fabrication or purchasing of components and materials, and the

logistics management of deliveries and inventories.

Lean Assembly & Use

Lean assembly and Use are the last two phases of LPDS. Lean assembly begins with the
first delivery of tools, labor, and components to the site and ends when the keys are

turned over to the client. Use phase is to handle the stages after the client is handled the

28

product. Operations and maintenance, commissioning and decommissioning are parts of

this phase.

The application of LPDS can be on a full range of projects, but would be more
appropriate to temporary production systems. Afier knowing, the kind of project delivery
system and to what types of projects LPDS can be applicable, the following lean

construction principles can be outlined.

2.3.4.2 Lean Construction Principles

The Lean Construction Institute (LCI) has developed through its extensive research a
Lean Project Delivery System (LPDS) as shown in Figure 2.3 intended as a template for
applying lean construction to the industry. A lean approach to manage construction
projects is different from the traditional method. Lean construction has the following

essential features (Howell 1999) and (LCI 2001):

1. Clear set of objectives to be established for the delivery process. Customer needs
and requirements are well understood.

2. A cross-functional team designs product and process concurrently, to give more
value to the customer needs - this process of parallel design helps positive
iteration within the process and negative iteration is reduced.

3. Shifting design work along the supply chain to reduce the variation and match the

work content.

29

4. Work structuring of the entire process increases value and reduces waste at the
project delivery level. Efforts to improve performance at the planning level

increases performance at project level.

“Making things happen” rather than “monitoring results” is lean production
control. Measuring and improving the performance of production control systems
improves the project performance and hence value to the customer is achieved.

Production control is applied from design to delivery.

2.3.5 Research In Lean Construction

Various research projects have focused on tailoring lean concepts to the construction
industry. Most of have been concerned with applying the lean production principles in the
site-built construction. The Lean Construction Institute, commonly known as LCI, is an
active research institute working to adopt lean principles and develop this innovative
project management approach. LCI has developed a new project delivery system for the
construction industry based on these lean construction principles. “Last Planner System
of Production Control ” by Glenn Ballard, is a part of this model and is developed
specifically to be used a production control technique for construction projects. Parts of

the Last Planner concepts are utilized in this research.

Research efforts in applying the lean production principles in construction
industry have shown excellent results. Various projects were executed as part of the
research work conducted by the Lean Construction Institute researchers in the site-built

construction. Most of the projects were commercial and large-scale projects. In spite of

30

the heavy complexity and uncertainty in such projects, they have shown good results
using the lean approach. Such results are always an encouragement and the same results

are driving many construction companies to become lean.

Research paper by Amarjit Singh (2002) titled “Lean Engineering for Mass
Housing — Design, Manufacture and Site Erection” at University of Hawaii at Manoa,
talks about the tremendous shortage of housing in the world and how lean construction
can help to overcome such shortages. The paper discusses an innovative approach for fast
and low-cost erection of mass housing units that have architectural flexibility,
manufacturing flexibility, and erection flexibility (Singh 2002). It explores the flexibility
of design, manufacture, and erection in mass housing and presents an innovative housing
system using lean concepts, like reducing waste, product innovation, and agile

production.

Production control is of primary interest to this research because it is an essential
tool in coordinating work assignments and workflow. To understand clearly the
production control system for project delivery, the sections below describe traditional
project control versus production control followed by the Last Planner system of

production control.

2.3.6 TRADITIONAL PROJECT CONTROL

To deliver a project according to customers’ needs every project must exercise a form of
planning and control. Project controls, as commonly known in the traditional

construction industry and other industries using contemporary project management, play

31

an important role along with planning of the production process in any construction
project. In general, the objective of project controls in traditional construction project
management setting is to detect negative variances or deviations from target or baselines,

and then to develop corrective actions.

The main objects of traditional project control are time and cost. Time control
checks on production output or progress, while cost control is primarily concerned with
tracking actual productivity rates and comparing to the rates used in project budgeting.
This function is performed using different tools/methods to control time and cost in a
construction project in the traditional methods. The basis of all these tools/methods is the

Work Breakdown Structure (WBS).

The WBS is one key element of traditional project management. “The WBS
provides the framework for defining the project from top all the way down to its smallest
components and for accumulating the costs associated with each piece. In doing so, the
WBS provides a data base from which problem areas can be identified, forecasts made,

and corrective action can be taken (Diekmann and Thrush 1986).”

The traditional system of project control focuses more on keeping track of time
and cost, with corrective actions taken after the system is off-track. “Without corrective
actions a project control system becomes merely a cost/schedule reporting system
(Diekmann and Thrush 1986)”. Traditional project control systems assume that causes of

deviation are apparent and corrective actions will be taken accordingly.

32

2.3.6.1 Production Control

This control approach is different from the concept of controls under a project
management paradigm. ‘Production’ is most commonly understood as either
‘manufacturing’ or ‘making’ where many copies of a single design are manufactured.
“Within the world of construction, manufacturing in this sense is approached most
closely by manufactured housing” (Ballard, 2000). The construction industry’s
production cannot run through an assembly setting as the assembled product eventually
becomes too large to move. So, most of the making, production or construction takes
place on the site, which makes it fixed position manufacturing. Manufactured housing
plants are the only construction sector where ‘production’ can be viewed as a moving

assembly-line process.

Reviews of lean production literature and prior research done in designing a
production control system suggest the following guidelines and criteria for an effective

design of a production control system (Ballard 2000):

l. Variability in a system should be reduced and the remaining variability must be
managed.
2. All the assignments should be sound in regard to the prerequisite work.
3. Completion of assignments is measured and monitored.
' 4. Causes for failure in realization of such assignments are identified and removed.
5. A buffer of sound assignments is maintained for each crew or production unit.

6. Prerequisite work for upcoming assignments is actively made ready.

33

7. Pull systems replace the traditional push technique'.
8. Production control facilitates work flow and value generation.
9. The project is conceived as a temporary production system.

10. Decision-making is distributed in production control systems.

Several applications of the production control concepts to the AEC industry have
been implemented over the years. The focus of a production control system would be

appreciated when contrasted with the existing project control system.

Under traditional project management, the work is planned mainly with design
information and project objectives. Afier initial planning, a schedule of “Should be done”
activities are identified and committed to by executors. As shown in Figure 2.4, these
“Should be done” activities are then performed using available or ‘pushed’ resources.
Performed work is tracked as the ‘Did’ activities. The control process is responsible for
determining the differential between the “Shoulds” and the “Dids” and responding to

eliminate it.

Using the lean-based production control concepts mentioned earlier, a new
production planning and control system called Last Planner system was developed by
Ballard (2000). When included in the current planning system, the Last Planner system

would first transform what “SHOULD” be done into what “CAN” be done. Then, from

 

1 Push Systems: In this system materials or information are pushed or supplied based on scheduled
milestones that take into account customer demand not plant or site status. The current system of
construction scheduling follows Push mechanism.

Pull Systems: Suppliers are “authorized” to release material or information on certain dates based on plant
or site status.

34

the “CANS”, a weekly work plan of “WILL” be done activities can be formed. The
assignments that “WILL” be done are a commitment of the Last Planner to perform this

work. The Last Planner is the person or group of peOple who have the skill and authority

Represents the input for
0 an activity to take place

This represents an
activity execution
utilizing resources and
information provided

to make daily assignments.

 

 

 

 

 

 

 

 

 

 

Project
Objectives

 

 

   
     

   
 

 

 

 

 

 

Project
Planning the Control
work Should (variances)
Resources Executing the
plan

 

 

 

Fig 2.4: Traditional Planning and Control System (Source: Ballard, 2000)

Figure 2.5 represents the Last Planner planning process which makes the

“SHOULD” and “CAN” assignments to “WILL” in the form of weekly or daily work

35

packages ready to be executed with all the prerequisite work done. The production
control process compares, usually on a weekly basis, the “WILLS” to the “DIDs”. The
“DIDs” are expressed as a percentage of the “WILLS” and this percentage is called
Percent Plan Complete (PPC). The Last Planner system of production planning and

control will be explained in detail in the following sections.

 

 

Planning the
work

 

 

 

 

Last
Planner

 

 

 

 

Resources

 

 

 

 

Fig 2.5 Last Planner Planning Process (Source: Ballard, 2000)

2.3.7 LAST PLANNER

Manufacturing a product requires proper planning to deliver goods in time and quality.
There is a lot of decision making in the production process where the product moves

from one stage to the other. A person who decides last, if the work is ready to be

36

executed or not, is the Last Planner. The last planner could be a foreman, floor manager
or a group of people. Before understanding the whole process of last planner and how it
works, it would be necessary to understand the decision-making process in the current

construction industry.

Planning of the production process in site-built construction is typically
performed using the method of Work Breakdown Structure (WBS). In this method, the
project is broken down into activities and sub activities, which in turn are scheduled
ahead of execution. Each activity would also be assigned necessary resources for
executing as scheduled. Not every activity is completed as scheduled and with the
assigned resources (budget). Some activities may need additional resources and some
might just need more time. These situations result from the uncertain and unpredictable

nature of construction projects.

Thus, the need for reliable planning is essential in order to produce better, both in
terms of quality and cost. The Last Planner system helps in better planning and
controlling the production process. The focus of Last Planner system is to improve the
quality of assignments in weekly (or daily) work plans and to shape and control work

flow.

Planning for assignments in the Last Planner system is performed through the
“Lookahead” process. Usually the lookahead involves consideration of potential
assignments for the next 3-12 weeks based on the project characteristics. The lookahead
process takes its input from traditional project planning techniques, which result in a

master schedule with project milestones and phase schedules as shown in Figure 2.6.

37

Master schedules involve the development of logic and sequence that helps identify the
commitments through out the project. The phase schedule involves greater detail of
planning where the project components are tested for logic and the work is divided into
phases to identify constraints or related work. This system of division into master and

phase schedules lacks a tool of detailed work structuring.

 

 

 

 

 

. , Master Schedule
Traditional Strategic Planning
Techniques Project Milestones Adjusting the
Logic and Sequence plan through
Identifying Long Lead Delivery requirements measured
Drilling down progress and
from strategic lessons learned
scheduling into .
production Detailed (Phase) Schedule
planning and .More detail
detailed work ProgcttPOflpoaems
- es rng og1c
aSSIgnments Identifying Related/Constraint Groups

 

 

 

 

Last Six-Week Lookahead
P1811119." Constraints analysis
Techniques Responsibility Identification
“To-Do” list
Resource leveling
Weekly updates

J,

Weekly Work Plan
Detail work assignments
Decentralized Crew Planning
Measurement of Percent Planned Complete
Learning
Immediate Make-Ready Needs
Workable backlog

 

 

 

 

 

 

 

Figure 2.6: Construction Planning vs. Last Planner
(Modified: Kaufman Consulting Group, LLC)

38

In the Last planner system, the activities, before entering into the lookahead
window, are exploded from master schedule or phase schedule into a level of detail,
appropriate for an assignment on a weekly work plan. This typically yields multiple
assignments for each activity. As each assignment appears in the lookahead window (a 3-
6 week time period), it is subjected to constraint analysis to make sure it is ready to be

executed (Ballard 2000).

Assignments that are made ready for execution enter into a workable backlog.
The assignments entering the workable backlog are all constraint free and in the proper
sequence for execution. If the last planner finds a constraint that could not be removed in
time, the assignment would not be allowed to move forward. The last planner should
maintain a backlog of work ready to be performed, with assurance that everything in the

workable backlog is indeed workable (Ballard 2000).

Weekly work plans are formed from the workable backlog. Such assignments
help improve the productivity of those who receive them and increase the reliability of
workflow between the production units. The last planner system can be viewed as a
needed supplement to traditional project management for better production. The analysis
of reasons for plan failure reveals more information regarding how the production system

actually functions and what could be done to improve it.

The planned assignments by Last Planner can be successfully controlled by
keeping track of the percentage plan complete. (PPC, Percent Plan complete is the
number of completed assignments and expressed as a percentage of planned

assignments.) Failure to achieve a 100% PPC are investigated to find the root cause. This

39

method of getting to the root causes is termed constraint analysis. In essence, PPC is a

measure of the reliability of the production planning system.

2.4 SUMMARY

In Chapter II, a detailed description of terms and existing research related to
manufactured housing industry, the kind of production process in a MH plant are
discussed. The second part of Chapter II summarizes the different production processes,
lean production, lean construction, and the existing research in these areas. Finally a
background of production control and last planner are described. The next chapter,
Chapter 111 describes the detailed methodologies and tools to be used in achieving the set

goal and objectives of Chapter I.

40

CHAPTER III

METHODS

41

3.1 INTRODUCTION

In chapter 1, manufactured housing was introduced and an argument was made for
production planning improvement. The goals and objectives of this research were also
discussed. Relevant literature on the construction and manufactured housing industries,
and the production management principles applied in these areas were explained in
Chapter 2. Lean construction principles and the Last Planner system were discussed in
Chapter 2 to provide the reader with a basic rounding on the origin of the tools and
methods that will be used to achieve the set objectives. In this chapter, the methods, tools,

and procedures used will be presented.

While the manufactured housing industry may be able to meet demand, it is
important for the industry to lower costs of production arising from both fixed and
variable costs of production. This is possible by performing a critical assessment of the
production planning processes, which would help identify better methods of streamlining

production and maintaining reliable workflow.

The main aim of this research is to evaluate the production planning process
through a lean production lens and suggest process improvement initiatives. Generally, a
process improvement study involves various steps to anive at process alternatives leading

to better performance. These steps are as follows:

1. Collect adequate data.
2. Analyze the data for problem patterns and trends.

3. Identify root causes of the problem.

42

 

4. Develop a new or revised method.
5. Implement the new method.
6. Evaluate impact of new method and refine approach, if necessary.

7. Permanent implementation.

This research is limited to the first four items on the list. Steps 5-7 are not in the
scope of this research and are, in fact, quite difficult given the required approvals and

cooperation manufactured housing companies.

3.2 METHODS AND TOOLS FOR OBJECTIVES

In the following sections, the methods used to achieve the research objectives in the
context of the process improvement steps outlined will be presented. The author also
visited two manufactured housing plants to understand the current production planning
process and also the production process. Information on the data collected from these

pilot visits will also be discussed.

3.3 OBJECTIVE 1

The first objective of this research is to document the current production planning process

utilized in a manufactured housing plant. This will be done in two steps:

1. Study and understand the production process in a manufactured housing plant.
2. Study and document the current production planning process in a manufactured

housing plant.

43

3.3.1 Understanding Production Process in Manufactured Housing

To study and document the current planning process in a manufactured housing plant it
was important to first understand the production process of a housing unit. Existing
literature was reviewed and plant visits were conducted to understand the production

process.

A part of a typical plant layout is shown in Figure 3.1, found in Senghore (2001),
with production stations; setting of roof, roof insulation, and exterior boards with

different activity relationship among stations.

 

 

 

 

Roof _‘ Roof Cover
' . '—'> . F
Insulation Deckrng Decking Shingles

with Paper

Setting the
Roof

Tl

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Installation
Exterior p of Exterior I . .
Boards Doors and Sldlng
Windows

 

 

 

Figure 3.1Example of production operations in MH
(Senghore, 2001)

As shown in Figure 3.1, there are various activities at each station and the
relationships among these activities are very important. Work flows from one station to
another and among the activities in each station. For example there are three different

activities in the roof decking station, namely, roof decking, covering decking with paper,

44

and placing shingles. Work flows from placing decking to covering with paper and then

placing shingles.

A different way of representing workflow and work requirements can be
performed using the Activity Definition Model (ADM). As shown in Figure 3.2, the
ADM represents directives, prerequisite work, and resources as inputs to the assignment.
The output of the assignment is also represented in the figure and this could be a product
or a release of workspace to another crew. This model of defining process activities
improves the planning and sets in place a work ethic that leads to well planned
assignments. It is also important to note that if an output does not meet set criteria, the
work has to be redone. Needless to say, outputs that meet criteria are allowed to trigger

downstream assignments.

     
   

Meets
Criteria?

 

 
     
 

Prerequisite
work

Roof Insulation

 

 

 

Resources

Figure 3.2 Activity Definition Model for Roof Insulation activity
(Modified: Ballard, G. 2000)

45

For the stations under study in this research, a representation of work similar to
that of figure 3.2 was used. Representing activities in this manner would help understand

the production process and the production planning process.

3.3.2 Documenting Production Planning Process

With the review of existing literature on production process of manufactured housing
plant, the author collected information on the current production planning process by
visiting manufactured housing production plants. Personnel involved in production

planning were contacted to both understand and map the production planning process.

3.3.2.1 Data on Production Planning Process

The type of data required from a manufactured housing plant to assess the production

planning process in building a manufactured housing unit was as follows:

1. How are assignments planned for crews working at a specific station?

2. How many assignments are planned for a day or an hour?

Other questions about the planning process also come from Figure 3.2. For
example, a question was asked regarding the type of directives needed for a particular

assignment and so on.

To answer these questions, the following two methods were be used:

1. Observing (videotaping) the process to develop actual records of the current

operations.

46

"I

2. Interviewing management and labor involved in the operations.

Both methods have respective advantages. On the one hand, asking the personnel
involved would help detect problems or lead to suggestions for improvements. On the
other, observation is an important tool for gathering information on a work process or
method. Detail recordings of people’s actions, flow of materials or information allows
later studying and analysis. What, When, Where, How, and Who can be answered by
observing the current practices, but only asking would explain the very important
question ‘Why.’ This is because personnel involved in day-to-day operations are
knowledgeable about what goes on and have excellent insight to sources of problems. It
is also important to include front line workers in the interview process and not focus only
on management. It is well documented that workers or foremen often have a great
knowledge and better insight of situations on the line (Oglesby et al 1989). This very

important source was tapped in this study.

As mentioned earlier, this phase of research followed the two mentioned
approaches of observing (includes videotaping) and interviewing, and the choice is based
on scope limitations. Data collection will include direct observation in the manufactured
housing plant, documentary analysis, and interviewing the personnel such as production

plant manager, production planner, foremen and workers.

47

 

3.4 OBJECTIVE 2

The second objective of this research was to evaluate the production planning process
performance using lean construction principles. This objective will be achieved in two

phases:

a) Measuring the quality of assignments. (Is the work being completed as
planned?)
b) Assessing the performance of work crews. (How efficient are work crews in

completing the work assignments?)

The two main techniques, which will be used to reach this objective, are the Last
Planner system and productivity rating studies. These techniques will be explained in the

following sections.

3.4.1 Last Planner

In this phase of the research, the reliability of the production planning process will be
assessed using the Last Planner system with PPC as the assessment metric. Data for PPC
calculation will be collected for workstations found in Senghore (2001) for a period of 2

weeks. For a typical week, the following are the steps to calculate PPC:

1. The first step in calculating PPC would involve collecting planned daily
assignments at work stations of interest. The number of assignments planned will
be collected from the production manager or foreman. Table 3.2 shows the form

for collecting this information.

48

2. The number of completed assignments at the end of the day will be determined by

asking the production manager.

3. The ratio of completed assignments to those planned activities will be calculated.

The resulting percentage is the PPC.

By completing the data collection for PPC as described above, it will be possible to
determine the average PPC at a particular station and across stations. The next step is to
analyze the failures in meeting planned assignments. This will be the topic discussed

under the third objective.

Table 3.2 Sample format of PPC data collection

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Station Week Week
Station observed No. May4-May8 Mayl 1-May15
M l TjwlTIF MITIWITTF
833335.253: 32113113. Remake
1.Build truss Y Y
2.Attach ceiling 6 Y
3.1nsulate ceiling N Y
4.Paint ceiling N
Completed Assignments 1 3
PPC 50% 75%
ESTER? 31:11:31"; Remarks‘
l.Clean the floor
2.Lay carpet 12
3.1nstall lighting and
electric fixtures
4.Final cleanup
Completed Assignments
PPC

 

 

 

 

 

 

 

 

 

 

 

 

 

49

It is important to note that PPC does not provide a measure of how efficient the
assignments were conducted. In other words, a PPC of 100% does not indicate whether a
crew was idle or working under an efficient combination. Hence, PPC measurements, at a
particular station, need to be complemented with work performance or work efficiency
measurements. This determination of efficiency can be made using work sampling
techniques. This topic is part of the second objective and will be discussed in the next

section.

3.4.2 Work Sampling

Work sampling is used as a data gathering technique for performance-improvement
applications. Because operations cannot be observed over a long period of time, sampling
offers an alternative. Sampling of operations on a site is rather an approximation of the
reality within acceptable limits. The sampling results can form a basis for judgments

about productivity problems (Oglesby et al 1989).

Work sampling can be done by various approaches, namely, field ratings,
productivity ratings, and 5-minute ratings. The most comprehensive and suitable for this
research is the productivity ratings method, because it explicitly accounts for work that is
essential to produce a product versus work that is contributory but non-essential. The

following section explores this method further.

3.4.3 Productivity Ratings

During the operations in a manufactured housing plant, the workers perform many tasks

similar to the ones performed in site-built construction. As discussed, to assess how well

50

these tasks are planned, the PPC metric will be used. The author also observed on two
site visits to manufactured housing plants that complementing PPC with productivity
ratings would give a better insight of the crew performance of planning operations in a

manufactured housing plant.

Productivity rating techniques are typically used to assess the amount of
productive work performed by workers. Productivity ratings refer to ratings of a crew or
crewmember performance, while assignments are conducted. The ratings are typically
stated as percentages of productive to non-productive work. These ratings are determined
by sampling the activities of crews or crewmembers. In the next section sample size
determination and classification of work under the productivity ratings method will be

discussed.
3.4.3.1 Sample Size

Work sampling involves making and analyzing the results of fixed observations. This
helps determine what individual workers are doing at specific instants in time. Any
sample size taken for observation has to satisfy a set of conditions based on the type and
nature of research performed. According to Oglesby et a1 (1989), a 95% confidence level
and limit of error of plus or minus 5% gives a good indication of the overall effectiveness

of an organization or of an operation.

The Confidence level mentioned refers to the level of dependability in a result. If,
say, the confidence level is 95%, then the answer can be relied on 95 percent of the time

or, conversely, that the answer may be wrong 5 percent of the time. The choice of

51

confidence level is based on the purpose of the sample. If the subject involves humans,
then the level of confidence is as set high as 99.99 percent. If the sampling is for a
construction activity, then confidence level of 95 percent is acceptable. The higher the
confidence level, the larger the number of observations or samples that must be made

(Oglesby et al 1989).

The other important factor in determining the sample size is limit of error. Limit
of error refers to the accuracy of the estimated result. With a given confidence limit, the
percentage variation on either side of the expected value of the sample is the limit of
error. The true values are expected to fall in this limit. For example, with an error limit of
10 percent and confidence level of 95%, it can be stipulated that the estimate of non-
productive work based on sampling can fall within plus or minus 10 percent of the total
work situations that were actually non-productive and this result could be depended on 95

times out of 100 (Oglesby et a1 1989).

For the purpose of this research, and based on prior construction research, a
minimum sample size of 384 with a confidence level of 95 percent and limit of error of 5
percent is satisfactory. This sampling can be applied to crews or projectsof any size,
because sample size is not related to how many number of individuals are observed.
“However, it would be necessary to sample a crew of 100 men 4 times or a crew of 10

men 39 times to meet the minimum number of 384 observations” (Oglesby et. 31.1989).

52

3.4.3.2 Classification of work

With the sample size determined, productivity ratings can be achieved by classifying

work into different categories according to performance. Under the productivity ratings

method, work is classified into three main categories as follows (Oglesby et al 1989):

1.

2.

3.

Eflective work: Work or activities that are directly involved in the actual process
of making a unit or adding to the unit is considered as effective work. Work, such
as, assembly of a roof truss unit or the work involving activities essential to the
process of building a roof truss unit in a manufactured housing plant is effective
work. Basically, effective work is work that leads to a change in shape, size, or
form of material resulting in an end product to emerge.

Essential Contributory work: Work done through associated processes essential in
finishing the unit, such as material handling, cleanup, checking drawings making
measurements, etc.

Non-useful or Idle work: Ineffective work like being idle or doing something that
is unnecessary to complete the job can be classified as non-useful work or Idle

work.

This general categorization of work can be used in finding productivity ratings in

any kind of construction work. Table 3.3 gives a sample of productivity ratings for a

number of construction trades. As shown in Table 3.3, every trade in construction work

has some amount of non-productive work. Theoretically, ideal work condition would be

100% productive work, but practically this is hard to attain. With the similarities in work

53

between MH plant and general construction, MH productivity ratings can be expected to

be similar to those shown in Table 3.3.

 

 

 

Percent of total time in category
Trade or craft Effective Contributory Not useful
Painter 45 24 31
Carpenter 36 35 29
Electrician 42 37 2 1
Laborer 35 29 36
Carpet layer 29 41 30
Equipment operator 47 35 18
Plumber 39 32 29
Shingle layer 42 29 29
Finishing worker 48 3O 22

 

 

 

 

 

 

Table 3.3 Productivity ratings for several construction trades
(Source: Oglesby et al 1989)

Worker activities in a manufactured housing plant were identified and classified
into the above—mentioned categories of work. Work identified as effective, contributory,
and idle from the sample collection would help in more than one way. For example, it
helps the management in identifying specific situations in which the work can be done
more effectively and at a lower cost. After productivity ratings are obtained, they serve as

input for calculating Labor Utilization Factor, which is detailed in the next section.

54

3.4.3.3 Labor-Utilization Factor

A labor-utilization factor, based on the productivity ratings method, is a percentage
obtained by summing the number of observations of effective work and one-fourth the
number of observations of contributory work, and the sum divided by the total number of

observations. The following equation shows this calculation (Oglesby et a1 1989).

Effective work + M: essential contributory work

 

Labor-utilization factor =
Total observed

Total observed = Effective + Essential contributory + Ineffective

The above equation used to calculate Labor-Utilization Factor (LUF) is taken as
formulated in the construction literature (Oglesby et a1 1989). As described above, the
work is categorized into three — Effective, Contributory, and Idle work. To calculate the
amount of productive work performed, it is necessary to eliminate the work that is not
adding value to the unit being built. Effective work was described as work that totally
adds value to the unit, but in the case of contributory work, only a part of the work adds
value to the unit. In construction industry, it has been formulated that only 1/4th of the

contributory work can be considered as productive work.

Based on this literature, one-fourth of essential contributory work is used in
addition to effective work to compute the Labor Utilization factor. Supporting data can be
found in Oglesby et al 1989. The essential contributory work could range anywhere from

1/4 to 1/2 based on type of work in different industries.

55

Though the main goal of any work-improvement process is to increase the
number of employees engaged in the effective-work category, some amount of
contributory work is necessary in completing the work. Therefore, the LUF recognizes

the essential contributory work when assessing the overall work performance.

In this research, the same workstations identified for PPC measurements will also
be observed and analyzed for productivity ratings. The productivity ratings will be used
to determine the LUF at each workstation. Recall also that PPC is used to assess the
production planning process. The ideal scenario of the performance of the planning
process that can be expected could be a high PPC with a high work performance
represented by a high LUF. In other words, PPC and LUF would have a linear relation as
represented in Figure 3.3. However, it is expected to find a non-linear relation between

PPC and LUF, similar to that in Figure 3.3, is also possible.

A

Labor-Utilization Factor
(LUF)

 

 

Percent Plan Complete (PPC)

Figure: 3.3 Comparison of Percent Plan Complete vs. Labor-utilization factor (LUF)

56

However, it is also possible to find a non-linear relation between PPC and LUF,
similar to that in Figure 3.3. Thus, LUF and PPC would be a good measure to assess the
performance of the planning process. While PPC gives the measure of the quality of
assignments, LUF gives the efficiency of workers that execute their assignments. The

investigation of this relation is one of the unique contributions of this research.

As mentioned, PPC will be calculated based on daily operations in the
manufactured housing plant. It could range anywhere from 40% to 100%. If PPC is low,
the last planner techniques can be implemented to improve it. Using techniques like
constraint analysis and lookahead process, PPC can be increased to 90% and above. This

has been successfully proved in prior implementations in different kinds of projects.

3.5 Objective 3

The third and final objective of this research is to identify causes of off-target
performance through constraint analysis techniques. In this research, Pareto analysis and
Fishbone diagrams, were used to suggest opportunities for production planning process

improvement.

3.5.1 Constraint Analysis

It is important for a company to be prevention oriented, rather than detection oriented,
engendered by traditional methods of production management (Markland et a1 1998). A
prevention-oriented approach can be achieved using various techniques. One such
technique is constraint analysis that identifies reasons for failure in completion of

assignments and suggests improvement. The author proposes a 3-step process of

57

constraint analysis for process improvement of production planning process in a
manufactured housing plant. The three steps in the proposed constraint analysis technique

are I

1. Identifying incomplete and delayed activities from the PPC data.
2. Identifying the most trivial problems using Pareto analysis (Markland et al 1998).
3. Identifying the assignable causes of the out-of-control condition using Cause-and-

Effect diagrams and suggests improvements (Markland et al 1998).

Assignments that are not completed or delayed can be identified from the Percent
Plan Complete (PPC) data. Such assignments are identified at each station from
observation and input from the supervisors of the station under study. These assignments

are investigated to identify the reasons for non-completion.

3.5.2 Pareto Analysis

Pareto analysis is a technique that identifies the ‘vital few’ factors that are responsible for
most of the problems. Most of the time, it is those few causes (20%) that create great
majority of the problems (80%). Pareto analysis uses a bar chart to indicate such ‘vital
few’ and ‘trivial many’ factors that effect the proper functioning of any process in

general.

For example, out of many reasons for non-completion of assignments in a
manufactured housing plant, assume that design failures occur frequently. The next most
frequent problem is inadequate supply of material and many other reasons that occur not

so frequently. Pareto charts identify these very frequently occurring problems and those

58

that have the greatest effect on the overall production planning system. Identifying of

such vital few problems can be graphically represented as in Figure 3.4.

The next step of constraint analysis is to find the root cause of the problem as
identified in Pareto analysis. This is performed using the cause-and-effect diagrams as

described in the next section.

 

 

Magnitude of Concern

 

 

 

Vital Few Trivial Many

Concern Categories

 

 

 

 

 

Figure 3.4 General Format for a Pareto Chart
(Modified from Markland et al 1998)

3.5.3 Cause-and-Effect Diagrams

Cause-and—Effect diagram commonly known as Fishbone diagram or an Ishikawa

diagram is a tool for clarifying the causes of a problem (Markland et a1 1998). Identifying

59

the assignable cause of a problem is done using this fishbone diagram with input from the
personnel involved in the planning process and the execution of the assignments. Figure
3.5 illustrates the general structure of a cause-and-effect diagram. The problem or
concern being addressed is on the right side of the diagram. This is the effect. The factors
influencing the effect are generally classified as labor, material, equipment, and methods.
The factors can be different based on the type of process. These four factors are the four

principal branches (fish bone), from which different smaller branches are listed.

Equipment Material

C
auses\ \ \ / Causes

/ \ /
\ \\ 5

Causes

7— \ \
Causes /
/ \ \

Methods Labor

 

 

 

 

 

 

 

 

Figure 3.5 Cause-and-Effect Diagram (Fishbone Diagram)
(Source: Markland et al 1998)

For each major branch or fishbone, the method of process improvement involving

management and workers would help identify a set of strategies for corrective action.

60

'—Y

These set of strategies are developed from the information gathered from the production

manager, foreman and the workers.

In this research, the information collected was categorized for root causes. For
example, if a certain resource was unavailable at the time of execution, the reason for
non-availability of the resource was investigated. Such information would be helpful for
management to prevent future reoccurrences. Maintaining a log of such frequently

occurring problems will help improve production planning.

These three steps of constraint analysis would help identify most of the root
causes that send the production in a manufactured housing production plant off-target.

The fault may be in the production planning process, production crew, material etc.

3.6 PRODUCTION PLANNING PROCESS IN MANUFACTURED HOUSING

PRODUCTION PLANTS

To verify how realistic and achievable the proposed methods are, the author visited a MH
plant to learn about the production planning process currently taking place. The following

sections give an account of the visits.

3.6.1 Brief about the Production Plants

The two plants visited by the author are different in size and number of units produced.
For reference the plants would be represented as Plant ‘A’ and Plant ‘B’. Plant A is one
of four production plants and the one with a higher production rate among the four. It is a

medium size plant with 13 workstations along the assembly line, a few subassembly

61

stations and feeder stations to supply material to these workstations. The plant has a
production manager, three foremen and 130 workers. Foreman 1 supervises workstations
1-3, foreman 2 supervises 4-7, and foreman 3 manages stations 8-13. On normal
production days, the plant produces 5 houses (10 floors) per day (8 hour day). Due to the

variations in demand and sales, production takes place only 4 days a week.

Plant B, owned by a different company, is smaller in size, has lower production
compared to Plant A. Plant layout and crew make-up are very similar to Plant A.
From the two plants visited the author believed plant A is better in production and
capacity than plant B. In addition, plant A was studied previously in Senghore (2001)

providing a basis for comparison and reference.

3.6.2 Production Process

In the following section the production process of Plant A is discussed. This section also

outlines the planning and execution of production operations.

3.6.2.1 Macro level planning

Macro level planning is done at the company level for all the plants in the company, in a
hierarchy as shown in Figure 3.6. Production volume is planned afier the General
Manager, Sales Manager, and the Operations Manager meet to discuss the company’s
sales and needs. After the sales manger determines the number of manufactured houses
that can be sold/marketed in a month, the team decides on the number of houses to

produce in that month. Operations are thus planned for the entire month. The operations

62

manager and the production managers meet to decide the number of days in a week the

plant will function.

Every morning before the plant starts, the production managers of all the plants
meet and discuss the day’s operations in a pre-production meeting. The production
manager then meets with the foremen in their respective plants and check on material,
equipment and labor. For example, if a worker is absent from work, the production
managers manage to continue work by substituting the absentees with “Swing Men”

(workers who can do any kind of job in the plant, but not skilled in any particular area).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

President
G.M.
Engineering Sales Materials Production
Manager Manager Manager Manager
Quality Production
Assurance Foreman (3)
Manager .
Quality
Inspectors
(3)

 

 

 

 

Figure: 3.6 Organization Chart for Plant A
Flaming of daily work is decided based on the number of layouts that have to be
produced. The work starts at 6 in the morning with a target of three floor sections to be

ready by 8:30, which is also the time for the foremen and the production manager to

63

meet. If this target is not achieved, red flags go up in the plant and the foremen along
with the Production Manager responds by putting more employees at the stations behind

schedule to speed up the work.

The plant aims at producing 10 floors (2 floors = 1 house) by the end of the day.
If the target units are not completed, the assembly line is stopped and the work is
continued the next day. On the next day in order to get back on schedule, additional

workers are assigned to the stations where problems existed.

3.6.2.2 Problems in the production process planning

The problems as seen by the production manager vary from planning to lack of

coordination in the work activities. Some of the difficulties faced in the plant were:

1. Size of the units varied a lot and some of the units are huge in size and it becomes
difficult to move it quickly through the line.

2. Inaccuracy in prefabricated components such as doors and windows.

3. Lack of space to move around along the assembly line and overall insufficient

space (Or may be too much inventory in the plant!) to work on the house.

To avoid more problems along the assembly line, the bigger units and smaller units are

run alternatively (if possible) in the production line.

The observation and study of this manufactured housing production plant’s planning

process gives a clear picture that

64

1. Though there is monthly, weekly and daily planning to some extent, there is no
detailed planning of day-to-day operations.

2. The activities are planned at a macro level only considering the number of units to
be produced in a week and number of floors to be produced in a day.

3. Discussion with the PM clearly shows that corrective measures are taken only
when there is a problem and there are no serious initiatives taken to prevent such

problems from reoccurring.

The aforementioned results of the pilot result supports the need outlined in this
research. Plant A will be further studied using the methods described which plant

personnel considered feasible.

3.7 SUMMARY

In Chapters 1 and 2, the manufactured housing industry was introduced, the problem
statement was stated, goal and objectives of this research were discussed, and literature
on both manufactured housing industry and lean construction was explained. In this
chapter, methods and tools to achieve each of the three objectives set in Chapter 2 were
explained. This chapter also explained in detail each step of the PPC, productivity ratings,
and constraint analysis tools used in this research. Next, Chapter 4 will present the results

of explain the entire data collection, calculations, and data analysis.

65

Chapter 4

Data Collection and Analysis

66

4.1 Introduction

In Chapter 3, the methods and tools to achieve the goal and objectives of this research
were presented. This chapter reports on the process of data collection, production
planning in the manufactured housing production plants, PPC and LUF data collection,
Pareto analysis and finally a detailed discussion of the data collected. As part of the data
collection for this research, the author visited two manufactured housing production

plants in northern part of Indiana.

4.2 Data Collection

The first step in the data collection process involved understanding and mapping the
production planning process. This was followed by the process of collecting data from a
few stations along the assembly line in a manufactured housing production plant. This
data collection process included videotaping activities at workstations, interviewing
people involved in the planning and production process, and on-floor observation. Video
recording of activities was performed to enable the productivity rating study, and to gain

more insights to the production process.

The process of data collection included several visits to two different
manufactured housing production plants in the initial stages of the project. The names of
the plants visited are not mentioned here to maintain participant confidentiality. First step
of data collection was to understand the planning process in a manufactured housing
production plant. Second step of the data collection process focused on one production

plant consistent with the scope of this project.

67

4.3 Production Planning Process

As described in Chapter 3, to understand and document the production planning process
in a manufactured housing plant, it was important to be familiar with the production
process of a typical housing unit in a manufactured housing production plant. The author
collected data from existing literature and plant visits. The following sections describe a

typical production plant, with its stages of production.

4.3.1 Manufactured Housing Production Plant

When a company steps into the business of producing manufactured houses, a production
plant is the first major investment. The plant has to be equipped with the machinery to be
used, material storage space, assembly line, and administrative offices. Various factors
have to be considered, typically with the help of an industrial designer and plant layout
designer, such as investment in building infrastructure, constructing the production plant,
hiring the production staff, and so forth. Figure 4.1 represents the steps to be followed in
the setup of a manufactured housing production plant. A recent masters thesis by
Mehrotra (2002) at Michigan State University is a good reference for guidelines in layout

designs for setting up a manufactured housing production plant.

68

 

 

 

v

 

Company/ Manufacturer

 

 

 

 

V

 

 

 

Purchase plant

 

site

 

 

Plant construction

 

 

 

 

l

Consult Designer

 

 

 

 

Meet Quality Standards
of HUD

 

 

l

 

 

 

Production Crew
and Staff

 

 

 

Plant Production

 

v

 

 

 

 

l

Design and

Management Team

 

 

 

Quality Inspection by
External Agency and In-
House Quality Manager

 

 

l

 

 

Product

 

 

Figure 4.1 Manufactured Housing Production Plant Setup

4.3.1.1 Manufacturing Codes and Standards

To build and ship a manufactured house, the manufacturer has to comply with the

National Manufactured Housing Construction and Safety Standards regulated by the US.

Department of Housing and Urban Development. The production plant has to be certified

of quality production to meet the HUD standard (Hullibarger 2001). Every unit that is

69

 

produced by the manufacturer must be engineered to meet both structural and quality

standards of HUD code.

Every unit model that the manufacturer produces with the standard and optional
features based on customer requirements has to be approved by a HUD-approved entity
known as Design Approval Primary Inspection Agency (DAPIA) (Hullibarger 2001).
Most of the models built by a manufacturer are standardized with some optional features
depending on customer needs. If the manufacturer changes the standard approved design,
a new approval of the design and engineering of the house has to be obtained from

DAPIA.

The manufacturers usually maintain a set of design and models for the customers
to choose. These standard designs are approved once and they make small modifications
based on customer needs and choice. Such units with little modifications do not need
approval each time. Usually manufacturers change the design of models once in a year
and they get them approved from DAPIA. Once approved the manufacturer starts the
production of houses. It is important to note also that DAPIA approves the

manufacturer’s quality assurance plan, known as Quality Manual.

Houses built in the plant are monitored by an external agency for compliance with
Manufacturer’s Quality Manual as governed by the HUD Code. Manufacturers usually
contract with a HUD-approved Inplant Primary Inspection Agency (IPIA), which is the
equivalent of building inspectors for site-built housing (Hullibarger 2001). Again, each
manufacturer has their own quality assurance program by which they build the units in

the plant.

70

The plant visited by the author was inspected daily by Inspection agency
personnel who monitor the regular production and check the quality of the plant
production. The inspectors go around the plant checking each house on the assembly line
to check if they are being manufactured according to the codes and standards. If a house
does not comply with the standards, the F oremen and Production Manager are informed

to correct non-conforming items according to the HUD code.

4.3.1.2 Size and Capacity

Manufactured housing plants are of different sizes and shapes based on the production
capacity of the manufacturer. Plants usually vary from a size of 30,000 square feet, with a
production capacity of two to three floors per day, up to 250,000 square feet, capable of
producing 20 floors per day (Hullibarger 2001). An average plant size would be around

100,000 square feet with a production capacity of about 8-10 floors per day.

The plant studied for this project is just over 150,000 square feet, employs about
130 people, and is capable of producing about 10-12 floors a day. The measure or
capacity of a manufactured housing plant is the number of ‘floors’ it produces. Each floor
can be an individual single section, or half module of a double section or a part of a multi
section house. The plant is designed to accommodate fiJture needs of building larger
spans, taller buildings, and more floors. The plant, which is a recent addition to the
company’s existing plant, has the entire infrastructure necessary to build HUD-complaint

housing.

71

The plant has equipment not traditionally used and some hi-tech equipment used
only in newly built plants. One such equipment is the air bladder system used to move the
houses down the production line. The air bladder, which raises the house only a quarter
of an inch off the floor, allows the house to be moved by two people, and the movement
is very easy to control with air valves. Shutting the valve lowers the house to floor level.
This system also facilitates fixing tires to the unit at the end of the production line. This
results in easier transportation of the house through the plant because of decreased

weight.

4.3.1.3 Manufactured Housing Plant Assembly Line and Stations

Manufactured housing production plant assembly layouts are designed in different
shapes. The assembly line is straight, U- or L-shaped with storage and sub-assembly
areas alongside the main line. Mehrotra (2002) suggests a good approach for the layout
design using the plant layout sofiware, FactoryPLAN. The other shapes of assembly line
layouts suggested are S- and Z-shaped. In general, house sections move along the

assembly line either end-to-end or side-by-side (Nutt-Powell 1982).

The assembly line is organized into work stations, with sub-assembly and feeder
stations along the main assembly line. The number of main stations ranges from about 10
to sometimes more than 25 (Hullibarger 2001). Different activities take place at different
stations. Some stations carry out just one activity like floor decking, and other stations
perform 2 to 3 activities, like roof truss insulation, roof decking, and at the same time,

external board fixing, and doors and windows installation.

72

The work force is usually divided into departments, which cover the major work
divided by clusters of stations. Most often the different stations are divided in clusters
and grouped based on the kind of work. The major clusters in a typical manufactured

housing plant are:

1. Frame/ Floors

2. Cabinets / Interior and Exterior Walls
3. Roof

4. External and Internal finishes

5. Final finish and repairs

Some of the manufactured housing plants may be broken down into even more
individual tasks. Each activity has a crew of workers to keep the assembly line
production running continually. More details on the planning operations at these clusters

will be discussed as observed at the production plant in later sections (Section 4.4).

4.3.2 Organizational Chart in a Manufactured Housing Plant Production

As mentioned, a manufactured housing production plant will be divided into different
clusters each managed by a supervisor who reports to the production manager or the plant
manager as shown in Figure 4.2. A production manager, quality manager, and purchasing
manager oversee the plant. These managers in turn report to the corporate office and

work in coordination with the sales department.

The corporate office and sales department work in coordination with the plant

management giving input about existing orders or forecast orders. The sales department

73

requests plant management to produce the required number of houses. With the given

target, the plant management meets and prepares a schedule to achieve the target.

l

 

 

President
G.M.

 

 

l

l

l

l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Quality Sales Materials Production
Assurance Manager Manager Manager
Manager
Quality Production
Inspectors Foreman (3)
(3)

 

 

 

 

 

 

 

Labor force / Maintenance

 

 

 

Figure 4.2 Organization Hierarchies in a Manufactured Housing Production Plant

The sales manager and his/her staff are responsible for maintaining relationship
with the retailers and developers. The sales department does the job of both forecasting
expected sale orders and obtaining orders from retailers, developers and individual
customers. A lot of these orders depend on the market condition. The economy plays a
big role in sales of houses, similar to site-built houses. Currently, manufactured housing
industry is seeing a fluctuation in demand and companies are facing a tough challenge in

the unstable market.

74

4.3.2.1 Hierarchical Planning

In any manufacturing organization, there are three levels of planning operations, namely,

long-range, intermediate-range, and short range planning (Chase et al 1998).

1. Long-range planning: This is generally performed annually, focusing on a
longer period of time, usually a year or more. Very few big manufactured housing
companies do such long-range planning.

2. Intermediate-range planning: This usually covers a period of 6 months to one
year. The planning ranges from quarterly, half-yearly to annually. Based on
forecasted sales, manufactured housing companies plan their operations either
half-yearly or annually.

3. Short-range planning: This covers the day-to-day up to six months of
operations. Manufactured housing plants usually perform daily, weekly, and

monthly planning of their operations.

Figure 4.3 shows the various decision-making processes at different levels based
on the forecasts of a firm’s production management team. The hierarchical procedure of
planning divides the decision-making, with top management allocating work among
different production plants based on the forecasts. At the same time the lower level of
management (production plant management) determines the production at the plant.
Involvement of top management in lower management’s detail decisions and lower

management’s involvement in top level planning is not advisable.

75

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Decision Level Decision Process Forecasts
Corporate Level . Allocate production ‘ Annual
among plants demand/forecasts
. Determines Monthly/Weekly
P t M
roduc ion anager r—> monthly/weekly <— demand
production
Determine/Manage Daily
Plant Manager —> dailyppggggtion <—— demand/target

 

 

 

 

 

 

 

 

 

 

 

 

Figure 4.3 Hierarchical Planning Process
(Modified from Chase et al 1998)

4.3.2.2 Production Planning System and Strategy

Any manufacturing system’s production planning includes determining the optimal
combination of production rate, the workforce level, and inventory on hand (Chase et al
1998). Production rate is the number of units completed per unit time, usually a day or
week. Workforce level is the number of workers needed for production. Inventory on

hand is the balance of unused inventory carried over from the previous period (Chase et

al 1998).

The form of production planning varies from company to company. In some
firms, the approach to plan production operations is derived from a long-term annual or

half-yearly plan. Based on the annual plan, the planning group determines how to meet

76

these requirements with the available resources. Available labor, material, and equipment
are compared to anticipated demands. If capacity is inadequate, additional requirements

for workers, equipment, and so forth are planned.

There are various internal and external factors that affect production planning in
the manufacturing industry. Figure 4.4 illustrates these factors that constitute the

production-planning environment (Chase et al 1998).

 

 

 

 

 

 

 

 

 

 

 

 

 

  

 

 

   

 

   

 

 

  

 

 

 

 

Competitors’ Raw material Market
behavior availability demand
External to
firm
Planning
External capacity for Economic
(e.g. Subcontractors) production conditions
. . . Internal to
'fl‘.”_..l_e!vel,s_ .. firm

1

>“" ”.1. . |

 

 

 

 

 

 

Figure 4.4 Required inputs to the Production Planning System
(Source: Chase et al 1998)

Though most of the external factors are not directly in control of the planning
group, the internal factors can be managed and controlled but with varying degrees. For
example, current physical capacity of a plant is usually fixed and cannot be easily
increased; work force size can be more flexible but with constraints; and inventory levels
tend to fluctuate. Chase et al (1998) suggest three production-planning strategies to

manage the internal factors in a production planning system. The three strategies

77

I

essentially involve a process of trade-offs among workforce size, working hours,

inventory, and backlog. The strategies are explained as follows:

1. Chase Strategy: ‘Hire and fire’ according to demand. Hiring workers when orders are
high and laying them off when orders are low is a strategy that depends on the
availability of a pool of easily trained applicants. It also depends on the extent of work
that can be performed by unskilled labor. Manufactured housing is one such industry that
uses unskilled labor and can implement this strategy. But, a drawback of this kind of
strategy is its motivational impact. The approach of workers would change based on

existing orders, due to the fear of being laid off as soon as work orders are completed.

2. Stable workforce-variable work hours: Vary the output by varying the number of
work hours to match production quantities with demand. Management can plan for work
through flexible work schedules or overtime. This kind of strategy provides workforce
continuity and a stable workforce that can be relied upon. It also avoids various costs of

hiring and firing associated with the Chase strategy.

The author found that the current manufactured housing industry appears to be
following this approach. Despite the low market demand, the plant visited by the author
had a stable workforce, working at flexible schedules to achieve the targeted output.
From observations at the production plant, it was typical for few workers to leave after
spending 5-6 hours as long as their production targets were met. Other workers stayed

beyond normal working hours of the factory in order to complete the work.

78

3. Level strategy: This strategy refers to producing a constant output rate with a stable
workforce. This may result in shortages and surpluses of inventory levels, causing order
backlogs and lost sales. Workers benefit through this strategy at the expense of poor
customer service levels and increased inventory costs (Chase et al 1998). This strategy
may also be a threat to the excess inventory levels, which may become obsolete over a

period of time. But it is known that Japanese companies have applied this successfully.

The reviewed literature, site visits, and anecdotal evidence indicate that the
manufactured housing industry uses one or a mix of these strategies. With only two
plants studied in this project, it is hard to make an assessment of the most used planning
strategy in the manufactured housing industry. However to understand the type of
planning involved in the operations at a manufactured housing production plant, it is
necessary to study the methods and processes involved in the daily operations. The next
sections provide a brief description of the planning operations of a production plant as

observed and studied by the author.

4.4 Production Planning Operations at a Manufactured Housing Production Plant

The manufactured housing production process consists of a set of interrelated tasks that
must be planned carefully to complete the product. As mentioned earlier, the
manufactured housing production plant consists of three different parts and three

different basic operations. The three different parts of a plant are:

1. Assembly line with main stations.

2. Subassembly stations.

79

3. Material storage spaces.

The assembly line runs throughout the production plant and also defines the shape
of the plant. Subassembly stations are those workstations that perform work and add to
the actual product on the main assembly line. Various material storage locations provide
raw material to production activities at the main assembly line and subassembly stations.
In addition, administrative offices are found inside the plant where most of the planning

for operations takes place.

At these three different parts in the manufactured housing plant, many activities
are performed to complete a manufactured house. Among the many different activities
performed in a manufactured housing plant there are three main activities that
manufacture the product. These three basic activities involved in a manufactured housing

production system are:

1. Main assembly activities or actual construction of the product.
2. Material handling and storage.

3. Workforce management.

According to Bernhardt (1980), the assembly activities are determined by and
arranged according to the nature of the product to be built. Assembly activities include
main assembly and subassembly activities. Handling and storage of materials is also an
important part of the production planning operations, as it is required to ensure

production runs are not interrupted. Last but not least, workforce management plays a

80

vital role in the plant production operations. If managed properly, a dedicated workforce

adds a value to the manufactured product.

Manufactured housing production plant operations will run incessantly when
these three basic activities are planned and organized properly. The in-house planning
team consisting of a production manager, quality manager, materials manager, and

foremen, plan and manage these three basic activities as described.

The foremen play an important role in the day-to-day operations on the floor of
the production plant. In the plant visited, there were five foremen responsible for plant
production. They manage the operations based on pre-divided work clusters in the plant.
The following section explains the operations at these clusters and the foremen planning

and management role.

4.4.1 Cluster Operations

As discussed in the previous sections, the production operations are categorized into
different clusters along the assembly line where a foreman supervises each cluster. The
clusters closely studied in the production plant are frame/floors, cabinets/interior and
exterior walls; roof external and internal finishes; and final finish and repairs. Frame and
floors sections are managed by foreman 1, interior and exterior walls sections are
supervised by foreman 2, roof stations, which include all the stations working on building
and finishing the roof of a unit by foreman 3, internal and external finishes by foreman 4,

and the final stations of finishes and repairs are supervised by foreman 5.

81

These foremen supervise all the operations at these stations with instructions from
the production manager. Workers at each station have to complete the assignments in
their normal duration according to the design and instructions from the foremen and
production manager. Production managers at each plant head the in-house production
operations and planning team. They are responsible for daily output of the plant and are

helped on this by input from the foremen about the status of the plant.

In plant A, the assembly line was organized and the operations are planned in
such a way that the unit has to move from one station to another every 20 minutes. Plant
B did not have this time target to achieve, hence the operations were more relaxed. At
plant B, each station had a fixed number of units to produce everyday and the target is
achieved by working at their own pace. During site observations, it was noticed that some
workers waited for long periods of time because the work was not released to them fi'om

upstream stations on the assembly line.
4.4.2 Daily and Weekly Operations

Information regarding the daily and weekly operations along the assembly line at
different stations was obtained by interviewing the plant personnel. As described in
Chapter 3, workflow and work requirements were performed using the Activity definition
Model (ADM). The ADM was used to ask questions during interviews with the plant
personnel. As shown in Figure 4.5, each station observed for data collection was

represented using ADM diagram.

82

   

   
    

Meets
Directives Criteria?
from

foreman

Yes
R elea se

      
 
    

  

 

 
 

Roof deck
complete

Roof Shingles

 
 

 

 

 

Shingles in

place

 

Figure 4.5 ADM diagram representation for Roof Shingles Station

Each plant has a team of production management that plans daily and weekly
operations of the plant. In the plants visited by the author, both plant A and plant B had a
similar operational planning but differed on execution. In both, management at the
corporate level decides on the target sales for the company and distributes the number of
units to be produced by each plant based on plant capacity. This is revised every month
or up to six months based on the size of the company. The target sales information is
communicated to a sales department in the production plant, which in turn meets with the

in-house production team to determine plant operations on a weekly basis.

83

After deciding the number of units to be built in a week the production manager
meets with the foremen and other managers in the plant every morning to plan the
operations for the day. The team meets again during the day to discuss progress and
status of the assembly line. All the foremen report any troubles in the assembly line.

Incase of serious problems an emergency meeting is held to find solutions.

At the individual station levels, foremen play a critical role in managing
production operations. Release of work from station to station depends mostly on the
planning and supervision of the foremen at their stations. The target output is usually
achieved by the end of the day, but at the same time there are various drawbacks in
production planning, which result in delays and non-completion of assignments. The
observed plant had many such issues that delayed the production and incomplete units.
Many of these aspects of planning failures will be discussed in later sections of this

chapter.

4.5 Summary

The previous sections provided an overview of the production planning involved in a
manufactured housing production plant and a step-by-step documented process of
planning involved in the initial setup of the production plant and then the daily
operational planning involved. The sections also document the detailed production
planning as based on observations and information gathered from the production plant
visited by the author. The process involved interaction with workers, foremen, and the

production manager.

84

4.6 Percent Plan Complete

The second objective of this research was to evaluate the production planning process
performance using lean construction principles. This objective was achieved in two
phases: measuring the quality of assignments by percent plan complete (PPC) and
assessing the performance of work crews using productivity ratings. As described earlier
in Chapter 3, percent plan complete is the percentage of number of assignments
completed to the total number of planned assignments during a given duration. The
following sections will describe the planning of station-level assignments for PPC, PPC

calculation, and finally PPC data and analysis.

4.6.1 Production Planning

Earlier sections in this chapter discussed production planning in a manufactured housing
plant. The following sections will discuss planning of daily production operations and
production assignments for each station, and how planned assignments are completed. In
addition, documentation of the current production planning undertaken by the
manufactured housing production plants to meet the uncertain future demands of the

customer is presented.

In general, production planning includes decision making on production and
inventory quantities. Any planning problem starts with a specific customer demand that is
to be met by the production plan. Most of the time future demand is either only partially

known or not known at all. This leaves any production planning team to rely on forecasts

85

of firture demand. Any forecast is inevitably inaccurate and thus the production planning

team needs to decide on how to account or react to demand fluctuations (Graves 1999).

Generic production planning in a manufacturing industry setting was discussed
earlier in this chapter. Three strategies of production planning were described in Section
4.3.2.2. The manufactured housing industry production planning follows one of these
strategies or a combination thereof. It is very difficult to generalize the kind of planning
strategy the manufactured housing industry follows, but this research makes an effort to

investigate the planning strategy of two manufactured housing production plants.

4.6.2 Production Planning at Two Plants

Investigating the method of production planning undertaken in the manufactured housing
industry was achieved by studying related literature, actual observation of production
operations, and interviewing plant production personnel. Both plant A and plant B had
similar planning strategies of production as represented in Figure 4.6, but the

implementation varied due to various reasons.

Plant A, was one of the four production plants the company operated and the
planning was performed in three levels. The first level of planning takes place at the
corporate level of the company. Input from the plant sales department and demand
forecasts targeted by the company set the target output of each plant every month. Second
level of planning is done at the individual production plant. Production operations are
planned every week and daily target of the assembly line is outlined. Production

personnel who include the plant production manager, quality manager, material manager

86

and foremen meet every week to decide on how many units to produce each day of the
week. The third level is daily planning including the process of material and labor
handling, supervising assembly operations and individual stations. Plant B had a very

similar production planning strategy as described here for plant A.

 

 

 

Time Production Planning Personnel
Planning of distributing Corporate planning
number of units to be team, Sales personnel,
Monthly produced by each Production Managers
production plant and Quality Managers

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Plant Production
Plant level production Manager, Quality
Weekly planning and Weekly Manager, Materials
operations Managers, Foremen
Planning and Supervising Foremen, Production
Daily Operations, Manager, Quality
Daily Material and Labor Manager, Materials
handling and Quality Manager
Inspections

 

 

 

 

 

 

 

 

 

 

 

Figure 4.6 Production Planning Strategies at Plant A and B

From the observations and interviews with plant production personnel in plant B,
it was understood that the production planning is not really performed inside the plant.
Many reasons were given for this kind of unstructured production planning. Uncertain
market demand was often cited. Manufacturers are reacting to such imbalance in market

demand by exercising tight control on production and inventory levels. Inputs from

87

production personnel and author’s observations at the plant, were again confirming the
need that inspired this research, which was to increase the performance of the production

planning.

As described earlier, to measure the production planning performance of a
manufactured housing plant the first step was to calculate PPC and the next step was to
assess the worker crew performance. To measure PPC, the production operations were
observed by videotaping and also input from the production plant manager and foremen

were used.

4.6.3 Percent Plan Complete Calculations

To calculate the percent plan complete (PPC), it was required to collect data on the
number of assignments the production team plans for a period of time. Results from
interviews revealed that the plant does not have a fixed number of assignments per day or
per hour. The only assignment for the production crew of the plant was to achieve the
target output for the day. To calculate PPC the inputs required are the total number of

assignments planned and number of assignments completed. As explained in Chapter 3,

PPC = (Number of assignments completed/N umber of assignments planned) X 100

The first input for PPC was the total number of assignments planned for a day
based on the assignment made by the weekly planning team. Work by the author at the
plant included getting information on the daily production activity from the production
manager and the foremen. The observation started everyday from the time assembly line

starts production to the end of production. At the end of the day’s shift, data was

88

collected from the foremen on the number of assignments completed, which was

basically the number of units completed.

To have valid data, the criterion for deciding completed and incomplete
assignments was based on information from the foremen and author’s observations. The

criteria of non-completion of assignments was the following:

1. If the unit was not complete in scheduled duration at its station, it is
considered incomplete.

2. Incomplete work at the end of the day. Only completely finished units were
considered as completed assignments.

3. Repair works or final finishes performed on units outside the assembly line

were also treated as incomplete assignments.

For example, assignments at the Exterior boards station were to fix the exterior
boards to the walls and cut out openings for the doors and windows. If the work was not
complete at the station before the section moved to the next station, it was considered as
incomplete. Also, if the workers at the next station are waiting for the work to be done at
the exterior boards station or help in completing the work rather than waiting for the
work, it was considered incomplete. Finally, if there was any repair work or reinstalling

the boards after it moved from the station, it was taken as incomplete assignment.

The data of incomplete assignments was obtained from the daylong observations
and interviews with foremen and workers at each station. Due to the scope of work, a sub

sect of stations in the manufactured housing production plant were considered for

89

calculating PPC. Stations considered for PPC were randomly chosen based on the

convenience of observation and nature of work.

Some stations like carpeting and interior works of the house were not considered,
as it became too difficult to observe the activity inside an enclosed housing unit without
distracting the crew working at the time. Thus, most of the stations observed were
stations like roof truss, roof activity, wall stations, doors and windows, and external
finishes. The following section describes the stations considered followed by the data

collected from those stations and the PPC calculations performed.

4.6.4 PPC Data and Calculations

While considering all the stations in the plant would have been ideal, the research scope
allowed only for a few stations to be observed. Various activities in these stations were
observed for calculating PPC. The major activities observed and their corresponding
stations are listed in Table 4.1. PPC was calculated for the observed activities and the
PPC for the entire station was calculated by taking the average PPC of all activities

belonging to the station.

Table 4.1 Manufactured housing stations and activities at each station

 

 

 

Station Activity
. Floor joist attached to the chassis, and initial rigid
l. Chassrs . . .
insulation rs done
Floor joist is covered with decking and vinyl flooring
2. Floor decking is completed. At this station the unit is fixed with an air

bladder system for moving along the assembly line

 

 

 

 

9O

 

3. Plumbing and HVAC

Rough plumbing, HVAC installation and sanitary

appliances are fixed

 

4. Internal walls

Fabricated internal walls are placed on the main station

 

5. Cabinets / Fixtures

Cabinets and fixtures are fabricated at a subassembly

station and then installed at the internal walls station

 

6. External walls

External walls are placed and rigid insulation placed.
Rough electrical and plumbing work is also done at

this station

 

7. Exterior boards

External wall boards are fixed to the walls, openings

for doors and windows are cut out in the walls

 

8. Roof truss building

A subassembly station outside the main assembly line

where the roof truss is built in two halves

 

9. Exterior doors and

windows

Doors and windows are placed at this station. At the
same time roof truss is placed on the house and roof

insulation is done

 

10. External siding / Roof
sheathing

External siding, external electric fixtures like lamps are
installed. Simultaneously roof sheathing is fixed over

the truss

 

11. External finishes

Other external finishes are also carried out

simultaneously at this station

 

12. Roof shingles

Roof is completed with insulation paper and shingles

placement

 

13. Interior finish

Interior finishes like painting and wall finishes are

done at this station

 

14. Final finishes

Carpeting, appliances and final electric fixtures are

installed

 

15. Attaching wheels

The unit is removed from the air bladder system and

wheels are fixed for transportation

 

 

16. Final inspection and

repairs

 

This station is used for final inspections and essential

repair works

 

91

 

Different activities that have been considered for calculating PPC are:

8.

9.

. Internal walls — Station 4

Exterior walls — Station 6
Exterior boards — Station 7
Doors and windows —- Station 9
External siding — Station 10
Roof truss building — Station 8
Paint ceiling - Station 8

Roof board — Station 10

Roof shingles - Station 12

10. External finish — Station 11

Data collection was done for each activity/assignment for 5 days over a period of
two weeks. Then PPC of different activities/assignments were calculated. Table 4.2
shows an example of data collection and PPC results for the roof truss station. The roof
truss activity, built at a subassembly station, was one of the most critical activities in the
entire production process. Usually built by two workers, the activity crew is sometimes
assisted by workers from other stations when needed. Usually workers helping the roof
truss crew are from the next station downstream of the production line, which is paint

ceiling. If the assignment is not completed in the scheduled duration the workers at the

next station wait for the work to be released.

92

Table 4.2: PPC of roof truss building assignment/station crew

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ifictivity: Roof Truss building
ays Observed: 5
Crew size: 2
Average
Days Planned Completed Crew PPC Reasons/Delays
Crewmember 3 1 5
Day 1 I ' 50% Big truss/No proper
Crewmember 3 1 5 directives
II '
Crewmember
I 3 2 Completion of previous
0
Day 2 Crewmember 3 2 “'60 A) days work
11
Crewnlrember 3 2. 5
Day 3 Cr wmember 83.30% Material shortage
e 11 3 2.5
Crewrrliember 3 3
0
Day 4 Crewmember 100 /°
II 3 3
CrewnIrember 3 2
Day 5 C b 66.60% Directives
rewrplem er 3 2

 

Figure 4.7 shows the bar graph of PPC for roof truss building station collected for

a period of 5 days. This shows different variations in PPC, which indicates that the

production planning is not very reliable. Only on day 4, it was 100% i.e., all assignments

were completed as planned.

93

 

 

100%

PPC

20%

 

60% ‘

40% a

0%;

l 83.30%
80% 4

 

' 66.60%

50%

‘J- —-— r run-r.

 

 

 

 

 

 

 

 

1 00%

 

 

 

iw—wlw

l

 

 

 

Day

 

66.60%

t
F, .1

 

 

 

 

Internal walls is a subassembly station with a crew of two. Again, the activity
could not achieve 100% PPC everyday due to various reasons. Most of the reasons were
due to material shortage or inefficiency of the crew. Table 4.3 presents the PPC results
for this activity. Followed by the PPC data, Figure 4.8 shows the line graph of PPC for

the internal walls station. The graph format was changed from Figure 4.7 to avoid

Figure 4.7 Percent plan complete for roof truss station crew

possible confusion to the reader.

94

 

Table 4.3: PPC of Internal walls assignment/station crew

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Activity: Interior walls
lDays Observed: 5
Crew size: 2
Average Reasons/Delays
Days Planned Completed Crew PP C
meI 3 2
Day 1 66.60% Inefficiency of crew members
Crew 11 3 2
mel 3 2 -
Day 2 6 6. 60% belaayks due to material and
Crew 11 3 2 mist CS
Crew I 3 3
Day 3 100%
Crew 11 3 3
mel 4 3
Day 4 75% lMaterial shortage
Crew 11 4 3
Crew I 3 2.5
Day 5 83.3% Rework (Mistake by crew)
Crew II 3 2.5
100% —.

 
      
 
 

l 00%

 

83.30%

 

80% *-

 

 

 

I 66.60%

1 60% r
I E
I n.
I 40% j,
l
1 20% i

0% ~ —e~ --—— .,,_____,._ — a —_——__- _.__,L We- ,,___ —_ . , «a
l 0 l 2 3 4 5 6 |

 

 

Figure 4.8 PPC for Internal walls station

95

Another activity observed was exterior walls station with a crew of 3. This
activity was spread over two stations, a subassembly station and the other on the main
assembly line. The walls are manufactured at the subassembly station and then lifted to
the unit on the main assembly line and fixed in place. The main problem in this work was
difficulty in fixing the prefabricated wall on the edges of the unit. A lot of rework was

done due to improper fixing and nailing of the wall.

Table 4.4 PPC for External Walls assignment/station crew

 

ays Observed: 5

Fetivity: External walls
Crew size: 3

 

Average Crew

Days Planned Completed PPC Reasons/Delays

 

Crewmember I

 

Day 1 Crewmember II 100%

 

Crewmember III

 

Crewmember I

 

Day 2 Crewmember 11 66.60% 'Material shortage

 

NNNWUJ

Crewmember III

 

2.5
2.5 83.3%
2.5

Crewmember I

 

Waiting for the unit

Day 3 Crewmember II (Directives)

 

Crewmember III

 

Crewmember I

 

Delays due to

O
66'6 /° mistakes

Day 4 Crewmember II

 

Crewmember III

 

Crewmember I

 

Day 5 Crewmember II 66.6% Directives

 

wwwwuwwwwwwwwuw

NNNNNN

Crewmember III

 

 

 

 

 

 

 

 

96

The other activities considered for PPC are provided in Appendices. The
following Figures 4.9 - 4.12 represent the PPC graph charts of all activities for which

data was collected.

i L r» ,;_# -_______ _j

100% ~
80%
60%
40%
20%
0% 1 ~ . . ,1

1 1 2 3 4 5
L Day

Figure 4.9 Percent plan complete of exterior boards assignment/station crew

 

PPC

 

 

 

 

 

PPC

 

 

 

Figure 4.10 Percent plan complete of doors and windows installation station crew

97

Both assignments shown in Figure 4.9 & 4.10 are continuous in the assembly line
and had only one crewmember assigned. Work did not start at doors and windows station

until the exterior boards assignment was complete.

The effect of unreliable workflow can be clearly observed in this case in Figure
4.9. On day 4, the work was not released from exterior boards station and as a result the

PPC for doors and windows is just 50% on day 4.

Figure 4.11 is the line graph showing the PPC values of two of the activities,
namely, external siding and external finishes. These two jobs again had one crew each

and the workflow affected both stations.

 

 

 
  
    

 

 

l 100% 100%
100% 1
100.00%
30% ‘ i
‘ 66.60% 1
l 9 6M 1 66.60%

Q:

l “- l
: 40°/o '
l i
' 20% l I
0°/o t T, 7 _ _ i _ * 1 ____fi ;
Day 1
I i w- _tEX‘°rP§LSEing -_ L __":Externai Finish 1

_i

 

 

 

l .___ _-___

Figure 4.11 PPC for exterior siding and external finish assignment/station crews
The remaining three activities observed were paint ceiling, roof board, and roof

shingles. Figure 4.12 shows the PPC for these three stations in the assembly line. Paint

98

ceiling is a subassembly station while roof board and roof shingles are carried on main

stations along the assembly line.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

100% l ,
80% 1
U 60% ~
2
40% ~
20% ~
0% “I
1 2 3 4 5
ISeriesl_* 33.30% ; 7 66.60% 100% 75% , _____6_6.60%
ISeriesZ 66.60% _ 100.00fé 100% 753/0 _»_100.00%.
DSeries3 66.60% 100.00% 1 100% 75% 100.00%
" Day
Paint ceiling I Roof board E] Roof shingles

 

 

 

 

 

 

 

 

Figure 4.12 PPC for paint ceiling, roof board, and shingles assignment/station crews

4.7 Summary of PPC

The above sections presented the detail data collected for PPC and its calculations. The
PPC data was collected from the foremen in charge of the above assignments for each of
the crews. PPC presented above is the average crew PPC. Figure 4.13 summarizes PPC at
all the observed stations in the plant. Data collected for PPC is not a very large sample

and this research just makes an effort to provide a template for further investigation in

production planning.

99

 

09 . 0.83
l: 0.78 076 081 I
‘2 05 ‘
lfl"
. l
01+ ‘

\% <9 *9 go
04:9 43 for by @680 (9&9 6‘" $80 0&8? . f N
'9 ‘6 K 2’ ~o x9
{‘0‘ '00" .08 a“ q} '0‘ 0° .&° 0‘» GS“ '
8 .. 5- 8° l
‘ Q) Q33? ‘9 Q1? Q)
I Q Observed stations ‘

Figure 4.13 Average crew PPC of all observed stations
The PPC data will be related to the labor utilization factor (LUF) data collected to
study the productivity ratings of the crewmembers performing each of the above
assignments at their respective stations. The following sections present the productivity
ratings sample and labor utilization factor data for the above observed activities by the

different crews.

100

4.8 Labor Utilization Factor

Labor utilization factor calculation was described in Chapter 3. To restate, LUF is
calculated from the productivity ratings that complement PPC data so that both provide a
better insight to the production planning operations in a manufactured housing plant. By
definition, labor-utilization factor is the percentage obtained by summing the number of
observations of effective work and one-fourth the number of observations of contributory

work, and the sum divided by the total number of observations. In equation format, LUF

is given by Effective work + ‘/4 essential contributory work

 

Total observed
Labor-utilization factor =

Total observed = Effective + Essential contributory + Ineffective

Based on the criterion explained in the previous chapter, the productivity ratings
of the work performed were calculated similar to the following example. For better
understanding, the same example of Exterior board station explained for PPC is taken.

The example of productivity ratings for the crewmember at this station is:

Effective work: Lifting the board, nailing of the board to the external wall studs,

cutting openings for doors and windows.

Essential contributory work: Walking a distance to get the board.

Ineflective work: Waiting for the external walls station to complete the work,

waiting for the exterior boards to be delivered to the feeding station.

101

As described in the earlier sections, each station has a different crew size based on
the kind of activity taking place at that station. Some activities, which cannot take place
at the main station, are undertaken at a subassembly station like the roof truss building.
The observations were performed by video recording the activity of crewmembers at each
station. Each crewmember’s activity during the duration of the work assignment was
categorized into efficient, essential contributory and non-productive work, which
facilitated the calculation of LUF. For a crew of more than one, the average LUF was
calculated for the entire crew. Finally, the LUF data was compared to the PPC result

collected.

The next section gives a brief overview about the working environment and the
work method used by the working of the labor force in the visited manufactured housing
production plant. The productivity study sampling and LUF results calculated will follow

this section.

4.8.1 Labor Force

A typical manufactured housing plant employs workers from areas local to the plant.
These workers are usually educated at high school level and some are high school
dropouts. Located in an Amish populated region in the northern part of Indiana, most of
Plant A workers were Amish. Plant B, which was located in a bigger town hired mostly

high school dropouts and, hence, the work practices were very different from plant A.

The manufactured housing industry has not attracted many construction workers

due to its nature of work and low pay rate. Most of the work carried out in building a

102

manufactured house does not require high skills. And being an affordable option for
housing, the costs of production are kept low and, hence, the pay rates were also not very
high. From the author’s observations at both production plants and interviews with
production managers, the work culture was clearly different and distinguishable.
Videotaping was not permitted at Plant A because of worker preference there. Production
operations at plant B were video taped and then reviewed to categorize each

crewmember’s work activity using productivity ratings.

4.8.2 Productivity rating study

In a productivity rating study, performance of a crew or crewmember is rated to assess
the amount of productive work done while the assignments are carried out. The results
are usually stated as a percentage of productive to non-productive work. For construction
type work, a minimum sample size of 384 observations with a combination of confidence
level of 95 percent and limit of error of 5 percent is satisfactory. Choices of a detailed

sampling size, confidence level and limit of error were explained in Chapter 3.

In this research, the observations were made for each crewmember while
performing an activity. Eleven different assignments at different stations were observed
and video taped. The total observation period of each crewmember varied from 35
minutes to one hour based on the continuity of the activity. An individual observation of
the productivity status of a worker can be made every 1-5 seconds. In this research, 5
seconds was chosen as the observation interval. Hence, to get 384 observations for each
crewmember, a minimum time period of 1920 (5 X 384) seconds or 32 minutes was

needed.

103

Video recording was selected as the method of observation of the work
performance of a crew, because it was easy to review the videotape later and observe the
continuous performance of the worker. It also helped in video recording two or more
workers in a crew at the same time and hence saved the time of actual observation in the

production plant.

The plant studied, functioned from 6AM to 2PM, 5 days a week. Observing a
sample of 384 only one time a day would not give a representative sample. So,
productivity ratings were performed once in the morning and once again in the afternoon
for each crewmember. Thus each crewmember of a particular assignment was observed a
minimum of 384 times in the morning and again a minimum of 384 times of 5 seconds
duration in the aftemoon. The average observations from both periods were used to
determine the LUF for each crew. Most of the activities observed for productivity ratings

are the same as those made for PPC, and were as follows:

_

. Plumbing and HVAC — Station 3
2. Interior walls — Station 4

3. Exterior walls — Station 6

4. Roof truss building — Station 8
5. Paint ceiling - Station 8

6. Roof setting —— Station 9

7. Exterior boards — Station 7

8. Doors and windows — Station 9

9. External siding — Station 10

104

10. External finish — Station 11

11. Roof boards — Station 10

12. Shingles — Station 12

LUF data for the above activities was calculated by taking the productivity ratings
observations in the AM and PM on a day when PPC was collected for the same
assignment. It should be noted that PPC for a few assignments was not available and just

LUF was calculated. The following section presents detail data collected for calculations

of the LUF.

4.8.3 Labor Utilization Factor Data

This part of the data collection involved 5 days of video recording and more than 72
hours of watching the tapes to observe the performance of the crewmembers. There were
approximately 19 X 384 X 2 = 14592 observations at 5 seconds intervals, i.e., nearly 21
hours of continuous observation and data analysis. Twelve assignments with different

crew sizes were observed for collecting the data. The results of each crewmember’s LUF

is calculated and shown in Table 4.5.

 

 

 

 

 

 

 

 

 

Station Crewmember LUF Station Crewmember LUF
[External Siding I 0.61 Exterior Boards I 0.61
[External Siding II 0.54 Exterior Boards 11 0.4
External Finish I 0.49 Roof Setting (Crew 1) I 0.57
IExtemal Finish 11 0.49 oof Setting (Crew 1) II 0.57

 

105

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

[Paint Ceiling I 0.57 [Roof Truss (Crew 1) I 0.43

aint Ceiling II 0.47 lRoof Truss (Crew 2) I 0.4
[Plumbing (Crew 1) I 0.61 [Roof Truss (Crew 1) II 0.32
[Plumbing (Crew 2) I 0.62 [Roof Truss (Crew 2) II 0.44
[Plumbing (Crew 1) II 0.51 Doors and Windows I 0.43

lumbing (Crew 2) II 0.41 Doors and Windows 11 0.45
IInterior Walls (Crew 1) I 0.6 Shingles (Crew 1) 0.68
IInterior Walls (Crew 2) I 0.54 Shingles (Crew 2) 0.65
Ilnterior Walls (Crew 1) II 0.44 Shingles (Crew I) II 0.55
Interior Walls (Crew 2) II 0.45 Shingles (Crew 2) II 0.49
Exterior Walls (Crew 1) I 0.78 lRoof boards (Crew 1) I 0.68

xterior Walls (Crew 2) I 0.52 [Roof boards (Crew 2) I 0.59
Exterior Walls (Crew 1) II 0.24 (Roof boards (Crew 1) II 0.70

xterior Walls (Crew 2) II 0.49 WROOf boards (Crew 2) II 0.58

 

Table 4.5 AM and PM LUF of individual crewmembers at the observed stations

Data in the above Table 4.5 shows the AM LUF and PM LUF of each
crewmember at 12 different stations. AM LUF is the morning observation and PM LUF
is the afiemoon observation. The observations were done in the morning and afternoon to
get more representative data. This would also help observe the work pattern and
productivity based on time. It can be noticed from the above figures in Table 4.5 that, the
AM LUF is higher most of the times than PM LUF. It clearly shows that the productivity

is low towards the end of working hours in a shift. The average of AM LUF and PM

106

LUF, and average of the entire crew are calculated for at each station and are represented

in the bar graph of Figure 4.14.

 

LUF

 

 

 

 

 

 

 

s
s- ‘ b ‘9 99°
a. $6 4
Q
# 7 Station crews» ._ _

Figure 4.14 Average LUF of all observed station crews
4.9 Relation between LUF and PPC

A part of this research was to investigate the relation between LUF and PPC. An
argument for the need to find out the relation between LUF and PPC was made in
Chapter 3. From the data collected, it can be observed that percent plan complete of the
planned assignments was high, around 70%. This figure was based on the method of
calculating PPC, but actual completion was 100% most of the time, because though the

assignments were not completed at the assigned station and during the time observed, all

107

the work was completed by the end of the day. But, according to the criteria we defined

earlier this is still not complete.

From the observations as shown in Figure 4.15, it was seen that most of the
assignments planned were completed by the end of the day either at a different station or
outside the assembly line. It was also observed that some crew spent more than the
normal working hours to complete the job. Hence it can be said that PPC was 100%. But,
at the same time when PPC was 100%, it was observed that LUF of the crew was just
about 50%. From this basic analysis, it can be concluded that despite completing the daily
target production, the crews were hardly utilized at 50%. The average LUF of all stations

observed is 0.52 and the average PPC of total number of observed stations is 0.78.

If ,7

01 7

5.06 .
:05
‘gm
140.3
‘00 02!

 

 

\ ' \
‘ ‘b (b é o . Q \\ . 90
665‘ .064“ {0 $89 a}? Q05“ 056° , $.00 (39$
'6’ x9 '00 6 6° 0 ‘5' ‘9
\9 ‘6" {660 .29 +‘ ‘69 Q Q-
o 7 r 7,7 “ * ”i ‘
0° LI: Labor Utilization Factor [:1 Percent Plan Complete

Figure 4.15 LUF vs. PPC for all the observed stations

108

4.10 Coefficient of Determination

In order to observe the correlation between Labor Utilization Factor (LUF) and Percent
Plan Complete (PPC), data observed on a single day for LUF and PPC are plotted in
Figure 4.16 as a scatter plot. The data points taken to plot Figure 4.16 are listed in Table
4.6. Regression analysis is performed using this graph with PPC on X-axis and LUF on
the Y-axis of the graph. The intention of using regression here is primarily for description
and not forecasting of one variable from another. Performing Regression analysis resulted
in a Correlation Coefficient (r) of 25%. Also, below 80% PPC it appears there is no
difference in LUF. In the region above 80% PPC, the LUF figures slightly increase with

increasing PPC.

 

 

 

 

 

 

 

 

 

 

 

 

 

Stations PPC LUF
Interior Walls 0.83 0.5
Exterior Walls 0.66 0.5

Exterior Boards 1 0.53
Doors & Windows 1 0.44
External Siding 0.75 0.57
External Finish 0.33 0.49
Roof truss 0.5 0.39
Paint Ceiling 0.66 0.52
Roof Boards 0.66 0.63

Shingles 1 0.59

 

 

109

Table 4.6 PPC & LUF data of a single day for observing Correlation between LUF & PPC

 

 

 

 

 

 

 

 

 

 

1
2 .
R —O'0603 —Linear (LUF&PPC) 1
1 .
0.9 1
0.8 r
0.7 -
ta. 0.6 ‘ ° 0 o
3 0.5 . e— W
0.4 « o
0.3 a
0.2 r
0.1 1
O t W, x 1. r A~——.
; 0 0.2 0.4 0.6 0.8 l 1.2
' PPC

 

 

 

 

Figure 4.16 Scattered plot of LUF vs. PPC for all the observed stations

The interpretation of r2 as the proportion of the total variation of Y explained by
X is frequently taken literally. However, a regression model does not imply that Y
necessarily depends on X in a “casual” or “explanatory” sense. Moreover, a value of r2
close to 1 is not an indication of perfect inference of Y from X using the estimated
regression model. The usefulness of a regression model in predicting Y from X depends
upon the width of the confidence or prediction intervals. Neter et a1. (1990) emphasized

that:

“Regression models do not contain any parameter to be estimated by r
and [2. These coefficients simply are descriptive measures of the degree
of linear association between X and Y in the sample observations which

may, or may not, be useful in any one instance.”

110

P-value: In order to check the statistical significance of the test to investigate the relation
between LUF and PPC, p-value is calculated. P-value or attained significance level
associated with the test is a statistical quantity that represents the smallest value of a for
which the null hypothesis is rejected. Often (it-values of 0.05 or 0.01 are used to attain the
significance level. If p-value is less than or equal to the considered a-value, then the null
hypothesis is rejected. Otherwise, if the p-value is greater than the a-value, then the null

hypothesis cannot be rejected.

This test is done considering a null hypothesis that LUF and PPC are related.

Denoting PPC values by x and LUF values by y, we calculate r.

ngxiyi -§X1;Yi

11".;121121211

r = 0.2456

 

Using the formula, r =

 

As mentioned above, null hypothesis is that PPC and LUF are related, or
assuming (PPC, LUF) has a bivariate normal distribution, H0: p = 0 versus Ha: p at 0 is

tested. The value of the test statistic is,

(tilt—1104”“)

1" ,00
Usin ormula, z =
gf l/Vn — 3

 

 

z = 0.64

111

Because this is a two-tailed test, the a = 0.025 is the probability that ank will

fall in the rejection region of H0, The rejection region for H0 is when Zealc > Z (a/Z).

Z(a/2)= Z(0.025)= 1.96
Zealc 1' 0.64
Zea.C < Z (an), Hence Ho cannot be rejected.

P-value = 2*P (Z > 0.64) = 2*0.2611 = 0.52 > (1.

Hence, for any value of alpha less than 0.5, the null hypothesis cannot be

rejected. This strongly suggests that there is no correlation between LUF and PPC.

In this study, data was not sufficient to judge the true nature of the correlation
between PPC and LUF. In addition, the limited sample size prevented an outlier analysis
so that relevant data points are used. Also, the PPC and LUF measurements at each
station could have contributed to the nature of data collected. Criteria for PPC for
different assignments changed from station to station depending on the type of work
performed and assigned. Additionally, between the two variables, LUF and PPC,
requirements for collecting data for PPC were more relaxed than the requirements for
LUF. Data collected for LUF was more accurate and extensive than the data collected for

PPC.

With more data, and stricter requirements for PPC, it is possible that a higher
correlation will be manifested. However, the results of this study do indicate that the
correlation between PPC and LUF is not strong. On the one hand, a high PPC does not

necessarily mean a high LUF, i.e., that the crew is efficiently utilized. On the other hand,

112

it appears that a low PPC is associated with a Low LUF. This indicates that it is important
to consider both PPC and LUF when performance of production planning is considered.

LUF adds an important dimension to the analysis as this study indicates.

4.11 Summary

This chapter has detailed out the data collection and research results. It was an
arduous effort to visit the production plants and stay at the plant to observe the
data from the start to close of the production. The objectives stated in Chapter 1
were achieved using the methods and techniques in Chapter 3. The last phase of
this research, Chapter 5 will discuss and analyze the results. It also discusses
Pareto analysis and feedback from the industry on the results obtained and uses
the fish bone technique to find the constraints. The chapter finally concludes the

research with future areas of research.

113

Chapter 5

Discussion and Conclusion

114

5.1 INTRODUCTION

Chapter 1 introduced the topic of this research, its goal, and objectives. In Chapter 2,
literature related and necessary for this research was reviewed. Tools and methods to
achieve the set objectives were explained in Chapter 3, and finally Chapter 4 reported on
the data collection and data analysis. This Chapter will conclude the research with the

results, conclusion, contributions, and future areas of research.

5.2 CONSTRAINT ANALYSIS

In the previous chapter, the planning process was documented, and results of PPC and
LUF calculations were presented. As explained before the results were not 100% at all
times for both PPC and LUF. This shows that the production planning can be improved.
There could be many reasons for under performance by the production planning systems,
and to explore this, constraint analysis was performed simultaneously during the data
collection process. The constraint analysis technique used in this research was explained
in Chapter 3, but to repeat, it is a technique to identify all the reasons for non-completion

and delay of the planned assignments.

Constraint analysis, as used here, includes three steps. First, the non-completed
assignments are identified from the PPC data. Second, the most frequently occurring
problems are identified. Third, the reasons for such problems are determined. Finally,
improvement opportunities are suggested. From the data collected for PPC and the
analysis of results in Chapter 4, all the non-completed and delayed assignments were

identified and are illustrated in Table 5.1. This data indicates that, on average, 22% of

115

planned operations were not completed. For every activity, the reason for non-completion
or delay in completion was noted by asking the concerned personnel of that assignment,

mostly the workers and foremen.

Table 5.1 Percent of non-completed assignments

 

 

 

 

 

 

 

 

 

 

 

Station Percent NOT complete
(%)

Interior walls 22
Exterior walls 24
Exterior boards 17
Doors and windows 17
External siding 19
External finish 27
Roof truss 27
Paint ceiling 32
Roof boards 19
Shingles 12

 

 

 

 

The reasons indicated by workers and foremen for completing assignments as
planned were slow release of work, directives from production personnel, material

shortage, planning failures, and repairs. These problems are further explained as follows:

1. Release of work: In an assembly line production, release of work from the
upstream is extremely critical for maintaining work flow. Release of work in
proper fashion as planned keeps the production line running and production

smooth. Any delay in release of work causes downstream work to stop and wait.

116

 

This is one of the major problems in an assembly line production and was
observed in the manufactured housing production plant under study.

Planning failure: The planning of operations at each level and at each station on
the assembly line becomes very important and helps the production run
efficiently.

Directives: Execution of planned operations is as important as planning the
operations. Production operations directives and decisions taken at critical stages
of production are quite important in proper production of the product. The
planning personnel like foremen and production manager play a key role in
directing the operations and giving directives.

Material shortage: Material shortage seems like an easy problem to handle, but
the line progress is held hostage until the problem is resolved. According to the
plant production manager, less inventory and Just-in-time are lean production
ideals but have to be carefully used and managed.

Rework: This problem results due to various failures. It could be a crewmember’s
mistake, production planning defect, or any other unknown reason. The particular
production plant studied had a separate station for repairs at the end of the
assembly line. All reworks and repairs were performed at this station. If there are
too many units that need rework or repairs, they are kept out of the plant, and then
brought back in to the plant to complete the work. This is a sign of mass
production.

Other problems: There are few other problems, which are not very frequent and

unexpected, like changes in design, inspection agency changes, and so on.

117

Table 5.2 list all these problems and their frequency of occurrence (Number of

assignments they occurred on)

Table 5.2: Problems in production operations of a MH production plant

 

 

 

 

 

 

 

 

 

 

 

Problem Number of assignments
Work from upstream not released 14

Directives for operations 7

Planning failure 7

Material Failure 7
Rework/Repairs 4

Pthers 2

5.2.1 PARETO ANALYSIS

Pareto analysis is a technique that identifies the ‘vital few’ factors that are responsible for
the majority of problems. The data in Table 5.2 was used to create a bar chart to indicate
such ‘vital few’ and ‘trivial many’ factors that affect the proper functioning of the
production process in a manufactured housing production plant. Figure 5.1 depicts the

Pareto analysis for the plant under study.

Based on the Pareto analysis bar chart in Figure 5.1, it can be clearly observed
that there are a vital few factors that cause the majority of the problems in the production
process of the manufactured housing plant under study. The next step was to find out the
actual assignable causes of these problems using cause-and-effect diagrams as explained

in Chapter 3, section 3.5.3. The next section in this chapter explains the different causes

118

and traces the roots of various problems that occurred during the time of data

observations and collection.

1

 

 

Magnitude of occurence

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1 Factos causing problems in production
1 I Work from upstream not released I Directives for operations

El Planning failure El Material Failure

 

. I Rework/Repairs 3 Others

1__ __ __ ___

 

 

 

 

 

 

Figure 5.1 Pareto analysis chart depicting the factors responsible for problems

5.2.2 FISH BONE TECHNIQUE

Cause-and-effect diagrams commonly known as Fishbone diagram or an Ishikawa
diagram is a tool used to clarify the causes of a problem (Markland et a1 1998). The
various problems observed and gathered during the data collection were listed in the
above sections and the fishbone diagrams were constructed with the production personnel
at the plant to assign the root causes for such problems. Based on this process, the

following root causes were assigned for the problems observed:

119

1. Labor Issues: One of the plant’s major problems is the worker issue. In Chapter 4, a

section described the kind of labor force that usually works in manufactured housing

industries. The various root causes of the labor problems were:

a.

Slow work: Most of the workers working in the plant are slow when compared to
the other plant observed for this research project. The plant manager and foremen
also corroborated this.

Skill: Basically the workers employed in the manufactured housing industry are
not very skilled, but some trades require a little skill in performing the work. This
was lacking in some of the trades and caused many problems in the production.
Work ethics: The production plant manager felt that work ethics have changed
over the years. The workers move around slowly and are not motivated to
complete the tasks. Worker abstinence was also observed and thus caused
problems for the management in planning the work.

Uneducated work force: The manufactured housing industry employs a large
number of uneducated labor forces to cut down costs on production. But, this
causes various problems in the production such as workers inability to read the

drawings, inaccurately taking measurements, and other similar problems.

2. Material: Material supply is a big issue in any production environment and so is the

case in manufactured housing plants. According to workers, foremen, and management,

material supply to the plants is an issue, which causes problems in the production. The

management also cited the lack of a systematic way to track material usage.

120

From the observations, it was seen that much of the plant space was being
occupied by material not used for days. The author feels that imbalance in production and
improper production planning was one of the reasons for material shortages. If planning
is done well and the material is handled properly, using less inventory and JIT techniques

would be a good option in a manufactured housing industry.

3. Methods and directives: Methods of production and directives of the production
process play a vital role in a smooth and structured production flow. Most of the
problems, the ‘vital few’ were caused due to the lack of directives, and uneven workflow

from one station to the next.

Design issues of a manufactured house create many problems. For example it was
observed that a big roof truss was to built, and the design of the truss was not detailed and
thus the workers faced lot of problem building it. Because of delay in making the unit at

that station, the work was delayed on the entire assembly line.

One of the main principles of lean production is to structured workflow such that
the work is released from upstream to downstream at proper intervals to avoid stoppages
in production flow. During the observations, it was noticed that large delays in work at
one particular station were affecting the work at the remaining stations along the

assembly line.

4. Equipment: Most of the equipment used in the manufactured housing industry is
sophisticated and work is made much easy with such equipment. During the observations,

equipment was not really a big problem in completing the assignments. A few concerns

121

of the workers regarding the equipment in the plant did not really affect the production

process.

One of the concerns, which needed to be addressed, was the airlift system, used to
transport units from one station to the next. This system did not allow smooth movement
of the unit and it also did not allow work underneath the house. In another instance, a
failure in a crane, dedicated to lift trusses, delayed the subsequent operations

considerably.

5. Other issues: The other issues, which caused the problems in production and
completions of work as, planned were: Sales problem, market demand, and government
regulations. Due to the low market demand and other external problems, it was a tough
job for the sales department to make sales. In order to achieve make sales, the sales
department would sell the houses with special designs and other changes according to the
customer choice and also to attract the customers. This created problems for the
production planning team and the production crew to manufacture such units and hence
slowed down the production. Government regulations, which constantly changed and
clearance of quality inspection by the external inspection agency, were some of the other

trivial few problems affecting production.

Overall, the production planning should be revisited and planned carefully
without repeating the mistakes that were observed and identified during this research
project. The data in this research project can be used to identify opportunities for
improvements and making changes in the production planning to improve the operations

performance.

122

5.3 RESULTS AND CONTRIBUTIONS

The main goal of this research was to evaluate plant production planning process in the
manufacturing housing industry and to identify opportunities for improvement. The

objectives were:

1. Documenting the currently utilized production planning process in a
Manufactured Housing plant.

2. Assessing and quantifying the production planning process performance using
lean construction principles.

3. Identifying causes of off-target performance through constraint analysis
techniques and suggesting opportunities for production planning process

improvement.

These objectives were achieved by using the discussed methods and tools of
Chapter II and Chapter III, which were, literature review, interviews, last planner system,
productivity ratings, and constraint analysis. The results of this research project were

discussed in Chapter IV and Chapter V. The results of this research were:

1. Identifying the production planning process requirements.

2. Calculating Percent Plan Complete and Labor Utilization Factor for 12 stations of
a manufactured housing plant.

3. Determining reasons for not completing assignments using constraint analysis.

4. Identifying opportunities for production planning process improvement.

123

This research makes the following contributions to the research on manufactured

housing:

1.

A documented production planning workflow, production planning strategies, and
production planning work requirements in manufactured housing production
process.

A method to evaluate performance of production planning process in a
manufactured housing plant and identify process improvement opportunities. The
method demonstrated in this research was using Last planner’s PPC tool and
Productivity ratings through work sampling.

Investigated relation between Percent Plan Complete and Labor Utilization
Factor. For the present data set available it was found that a linear relationship
existed between PPC and LUF. The author at the same time feels that more data
points are required to access this relationship in further detail. For the available
data the relationship attained was a contribution of the author.

Demonstrated the use of Pareto analysis and Fish bone diagrams as constraint

analysis techniques to identify process improvement opportunities.

5.4 RECOMMENDATIONS AND GUIDELINES FOR THE INDUSTRY

After studying and understanding the whole production planning aspect of manufactured

housing production plants, the author has come up with the following recommendations

and guidelines.

124

 

1. Manufactured housing industry is a combination of site-built construction and
manufacturing industry (E.g.: Automobile industry). The industry can take
advantage of both the industry’s practice to improve its production planning.

2. The tools used like LUF, PPC, Pareto analysis and F ishbone technique can be
used to fit the manufactured housing industry needs. This research used these
tools to assess the production planning. PPC, out of Last planner can be used
as a production planning technique by the MH plant. The foremen to assess
the performance of each individual station can also use PPC at individual
stations. Similarly, the crew performance can also be measured using LUF
results and finally Pareto analysis can be performed to improve the
performance at each the station level, improving the performance of the whole
plant.

3. Each individual plant of a company could also use these tools and the CEO or
owner of the company can get a clear picture of performance of each plant.
For instance, the CEO of a company would get a clear picture of the
production capacity and the current productivity. If provided with a chart for
one month of observations of PPC and LUF, with constraint analysis, he could
determine the problem areas and take appropriate measures to streamline the

production process and reduce the waste in the production.

The next section describes future areas of research and other aspects of this research

project, which were not looked into due to the limits and scope of this work.

125

5.5 LIMITATIONS & FUTURE AREAS OF RESEARCH

This research project assessed the production planning of a manufactured housing plant
using lean production principles. Like any other project, this research had certain

limitations. The following were the limitations of this research project:

1. This project had a very small sample and it does not reflect the entire
manufactured housing industry’s production planning. Only two plants were
considered to achieve the set objectives.

2. Due to the time and scope of the work, this research project observed only a few
stations in the plant. Also, the selection of stations for this project was based on

facilitated access to the production plant.

As a part of this research the author studied two factories in the regions of
northern Indiana. Only few of the stations along the assembly line and other subassembly
stations were studied. The production planning process was analyzed and assessed.
Different problems and opportunities for production planning process were identified.
One of the major areas of future research could be the application of this production

planning process assessment technique to the complete factory.

Also, this method of production planning assessment can be applied to different
production plants and a set of problems can be identified based on the location of the
factory. The sample for this research project was very small and it does not give an
overall picture of the entire manufactured housing industry. Applying this technique to a

larger sample would give an overall picture of the manufactured housing industry.

126

 

Research can be carried out in order to map the exact material consumption
involved at every activity and sub activity station. A research on this line would help find
more accurate solutions to the problems presented in this research. Just-in-time concept

can be applied to see the material handling in a manufactured housing industry.

5.6 SUMMARY

This thesis report has been divided into five chapters. In Chapter 1, an overview of the
evolution of manufactured houses was introduced. Various aspects of the manufactured
housing industry, like cost analysis, demand issues were discussed and a problem

statement was formulated. The research goal and objectives were also presented.

In Chapter II, a detailed description of terms and existing research related to the
manufactured housing industry were discussed. Chapter II also summarized the different
production processes, lean production, lean construction, and the existing research in

these areas. Finally a background of production control and last planner were described.

In Chapter III, methods and tools to achieve each of the three objectives set in
Chapter 2 were explained. This chapter also explained in detail each step of the PPC,
productivity ratings, and constraint analysis tools used in this research. Chapter IV

reported on data collection and results.

Finally, Chapter V discussed Pareto analysis and input from the industry was used
to construct a fishbone diagram and find the main constraints. The chapter finished with
the research with conclusion, results and contributions, limitations, and firture areas of

research.

127

APPENDICES

128

 

ays Observed: 5

Ectivity: Exterior boards

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

rew size: 1
Crew Average
Da Planned Corn leted
ys p Crew PPC Reasons/Delays
Day 1 CrewI 3 3 100%
Day 2 Crew I 3 3 100%
Day 3 Crew I 3 3 100%
Day 4 Crew I 4 3 75% Work from upstream not
released

Day 5 Crew I 3 2 66.6% Work not released
ctivity: Doors and windows installation

ays Observed: 5

rew size: 1

Crew Average
Da s Planned Corn leted
y p Crew PPC Reasons/Delays
Day 1 CrewI 3 3 100%
Day 2 Crew I 3 3 100%
Day 3 Crew I 3 3 100%
Day 4 Crew I 4 2 50% Work from upstream not
released

Day 5 Crew I 3 2 66.6% Work not released (Directives)

 

 

 

 

 

 

 

129

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Activity: External siding
1Days Observed: 5
Crew size: 1
Crew Average
Da s Planned Com leted
y p Crew PPC Reasons/Delays
Assisting a different crew at
0
Day 1 Crew I 3 2 66.6/o .another station
Day 2 Crew I 3 3 100%
Day 3 Crew I 3 3 100%
Day 4 Crew I 4 3 75% Work from upstream not
released
Day 5 Crew I 3 2 66.6% Work not released (Directives)
ctivity: Paint ceiling
ays Observed: 5
rew size: 1
Crew Average
Da s Planned Com leted
y p Crew PPC Reasons/Delays
Work not released from roof
0
Day 1 Crew I 3 1 33.3 /o truss building
Worker absent / Roof truss
0
Day 2 Crew I 3 2 66'6 A, worker completes 2 out 013
Day 3 Crew I 3 3 100%
Da 4 Crew I 4 3 750/ Work from upstream not
y 0 released
Day 5 Crew I 3 2 66.6% 1Material shortage

 

 

 

 

 

 

 

130

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Activity: Roof board
[Days Observed: 5
Crew size: 2
Da s Crew Planned Completed Average
y Crew PPC Reasons/Delays
meI 3 2
Day 1 66.60% Work not released
Crew 11 3 2
meI 3 2 -
Day 2 6 6. 60% Crew not present/Shingles
Crew 11 3 2 workers do the work
meI 3 3
Day 3 100%
Crew II 3 3
mel 4 3 .
Day 4 75% Material shortage
Crew 11 4 3
mel 3 3
Day 5 100%
Crew 11 3 3
Activity: Roof shingles
1Days Observed: 5
Crew size: 2
Crew Average
Da s Planned Com leted
y p Crew PPC Reasons/Delays
Crew I 3 2
Day 1 66.60% Work not released
Crew 11 3 2
meI 3 3
Day 2 100%
Crew II 3 3
mel 3 3
Day 3 100%
Crew 11 3 3
Crew I 4 3
Day 4 75% Work not complete at roof board
Crew 11 4 3 station
meI 3 3
Day 5 100% Rework (Mistake by crew)
Crew 11 3 3

 

131

 

 

 

 

Activity: External finish
1Days Observed: 5

 

 

 

 

 

 

 

 

 

 

 

 

Crew size: 1
Days rew Planned Completed €133.33: Reasons/Delays
Day 1 CrewI 3 2 66.6% 1Material shortage
Day 2 Crew 1 3 1.5 50% 22:11:32? released and slow crew
Day 3 Crew I 3 3 100%
Day 4 Crew 1 3 2 6 6. 6% flSalt<;w work and work released
Day 5 Crew 1 3 1 33.3% 3:312:32? released and slow crew

 

 

132

 

 

 

Crew LUF of Plumbing and HVAC

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Crew Total Observations Efficient Contributory LUF
Crew 1 409 227 103 0.61
Crew 2 411 221 144 0.62

AM LUF 0.615
Crew 1 400 175 120 0.51
Crew 2 404 141 99 0.41

PM LUF 0.46

Average crew LUF 0.5375
Crew LUF of internal walls

Crew Total Observations Efficient Contributory LUF
Crew 1 407 215 117 0.6
Crew 2 409 204 68 0.54

AM LUF 0.57
Crew 1 425 165 102 0.44
Crew 2 425 158 134 0.45

PM LUF 0.445

Average crew LUF 0.5075

 

 

133

 

 

 

 

Crew LUF of external walls station

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Crew Total Observations Efficient Contributory LUF
Crew 1 406 305 49 0.78
Crew 2 406 201 43 0.52

AM LUF 0.65
Crew 1 424 77 105 0.24
Crew 2 424 175 141 0.49

PM LUF 0.46

Average crew LUF 0.365
Crew LUF of Exterior boards

Crew Total Observations Efficient Contributory LUF

Crew 1 384 230 102 0.66

AM LUF 0.66
Crew 1 394 131 111 0.4
PM LUF 0.4
Average crew LUF 0.53

 

 

134

 

 

 

Crew LUF of External Siding

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Crew Total Observations Efficient Contributory LUF
Crew 1 410 222 115 0.61
AM LUF 0.61
Crew 1 432 201 142 0.54
PM LUF 0.54
Average crew LUF 0.575
Crew LUF of External Finish
Crew Total Observations Efficient Contributory LUF
Crew 1 397 167 117 0.49
AM LUF 0.49
Crew 1 432 181 136 0.49
PM LUF 0.49
Average crew LUF 0.49

 

 

135

 

 

 

 

Crew LUF of Roof truss building

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Crew Total Observations Efficient Contributory LUF
Crew 1 400 152 88 0.43
Crew 2 404 140 95 0.4

AM LUF 0.415
Crew 1 432 115 98 0.32
Crew 2 432 148 173 0.44

PM LUF 0.38

Average crew LUF 0.39
Crew LUF of Shingles Station Crew

Crew Total Observations Efficient Contributory LUF
Crew 1 418 270 58 0.68
Crew 2 418 262 51 0.65

AM LUF 0.665
Crew 1 398 212 35 0.55
Crew 2 398 187 47 0.49

PM LUF 0.52

Average crew LUF 0.59

 

 

136

 

 

 

 

Crew LUF of Paint ceiling

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Crew Total Observations Efficient Contributory LUF
Crew 1 432 233 61 0.57
AM LUF 0.57
Crew 1 401 168 94 0.47
PM LUF 0.47
Average crew LUF 0.52
Crew LUF of Roof setting in place

Crew Total Observations Efficient Contributory LUF
Crew 1 407 210 101 0.57
AM LUF 0.57
Crew 1 408 206 113 0.57
PM LUF 0.57
Average crew LUF 0.57

 

 

137

 

 

10.

11.

12.

13.

14.

REFERENCES

. Albem, W. F. and Morris, M. D. (1997). Factory-Constructed Housing

Development: Planning, Design, and Construction. CRC Press.

Balakrishnan, R. (2001). The Toyota Production System: A Case Study of
Creativity and Innovation in Automotive Engineering. University of Bristol.

Ballard, G. (2000). The Last Planner System of Production Control. PhD
dissertation, University of Birmingham, UK.

Ballard, G. and Howell, G. (1998). What kind of production is Construction?
Proc. Annual Conference of International Group of Lean Construction, 1998.

Ballard, G. (1997). Improving Work Flow Reliability. Proc. 7th Annual
Conference of International Group of Lean Construction, Berkeley, CA, July 26-
28, 1999.

Bhattacharyya, G. K. and Johnson, R. A. (1977). Statistical Concepts and
Methods. John Wiley & Sons; lst Edition.

Chase, R., Aquilano, N., and Jacobs, F ., Production and Operations Management:
Manufacturing and Services. Eighth Edition, Irwin McGraw-Hill.

Diekrnann, J.E., Thrush, KB, and Wilson T. (1987). Project Control in Design
Engineering, Cost Engineering, Vol. 29, No. 3.

Horman, M. and Kenley, R. (1997). The application of lean production to project
management. The University of Melbourne.

Howell, G. and Ballard, G (1994). Lean production theory: Moving beyond ‘Can-
Do’. Proc. Annual Conference of International Group of Lean Construction, 1994.

Hullibarger, S. (2001). Developing With Manufactured Homes. Manufactured
Housing Institute, VA.

Hunt, V. (1993). Reengineering. Oliver Wright Publications, Inc., VT.
Kaufman Consulting Group, LLC, (2000).
Koskela, L., Howell, G., Ballard, G., and Tommelein, I. D. (2001). The

Foundations of Lean Construction. Department of Civil and Environmental
Engineering, Stanford University.

138

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

Koskela, L. (1993). Lean Production in Construction. lst Workshop, International
Group for Lean Construction, IGLC-l, 11-13 August 1993, Technical Research
Center of Finland (VTT), Espoo, Finland.

Koskela, L. (1999). Management of Production in Construction: A Theoretical
View. Proc. Annual Conference of International Group of Lean Construction,
1999.

Koskela, L. (2000). An exploration towards a production theory and its
application to construction. Helsinki University of Technology, Finland.

Lean Construction Institute. Web address: http://www.1eanconstruction.org[

Markland, R. B, Vickery, S. K., Davis, R. A., and Sounderpandian, J. (1998).
Operations Management: Concepts in Manufacturing and Services: Om Sheets:
Excel Spreadsheet Problems. Southwestern Publishing Co., January 1998.

Manufactured Housing Institute. Web address:
http://www.manufacturedhousing.org

Manufactured Housing: Origins and Evolution, Research paper, Harvard
University, 1997.

Mendenhall, W., Scheaffer, R. L., and Wackery, D. D. (1990). Mathematical
Statistics with Applications. Fourth Edition. PWS-KENT Publishing Company,
Boston, MA.

Nutt-Powell, T. E. (1982). Manufactured Houses: Making sense of a Housing
opportunity. Auburn House, January 1982.

Oglesby, C.H., Parker, H., and Howell, G. (1989). Productivity Improvement
Studies in Construction. McGraw Hill Book Company, New York, NY.

Perkins, B. (1999). Manufactured Housing Remains Growing Force In Housing
Market. Realty Times Columnist, March 26, 1999. www.realigtimes.com.

Porter, ME. (1985). Competitive advantage: Creating and sustaining superior
performance. Free Press, New York.

Senghore, O (2001) “Production and Material Flow Process Model For
Manufactured Housing” Michigan State University.

Singh, A. (2001). Lean engineering for Mass housing - Design, Manufacture and
Site erection. Journal of Construction Research, Vol. 3. No. 1 (2002).

Vermeer K., Louie J. (1997). The Future of Manufactured Housing. Joint Center
for Housing Studies Of Harvard University.

139

 

30. Warner, K. and Johnson, R. (1993). Manufactured housing impacts on adjacent
property values. Report 4, Manufactured Housing Research Project, College of
Architecture and Urban Planning, University of Michigan, Ann Arbor, Michigan.

31. Womack, J. P., Jones, D.T., and R008, D. (1991). The Machine That Changed the

World: The Story of Lean Production. lSt Edition, HarperCollins Trade
Publishers, London, UK.

140

 

   

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

11111111111111

1
312

    

11