fig? . . may.

rt?! ‘ ‘ V \: :l

wmwkgfi

This is to certify that the

dissertation entitled

THE IMPACT OF SUPPLY CHAIN INTEGRATION MECHANISMS
ON SUCCESS IN NEW PRODUCT DEVELOPMENT PROJECTS"

presented by

Mulmumudi Jayanth Jayaram

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Business - Administration

 

Major professor --"*‘

Date 8/”1/ ?8

0-12771

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

 

 

 

 

 

 

 

DATE DUE DATE DUE DATE DUE
fitfiu‘xfxf s g
9 i“) m
_ .14 '
in 2 2 n5
[2 1622006

 

 

 

 

 

 

 

 

 

1M chlRC/Dahfinpesp."

 

 

THE IMPACT OF SUPPLY CHAIN INTEGRATION MECHANISMS
ON SUCCESS IN NEW PRODUCT DEVELOPMENT PROJECTS

By

Mulumudi J ayanth J ayaram

A DISSERTATION
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Marketing and Supply Chain Management

1998

Copyright by
MULUMUDI JAYANTH JAYARAM
1998

ABSTRACT

THE IMPACT OF SUPPLY CHAIN INTEGRATION MECHANISMS
ON SUCCESS IN NEW PRODUCT DEVELOPMENT PROJECTS

By

Mulumudi Jayanth Jayaram

Recent research has emphasized cross fimctional integration as a key facilitator of
success in new product development (NPD) projects. This dissertation developed and
tested a model using supply chain integration as a key enabler influencing both planning
of the new product and execution of the development process in NPD projects. Supply
chain integration was operationalized to include cross fimctional (for example, between
design and manufacturing) and boundary spanning integration mechanisms (such as
supplier integration). Product concept effectiveness (a construct captming the planning
of the new product), and development process performance (a construct measuring the
execution of new product development process) were theorized to influence overall
project success. Hypotheses drawn from this framework were tested on a multi-industry
sample (electronics, communications, semiconductors, and computers) of firms. Data
analysis was conducted using regression analyses, partial correlations, confirmatory
factor analyses and structural equation modeling.

The results from the data analysis indicated support for the importance of

integration mechanisms as enablers of success in new product development projects.

Specifically, infrastructure programs with key suppliers, design proactiveness and
‘capturing the voice of the customer’ were found to be the most critical integration
mechanisms that influenced NPD project success. Integration mechanisms had a
differential impact on individual measures of project success. The results support the
complementary influence of product planning and development process execution on
project success. The contingent influence of market and product related factors on the
integration mechanisms —) development process performance linkage was also supported.
Finally, the results suggest that successful attainment of the goal of reducing time-to-
market has associative benefits of attaining goals relating to lower product costs and
superior conformance and design quality.

Contributions of this dissertation to practice included identification of specific
‘supply chain’ integration mechanisms that were significantly related to project success
and to the mediating factors (product planning and project execution) of project success.
Also, integration mechanisms were shown to have a contingent role in NPD projects.
This dissertation contributed to theory by bridging the gap between two established
schools (rational planning and problem solving) in new product research by developing
and testing a model using constructs from both schools. Contributions to methodology
include development of new scales (for example, fit with firm capability) and verification

of measurement properties of existing scales at the NPD project level.

This dissertation is dedicated to my father, Mulumudi Jayaram, who would have been
extremely proud of this day; to my mother, Sundari J ayararn, whose love and support
continues to sustain me; and to my wife, Vasanthi, who has helped me in innumerable

ways throughout my graduate study and dissertation work.

ACKNOWLEDGEMENTS

Several people have contributed to the successful completion of this dissertation. I
would like to thank Dr. Ram Narasimhan for his support, guidance and direction
throughout the doctoral program. He helped me discover the joys of pursuing rigorous
and meaningful research. I am fortunate to be one of his students. In particular, I am
indebted to him for his unwavering support and encouragement in my search for an
academic position.

A special word of thanks to Dr. Roger Calantone for his valuable insight and
innumerable pieces of advice on professional and personal issues. I thank Dr. Robert
Handfield for his positive support and help in sharpening my empirical skills, and Dr.
Gary Ragatz for his cheerful disposition towards me, helpful suggestions dtu'ing my job
search process and support throughout the dissertation process.

I am grateful to the National Association of Purchasing Managers (NAPM) for its
support through the Doctoral Fellowship Award. This research project could not have
been completed without their generous financial support. A special ‘thank you’ to the
managers and executives who participated in this research project. Several managers
went out of their way to help me in this project, by suggesting new contacts and
providing timely feedback. I hope that the findings of this researCh, in some small

measure, provides some value for the time invested in this project.

vi

I sincerely appreciate the efforts of Dr. Shawnee Vickery and Dr. Cornelia Droge
who involved me in their research projects and also provided helpful suggestions in my
job search. I also thank Dr. Robert Nason and Dr. Michael Moch for their overall support
and encouragement. I would like to acknowledge the warm support, guidance and
helpfulness of Dr. Larry Stimpert, who demonstrated, by example, the joys of teaching
and also provided unique insights into the field of business strategy. To the staff of both
departments - Sue, Kathy, Lonnie, Laurie, Cheryl, Cindy, Renee and Marilyn — I would
like to express my sincere appreciation for all their hard work and enthusiastic support.

My greatest gratitude is to my family — my brothers (U dai and Umakanth), sisters-
in-law (Shalini, Padma, Jayanthi and Jyothi), wife (V asanthi), parents (Jayaram and
Sundari), in-laws (RVG Rao and Adhilakshmi) and Annamma. I would like to thank Dr.
S.A.N Moorthy (mathematician par excellence) for inspiring me to pursue an academic
career. I cherish the joyous moments with all the little bundles of joy in my life — my
very own Tara (you are truly a ‘star’ to mcl), Meera (whom I have watched grow into a

darling!), Kushal (the energetic and exuberant lad!) and Kunal (the silent toddler!).

vii

TABLE OF CONTENTS

LIST OF TABLES ............................................. x
LIST OF FIGURES ............................................. xiii
CHAPTER 1
OVERVIEW ................................................... 1
1.1 Introduction and Motivation for the Study ....................... 1
1.2 Research Objectives ......................................... 4
1.3 Research Methods .......................................... 5
1.4 Summary of Major Findings ................................... 6
1.5 Contributions of the Study ............................... ' ..... 7
1.6 Organization of the Dissertation ................................ 8
CHAPTER 2
REVIEW OF THE LITERATURE ................................... 9
2.1 Introduction ................................................ 9
2.2 Theoretical Perspectives of Product Development ................... 10
2.2.1 Product Development as a Rational Plan ...................... 10
2.2.2 Product Development as Disciplined Problem Solving ........... 10
2.2.3 Product Development as a Communication Web ................ 12
2.2.4 Development Process Performance — Schools of Thought ........ 12
2.3 Determinants of Product Development Success ..................... 15
2.3.1 Meta Analytic Findings ................................... 15
2.3.2 The Integration Literature .................................. 15
2.3.3 The Concurrency Literature ............................... 17
2.3.4 The Contingency Literature ............................... 18
2.4 Supply Chain Integration Framework of Product Development ......... 19
2.4.1 Introduction to the Framework .............................. 19
2.4.2 Summary of the Research Objectives ......................... 20
CHAPTER 3
SUPPLY CHAIN INTEGRATION FRAMEWORK .................... 21
3.1 Definition of Supply Chain Integration .......................... 21
3.2 Supply Chain Integration Framework .......................... 21
3.2.1 Product Development Project Success ...................... 22
3.2.2 Product Concept Effectiveness ............................ 23
3.2.3 Development Process Performance ......................... 24
3.2.4 Supply Chain Integration Mechanisms ...................... 26
3.2.5 Concurrent Process Management .......................... 32
3.3 Research Hypotheses ........................................ 34
3.3.1 Antecedents to Project Success ............................ 35
3.3.2 Antecedents to Product Concept Effectiveness ................ 37
3.3.3 Antecedents to Development Process Performance ............. 37
3.3.4 Hypotheses relating to Contingency Relationships ............. 38

viii

CHAPTER 4

RESEARCH DESIGN AND METHODOLOGY ........................ 42
4.1 Research Design ............................................. 42
4.1.1 Sampling Frame ........................................ 43
4.1.2 Unit of Analysis ........................................ 44
4.2 Research Methodology ........................................ 45
4.2.1 Survey Development ..................................... 45
4.2.2 Data Collection Procedures ................................ 48
CHAPTER 5
ANALYSIS AND FINDINGS ..................................... 52
5.1 Introduction ............................................... 52
5.2 Description of the Sample .................................... 53
5.2.1. Company Background .................................. 53
5.2.2. Respondent Profile .................................... 55
5.2.3. Project Characteristics ................................. 56
5.2.4. Descriptive Statistics of Contingency Variables .............. 60
5.3 Data Quality .............................................. 63
5.3.1 Reliability of Scales .................................... 63
5.3.2 The Supplier Integration Construct ....................... 63
5.3.3 The Design-Manufacturing Integration Construct ............. 65
5.3.4 The Customer Integration Construct . . . . . . . ................ 66
5.3.5 The Concurrent Process Management Construct .............. 67
5.3.6 Product Fit with Firm Capability Construct .................. 68
5.3.7 Product Development Process Performance Construct ......... 69
5.3.8 Product Development Project Success Construct ............. 69
5.3.9 Validity of Constructs .................................. 71
5.4 Findings .................................................. 77
5.4.1 Antecedents to Project Success ........................... 77
5.4.2 Antecedents to Product Concept Effectiveness .............. 89
5.4.3 Antecedents to Development Process Performance ........... 95
5.4.4 Contingency Relationships .............................. 102
5.4.5 Summary of Major Findings ............................. 109
CHAPTER 6
DISCUSSION AND CONCLUSIONS ............................... 1 11
6.1 Introduction ............................................... 111
6.2 Determinants of Project Success ............................... 111
6.3. Supply Chain Integration Mechanisms .......................... 112
6.4 Contingency Relationships ................................... 113
6.5 Synthesis ................................................. 114
6.5.1. Pattern of Influences ................................... 120
6.6. Managerial Implications ..................................... 122
6.7 Academic Contributions ..................................... 125
6.8 Limitations of the Research ................................... 128
6.9 Directions for Future Research ................................ 129
APPENDICES .................................................. 132
A. Survey Inshument ........................................ 132
B. Cover Letter ............................................. 139
C. Reminder Postcard ........................................ 140
D. Follow-up Letter ......................................... 141 -
REFERENCES ................................................ 142

ix

LIST OF TABLES

Table Page
2.1. Comparison of Research Streams in Product Development ........... 11
4.1. Assessment of Non-Response Bias ............................. 51
5.1. Company Age ............................................. 5 3
5.2. Number of Employees ...................................... 54
5.3. Industry Background of the Projects ............................ 54
5.4. Respondent's Functional Background ........................... 55
5.5. Cumulative Person Hours Worked ............................ 57
5.6. Location of Product in its Product Family Stream ................ 57
5.7. New Product Development Project Duration .................... 58
5.8. Duration in Market since Product Launch ....................... 59
5.9. Unit Sales Price of Product .................................. 59
5.10. Technical Complexity of New Product ........................ 60
5.11. Level of Product Innovativeness for Intended Market ............. 61
5.12. Market Uncertainty ........................................ 61
5.13. Stage in the Product Life Cycle for Targeted Product .............. 62
5.14. Factor Analysis : Supplier Integration .......................... 64
5.15. Factor Analysis : Design-Manufacturing Integration ............... 65
5.16. Factor Analysis : Customer Integration .......................... 67
5.17. Factor Analysis : Concurrent Process Management ................ 68
5.18. Factor Analysis : Product F it with Firm Capability ................ 69
5.19. Factor Analysis : Product Development Process Performance ........ 7 0
5.20. Factor Analysis : Product Development Project Success ............ 70
5.21 . Confirmatory Factor Analysis ................................. 74
5.22. Predictive Validity of Constructs ............................... 76
5.23. Effect of Product Concept Effectiveness on NPD Project Success ..... 77
5.24. Effect of Conformance Quality on NPD Project Success ............. 78
5.25. Effect of Design Quality on NPD Project Success .................. 79
5.26. Effect of Time-to-Market on NPD Project Success .................. 79
5.27. Effect of Cost Performance on NPD Project Success ................ 79
5.28. Effect of Components of Development Process Performance on

NPD Project Success ......................................... 80
5.29. Effect of the Index of Development Process Performance on

NPD Project Success .................................... 81
5.30. Effect of Components of Development Process Performance on

Components of NPD Project Success ............................ 82
5.31. Descriptive Statistics : Components of Development

Process Performance ..................................... 83
5.32. Partial Correlations of Components of Development

Process Performance ..................................... 84

Table

5.33.

5.34.

5.35.

5.36.

5.37.

5.38.

5.39.

5.40.

5.41.

5.42.

5.43.
5.44.

5.45.
5.46.

5.47.

LIST OF TABLES (CONT’D)

Partial Correlations between Components of Development

Process Performance and Market Share .......................

Partial Correlations between Components of Development

Process Performance and Profitability .......................

Partial Correlations between Components of Development

Process Performance and Break Even Time .......................

Effects of Product Concept Effectiveness and Development Process

Performance on NPD Project Success ............................

Hierarchical Regression Analyses: Effects of Product Concept
Effectiveness, Development Process Performance and their

Interactions on Components of NPD Project Success ................
Effect of Supplier Integration on Product Concept Effectiveness .......

Effect of Design- Manufacturing Integration on Product Concept

Effectiveness ...............................................
Effect of Customer Integration on Product Concept Effectiveness ......

Factor Score Regression Analyses: Effects of Supplier,
Design-Manufacturing and Customer Integration on

Product Concept Effectiveness ..................................

Effect of Product Concept Effectiveness on Development Process

Performance ................................................
Effect of Supplier Integration on Development Process Performance . . . .

Effect of Design-Manufacturing Integration on Development Process

Performance ................................................
Effect of Customer Integration on Development Process Performance . . .

Effect of Concurrent Process Management on Development Process

Performance .................................................

Factor Score Regression Analyses: Effects of Integration Mechanisms
and Concurrent Process Management on Development Process

Performance .................................................

5.48a. Moderating Effect of Product Innovation on Concurrent Process

Management (Factor 1) - Time-to-Market Relationship ..............

5.48b. Moderating Effect of Product Innovation on Concurrent Process

Management (Factor 2) — Time-to-Market Relationship ..............

5.48c. Moderating Effect of Product Innovation on Cross Functional

Communication — Time—to-Market Relationship .....................

5.49a. Moderating Effect of Technical Complexity on Concurrent

Process Management (Factor 1) — Time-to-Market Relationship ........

xi

Page

85
85
86
87
88
9O
92
93
94

95
97

98
99

100

102
103
104
104

105

Table

LIST OF TABLES (CONT’D)

5.49b. Moderating Effect of Technical Complexity on Concurrent Process

Management (Factor 2) - Time-to-Market Relationship ...............

5.50a. Moderating Effect of Market Uncertainty on Concurrent Process

Management (Factor 1) - Time-to—Market Relationship ...............

5.50b. Moderating Effect of Market Uncertainty on Concurrent Process

Management (Factor 2) — Time-to-Market Relationship ...............

5.51a. Moderating Effect of Product Life Cycle on Supplier

Integration (Factor 1) - Time-to-Market Relationship .................

5.51b. Moderating Effect of Product Life Cycle on Supplier

Integration (Factor 2) — Time-to-Market Relationship .................

5.51c. Moderating Effect of Product Life Cycle on Supplier

6.1.

6.2.

6.3.

6.4.

6.5.

6.6.

Integration (Factor 3) - Time-to-Market Relationship .................
Strength of Effects for the Product Fit to Capability

Components to Project Success Relationships .......................
Strength of Effects for Development Process Performance

to Components of Project Success Relationships .....................
Strength of Effects for Integration Mechanisms to Components

of Product Concept Effectiveness, Development Process

Performance and Project Success Relationships ......................
Strength of Effects for Concurrent Process Management

to Development Process Performance and

Project Success Relationships ....................................
Strength of Effects for Integration Mechanisms to Components

of Project Success Relationships ..................................
Strength of Effects for Concurrent Process Management to

Components of Project Success Relationships ........................

xii

Page

105

' 106

107

108

108

109

115

116

118

119

120

120

LIST OF FIGURES

FIGURE 1 : Supply Chain Integration Framework ..................... 22
FIGURE 2 : Pattern of Influences .................................. 122

xiii

CHAPTER 1

OVERVIEW

1. Introduction and Motivation for the Study

The last decade or so has seen a number of sweeping changes that have affected
firms in several industries. Hyper-competition, reduction in product life cycles, and
increasing need for mass customization are a few changes worth noting. To respond to
these changes, firms have used innovation as a source of differentiation, and have sought
to increase the pace with which innovation can be introduced to the market. The
underlying premise being that agile competitors and in particular, agile product
developers have a higher chance of success in the market place. Thus, most firms
consider the fast development of new products as a key strategic activity. Indeed, the
fortunes of firms are made and dissipated on the basis of product and process innovation.
However, empirical evidence suggests that quite a few new products fail in the market
place. Booz, Allen and Hamilton (1982) found that the failure rate of new products
introduced between 1963 and 1981 was as high as 35%. In a later stu'dy, Cooper (1990)
confirmed that new products continue to fail at a high rate as they did twenty five years
ago. To understand the persistence of this pattern, a number of studies have been
conducted to isolate the key factors that enable product development success. Three

separate meta analyses have been conducted to synthesize the results of this stream of

literature (see, Montoya-Weiss and Calantone, 1994; Brown and Eisenhardt, 1995; and
Kessler and Chakarabarti, 1996). Yet, the findings from these three reviews are
paradoxical. The broad correlates of new product development project success are not
consistent, both in terms of a presence of a relationship and in terms of the strength of
their impact, across a variety of settings. All the three reviews call for an integrated
examination of broad and new factors that affect new product development success.

Until recently, the dominant view among product development managers was to
pay attention to the product development process, typically visualized as a series of steps
that starts fi'om the inception of a new idea and ends with the introduction of a
commercially viable product. Moreover, these series of steps were treated as fixed and
the ‘quality’ of execution of the steps was inferred in terms of financial success of the
project. This view is limited because of the absence of a strategic context for evaluating
product development project success. More specifically, little was known about why
certain projects enjoy higher levels of success than others and whether the successful
projects also had products that were planned at the inception stage. This research
proposes to use the planning/execution classification, which has strong historical support
in the business strategy literature, as a means of more closely isolating the factors that
contribute to new product development success.

Using this classification, this research suggests that the success of product
development projects depend on two broad issues: product definition-related issues and
development process-related issues. Under “product definition,” typical research issues
are: “What requirements of the customer base does this new product satisfy uniquely?” or

“Does the firm have the capability to design, manufacture and introduce this product to
2

the market effectively?” On the other hand, “development process” related issues deal
with the quality of the execution of the development process, i.e., “What is the impact of
the various development activities (from concept to market introduction) on project cost,
product quality, and project lead times?” This dissertation will investigate these issues.

The interrelationships among components of development process performance
(quality, cost and time to market) is another area that has grown considerably recently.
Managers generally are interested in optimizing cost, quality and lead time objectives for
new product development projects. However, limited resources (cost, time, personnel
etc.) are available to managers to simultaneously minimize cost and lead times, and to
maximize product quality. In the literature, one camp (called the trade-off school) has
argued that high performance along one dimension of development process performance
(for example, superior quality) precludes high performance in other components of
development process performance (for example, low cost or low time to market).
Representative works from the trade-off school can be found in Cohen, Eliashbcrg, and
Ho (1996), Calantone and Di Beneditto (1996), and Gupta, Brockoff, and Weisenfeld
(1992). On the other hand, a parallel stream of research (called the synergy school) has
also found that there can be synergies in achieving two of the three development process
outcomes. Representative works of the synergy school can be found in Ittner and Larcker
(1997), (Flynn,l994), Raia (1991), and Valentino and Christ (1989). This dissertation
addresses this inconsistency in the literature.

The previous two research objectives pertained to the performance consequences of
effective product planning and effective execution of the development process. The final

objective of this research is to identify factors that contribute to effective product
3

planning and execution of the deveIOpment process. Prior literature has called for the
inclusion of integration mechanisms as predictors of success in technological innovation
(Shrivastava and Souder, 1987), process innovation (Ettlie and Reza, 1992), and product
development (Clark, 1989; Shrivastava and Souder, 1987). This research builds on this
body of literature by identifying three kinds of integration mechanisms -- supplier
integration, design-manufacturing integration, and customer integration -- as enablers of
project success, via the mediating influences of effective product planning and effective
execution of the development process. In other words, the three integration mechanisms
that collectively span the supply chain are hypothesized to have an impact on product
concept effectiveness and development process performance (i.e., product quality, project
cost and project lead time performance). Besides the three integratiOn mechanisms, the
construct of concurrent process management practices is also hypothesized to influence
development process performance. Support for the inclusion of this relationship is

provided in the next chapter.

1.2 Research Objectives

This dissertation has three research objectives. One objective of this research is to
understand the relative importance of “product definition” and “development process”
issues on project success. That is, for a given development project, is. product definition
more important than development process, or are both equally important? The second
objective of this dissertation is to understand how managers of successful product
development projects assign priorities to the three objectives of time, cost and quality.

The final objective of this research is to identify practices that influence product concept
4

effectiveness and development process performance. Four sets of practices are
specifically tested, three of which collectively form supply chain integration mechanisms
and the remaining set of practice constitutes the factor of concurrent process

management.

1.3 Research Methods

The unit of analysis for this research was a new product development project.
Firms from a variety of industries were selected in order to test for contingency
relationships across different settings. Data collection was conducted using a survey
instrument. The ideal respondent for the survey was specified to be “the person with
overall responsibility for overseeing product development projects”. The instrument was
analyzed for content validity by product development managers from three different
industries, before the large scale mail-out. Each manager completed the survey and
provided feedback on wording of the items, understandability, organization of the survey
and length of the survey. The instrument was refined based on this feedback.

The sampling frame was firms in “high tech” industries which was defined to
include frrms belonging to the Standard Industrial Classification (SIC) codes 35 to 38
comprising industries such as Industrial Equipment, Computers, Electronics and
Electrical Equipment, Motor Vehicles, Scientific Instruments and Medical Devices.
Firms belonging to these SIC codes were chosen because they represent industries
involved in the manufacture of relatively high value-added and in most cases high
technology products. An initial contact list of approximately 5000 firms was generated

from two groups of sources — membership lists in professional associations and
5

commercial mailing lists. A detailed description of the procedure followed to generate
this list is done in Chapter 4. From the contact list of 5,000 firms, 1,500 firms were
selected at random (without replacement) fOr the initial mailing. A follow up post card
mailing after two weeks of the initial mailing and a second mailing after four weeks of
the initial mailing yielded 338 usable surveys (24% response rate).

As has been noted in a recent meta analysis of empirical studies in product
development projects, researchers have focused on two kinds of links -- between practice
and performance, and between construct and performance -- interchangeably (see
Montoya-Weiss and Calontone, 1994). Thus, this research also investigated both types of
linkages. The data was analyzed using regression analyses and exploratory factor

analysis techniques.

1.4 Summary of Major Findings

The first research objective was to investigate the relative role of product concept
effectiveness and development process performance on new product development project
success. Both product concept effectiveness (as measured by fit with firm capabilities)
and development process performance were significantly related to project success. The
strength of influence was stronger in the case of development process performance
indicating that execution of the new product development process was more critical (as
compared to product concept effectiveness) in terms of its impact on project success.

The second research objective was to understand how managers of successful in
product development projects assigned priorities to the three objectives of time, cost and

quality. All the three indicators of product development process performance (Time,
6

Cost and Quality) significantly influenced project success. Time-to-market was the most
important indicator of development process performance in terms of its impact on project
success. This was followed by quality and then cost.

The final objective of this research was to identify practices (and factors) that
influence product concept effectiveness and development process performance. All the
three supply chain integration factors (supplier integration, design-manufacturing
integration and customer integration) were significant predictors to both product concept
effectiveness and development process performance. As hypothesized, concurrent
process management was also a significant predictor of development process
performance. The individual practices constituting each factor is described in Chapter 5.

The results indicate that supply chain integration mechanisms significantly impact
two mediators of project success, i.e., product concept effectiveness and development
process performance. Also, development process performance as a construct and time-to-
market as an indicator have a dominant influence on project success. Supply chain
integration mechanisms have a strong impact on project success primarily via their

intervening impact on time-to-market.

1.5 Contributions of the Study

Contributions of this dissertation to practice include the isolation of supply chain
integration mechanisms that are significantly related to project success and mediating
factors (product planning and project execution) of project success. Also, the contingent
nature of these integration mechanisms on performance are identified. Contributions to

methodology include development of a new scale (for fit with firm capability) and
7

verification of existing scales at the product development project level. All the scales
were found to have strong measurement properties (reliable, valid and confirmed factor
structures).

This dissertation bridges the gap between two established schools of rational
planning and problem solving in new product research. Using constructs from both these
streams, an integrated conceptual model was developed and tested using empirical data.
The results from the data analysis indicated strong support for the importance of

integration mechanisms as enablers of success in new product development projects.

1.5 Organization of the Dissertation

This dissertation is organized as follows. In chapter 2, based on the literature in
new product development, the framework of supply chain integration in product
development projects is introduced. In chapter 3, this framework is explained in detail
and research hypotheses are developed from the conceptual framework. Chapter 4
discusses the research design and methodology including the sampling frame and data
collection procedures. The description of the sample, psychometric properties of the
scales, and the findings from testing of the hypotheses are reported in Chapter 5. The
final chapter synthesizes the findings, discusses the primary contributions along with

limitations, and provides directions for future research.

CHAPTER 2

REVIEW OF THE LITERATURE

2.1 Introduction

The development of successful new products is a key strategic activity for most
firms. Indeed, the fortunes of firms are made and dissipated on the basis of product and
process innovation. In a survey of 700 firms (making industrial and consumer products),
Booz, Allen and Hamilton, Inc. (1982) found that over a five-year period, new products
accounted for an average of 28% of these companies’ growth. In a more recent survey
Wind, Mahajan and Bayless (1990) reported that 25% of current sales were derived from
new products introduced in the last three years. Despite this potential for superlative
growth, product development projects are inherently risky. Booz, Allen and Hamilton
(1982) found that the failure rate of new products introduced between 1963 and 1981 was
as high as 35%. Cooper (1990) later confirmed that new products continue to fail at a
high rate as they did twenty five years ago. More importantly, the reasons for the failure
of new product projects continue to baffle researchers and practitioners alike. While
there have been a number of studies that examined enabling factors that influenced
project success, such findings have not been robust enough to accumulate reliable
knowledge on new product success. Moreover, most empirical studies in the past used
relatively weak research methodologies (such as descriptive statistics, correlational
analyses etc.) to examine the research questions of interest. In response to these

shortcomings, Brown and Eisenhardt (1995) call for an integrated examination of the

links among process performance, product factors and financial performance using
multivariate data analysis techniques. They also emphasize the need to include new

enabling factors that influence project success.

2.2 Theoretical Perspectives of Product Development

Brown and Eisenhardt (1995) provided a schema for classifying the work in the
product development literature (see Table 1). Three distinct research streams were
identified: 1) rational planning; 2) communication web; and 3) disciplined problem
solving.
2.2.1 Product Development as a Rational Plan

Research in this stream have assruned that product development as a strategic
activity that has to be planned in advance, executed well and have top management
commitment to the product. This stream contains the largest portion of the reported
empirical literature. The underlying focus of this stream is to identify, in a broad sense,
the correlates of project success. The research methodologies employed were generally
weak and the works were largely atheoretical.
2.2.2 Product Development as Disciplined Problem Solving

This is the most recent stream of research. Product development was treated as an
iterative, problem solving activity with many “hit and miss” trials and errors before the
development of a successful product. Like the communication web stream, the focus on

this stream was also narrow with the objective of explaining complex phenomena such as

10

product vision, characteristics of heavy weight product leaders, and ingredients of an

effective product concept.

Table 2.1. Comparison of Research Streams in Product Development

 

 

 

Concepts Rational Plan Communication Disciplined Problem
Web Solving
Key Idea Success via superior Success via internal and Success via p-roblem
_ product, attractive external communication solving with discipline
market, rational
organization
Theory Mostly atheoretical Information and Information including
resource dependence problem solving
Methods Bivariate analysis; single Deductive and Progression from
informant; many inductive; multivariate; inductive to deductive;
independent variables multiple informants multiple informants;
single industry, global
studies
Product Product advantage-cost, _ Product integrity -
quality, uniqueness, fit product vision that fits
with core competence with customers and firm
Market Size, growth, __ _
competition
Senior management Support _ Subtle control
Project team X-functional, skilled _ X-functional
Communication High cross-functional High internal, high High internal
external - various types
and means
Organintion of work Planning and “effective” _ Overlapped phases,
execution testing, iterations, and
planning
Project leaders Politician and small Heavyweight leader

 

group manager

 

 

Customers Early involvement __ _
Suppliers Early involvement _ High involvement
Performance Financial success Perceptual success (team Operational success
(dependent variable) (profits, sales, market and management (speed, productivity)
share) ratings)
Pioneering study Myers and Marquis Allen (1971) Imai et a1 (1985)
(1969)

 

Source : Brown and Eisenhardt (1995)

ll

 

2.2.3 Product Development as a Communication Web

In this stream of research the object of investigation is typically the new product
development team and most researchers were interested in investigating the
communication pattern of members in the team and their impact on team effectiveness.
In particular, communication among team members and communication with external
members were studied. This stream of research drew strength from information and
resource dependence theories.

2.2.4 Development Process Performance — Schools of Thought

Several indicators of development process performance have been studied in the
new product development literature. Quality, time to market and cost are the most
frequently mentioned measures (Murmann,l994). In order to improve the profit
contribution of new product development projects, firms need to balance its efforts
towards three objectives - development speed, development cost and product quality
(Smith and Reinertsen, 1991). Rosenthal and Tatikonda (1993) also state that the
balanced consideration of quality, cost and lead times helps in resource deployment
decisions.

In terms of interrelationships among the indicators of development process
performance, there are two distinct schools of thought — synergy school and the trade-off
school. The synergy school argues that high performance on two or more components of
development process performance can be achieved simultaneously. The practitioner
press offers several examples of firms that have excelled in multiple measures of

development process. For example, companies such as Hewlett-Packard Co. (Hof, 1992;

12

Teresko, 1991), Honeywell Inc. (Larson, 1988 ), Intel Corp. (Zipser, 1992), and Xerox
Corp. (Raia, 1991) have reported simultaneous improvements in cycle times, product
quality, development cost and market share. Some authors have speculated that
improvements in time-to-market and quality are positively related. For example, Ittner
and Larcker (1997) report that performance benefits of accelerated product development
was greater when the perceived quality of the product design was higher. Himmerfarb
(1992) argues that concentration on speed to market forces a development team to focus
on those elements of a product design that are most related to quality. The business press
also cite the cases of Xerox and RCA (Thomson) Consumer Electronics as examples of
companies that have made simultaneous improvements in product quality and
development lead times (Raia, 1991; and Valentino and Christ, 1989).

In contrast to the synergy school, the tradeoff school posits that all development
process measures cannot be attained simultaneously. Improvements in one process
measure can only occur at a cost of another. For example, some researchers have
reported trade-off relationships between cost and time (Graves, 1989; Mansfield, 1988).
An economic model developed by McKinsey and Co. proposed that products that were
shipped six months late to market but on budget earned 33% less profit. In contrast,
products that were shipped on time but 50 % over budget earned only 4% less profit in
the same market (Dumaine, 1991).

The second series of examples of reported trade-offs in the NPD literature is
between quality and time to market. Eisenhardt and Tabrizi (1995) report that practices

such as rewarding designers for schedule which are deployed to accelerate development

13

time may actually impair quality performance as designers may not adhere closely to
product specifications or induce a high level of design quality in the product.
Conventional wisdom based on the recent experiences of US manufacturers seems to
indicate that fast product innovation and quality represent a trade-off that cannot be
simultaneously achieved (F lynn,1994). Munnann (1994) stressed that companies should
optimize their development cycle time by considering possible trade-offs between
speeding up the process and desired quality. Rosenthal and Tatikonda (1993) conducted
a series of seven case studies in the. electronics and electrical goods equipment industry
and found that four of the seven projects studied traded off time targets to concentrate
more on increasing product functionality. Several authors have contended that efforts to
reduce cycle time by skipping steps in the development cycle may lead to product defects
and manufacturing problems thereby eliminating any benefits from cycle time reductions
(Crawford, 1992; Gupta and Wilemon, 1990; Hise, O'Neal, McNeal, and Parasuraman
1989; Ulrich, Sartorius, Pearson, and Jakiela, 1993; Kessler and Chakrabarti, 1996).

The debate on the synergy versus trade-off issue is yet unresolved. Most studies
are still prescriptive and lack rigorous empirical verification. The reason why such trade-
offs or synergies occur is largely unexplained. Few studies have examined cost, quality
and time to market concurrently in their relation to NPD project success. Also, in the
NPD literature “quality” has been operationalized to include only conformance quality
and not other dimensions of product quality such as design quality. This research seeks

to address these issues.

14

 

 

 

 

J

El:

SUt

rel.

det

int:

B01

stut‘

h.)
{’J

inte
mat
Perl
indL
betv
deVe
in St

main;

2.3 Determinants of Product Development Success
2.3.1 Meta Analytic Findings

In a recent meta-analytic review of the new product literature, Brown and
Eisenhardt (1995) called for the discovery of new enabling factors that determine NPD
success. They stressed the need to be on a “discovery” mode for detecting new
relationships among the antecedent variables to NPD success. In a meta analysis of the
determinants of new product success, Montoya-Weiss and Calantone (1994) call for an
integrated examination of broad factors that affect new product development success.
Beth these reviews emphasize the need for including system wide variables in a single
study. The basis for this emphasis comes from the lack of understanding the robust links
that affect product success and the conditions under which the relationships exist.
2.3.2 The Integration Literature

A critical review of the literature suggests that some studies have employed an
integrative perspective to analyze product development success. Clark (1989) noted that
managing practices and policies in an integrated network can contribute to superior
performance in product development. In a comprehensive study 'of the world auto
industry, Clark (1989) found evidence suggesting that integration of the capability
between upstream and downstream firms is an important determinant of product
development success. However, the nature of this desired integration was not explicated
in sufficient detail. In a series of in-depth case studies of 12 main competitors in the

mainframe computer industry, Iansiti (1995) found that a high systems focus as indicated

15

by technical integration, exposure to systems integration and accumulation of interaction
knowledge predicted both lead time and productivity.

Similar recommendations can be found in the innovation literature. Shrivastava
and Souder (1987) warn that integration problems can severely inhibit new product
development and successful technological innovation. Ettlie and Reza (1992) advocated
the simultaneous use of external and internal integrating mechanisms as a way to
influence successful process innovation. In particular, they call for a market-directed
integration and integration directed at the value-added chain. -Market directed integration
places particular attention to product designs that are based on customer’ needs (Teece,
198 8). Internal integration seeks to match internal competencies with the desired
competencies to design, manufacture and introduce a customer-driven product.

The importance of integration in NPD projects is gaining attention. For example,
Kessler and Chakrabarti (1996) argue that integration enables a faster development
process. They proposed that integration can be achieved via four mechanisms: 1)
concurrentness in development process or task overlap; 2) design for manufacturability;
3) strength of firnctional norms relative to shared project norms; and 4) proximity of team
members. Using the extant literature, they suggest that faster development is associated
with development cost performance (i.e., lower values of cost are ‘better’). Similarly,
they deduce that faster development can be associated with higher product quality, when
the project is focused on customers’ needs. Finally, they suggest that faster development
processes are associated with higher project success. Thus, integration has been proposed

as a key enabler to development process performance and NPD project success.

16

2.3.3 The Concurrency Literature

The sequential approach of managing NPD projects was largely the dominant view
until recently. In this approach, the tasks from concept/idea stage to product launch
typically comprising NPD projects were done sequentially. In contrast, the concurrent
process management approach considers downstream and external requirements while
making upstream decisions. For example, both the product and downstream production
process could be designed simultaneously in the early stages of the NPD process. Within
the evolving literature relating to concurrency, two themes relevant to this research -
proactiveness and joint problem solving - are worth noting. Concurrent process
management entails being proactive on downstream oriented decisions. Krishnan (1996),
using case study data found that breaking the sequential mind set and pattern of execution
requires the concurrent management of, downstream phases with upstream phases by
using early upstream information. Thus, sharing information on an evolving or dynamic
basis appears to be critical for a proactiveness strategy. Yet, a mismanagement of the
concurrent approach can lead to expensive and time-consuming downstream rework. The
precise balance between proactiveness and reactiveness is still elusive.

Another theme underlying concurrent process management is the notion of joint
problem solving among different functional areas that are involved in the NPD process.
Swink, Sandvig, and Mabert (1996) using case study data found that in NPD projects
developing highly innovative products, joint engineering problem solving was used to a

large extent. Trygg (1993), in a survey of Swedish companies which practiced concurrent

l7

process methods found that 80% of the companies had formal joint design review
meetings among personnel from product engineering, production, marketing and quality
assurance. The performance consequences of adopting concurrent process management
are not clearly known. It has been maintained that early implementations of concurrent
process management techniques reported the benefits of lower product costs and
improvements in quality, however, more recent implementations have reported the
benefits of lowering time-to-market (Trygg, 1993). On the whole, concurrent process
management does appear to positively influence one or more components of development
process performance.
2.3.4 The Contingency Literature

Most previous work in the NPD literature have emphasized an unqualified effect of
antecedent practices on project success. In contrast, in the evolving contingency school,
the view that certain variables operate as moderators has started to emerge. Specifically,
in this school, it is proposed that the impact of practices on performance depends on the
levels of certain contingency variables. Within this literature, two types of contingencies
have been studied — market-related and product-related. Examples of market-related
contingencies include stage in the product life cycle and market uncertainty. Song and
Montoya-Weiss (1995) , in a study of 788 new products in Japanese firms found that
market uncertainty moderated the effect of several technical activities on new product
success. Eisenhardt and Tabrizi (1995) found that stage in the product life cycle
moderated the influence of supplier involvement on time-to market. Similarly, the effects

of product-related contingencies such as product innovativeness and technical complexity

18

of the product have been studied in the NPD literature. Song and Parry (1996) tested a
contingency path model that linked product innovativeness and new product success and
found that higher levels of product innovativeness actually weakened the direct links
between technical and marketing proficiency on new product success. Thus, the
contingency perspective is useful for testing several moderating influences in the
practice—-) performance relationships in the NPD context.

In sum, this dissertation proposes to use the insights relating to meta-analytic
findings and developments within three sub-streams within the NPD literature, i.e.,
integration, concurrency and contingency schools, to propose and test a holistic
framework of supply chain integration in the context of NPD projects. The framework is

briefly introduced in the next section and details are provided in the next chapter

2.4 Supply Chain Integration Framework of Product Development
2.4.1 Introduction to the Framework

As discussed in the previous section, the findings from recent meta-analyses and
recent work in the area of integration mechanisms point towards the need for “ a broad
brush” exploration of systemic links between integration mechanisms and NPD project
success. Fundamentally, this dissertation deviates from previous work in two ways.
First, the selection of integration mechanisms proposed to influence NPD project success
is guided by a focus on the supply chain. Second, integration mechanisms are
hypothesized to influence mediators of NPD project success, and not NPD project
success directly. The mediators of NPD project success that were selected are product

concept effectiveness (a planning construct) and development process performance (an

19

execution construct). This is similar to the approach taken in business strategy in which
the corporate strategy is comprised of a planning and an execution component (Andrews,
1987) 9
Recent trends in the concurrency literature support the inclusion of concurrent
process management as a key construct that impacts development process performance.
Similarly, the empirical findings in the contingency school point to the inclusion of key
moderating influences in the practice—>performance relationships in the framework. In
the interest of conciseness, only those practices (concurrent process management and
supplier integration) for which the findings are inconsistent are proposed to be tested in
this research. In summary, the proposed supply chain integration fiarnework is designed
with a view of testing new mediating relationships and controversial moderating
relationships among the practice—>performance linkages in NPD projects.
2.4.2 Summary of the Research Objectives
Formally, the purpose of this research is to examine whether supply chain integration
mechanisms have important performance implications in NPD projects. More
specifically, the research objectives are to examine the:
Impact of supply chain integration mechanisms on new product planning
Impact of supply chain integration mechanisms on new product development
process performance

0 Impact of new product planning and new product development process
performance on NPD project success

20

CHAPTER 3
SUPPLY CHAIN INTEGRATION FRAMEWORK

3.1 Definition of Supply Chain Integration

Supply Chain Integration can be defined as an approach or philosophy that
concurrently considers market-directed links and internal value chain links. Thus, supply
chain initiatives call for an interfunctional and interorganizational focus. Such an
emphasis is reflected through the deployment of supply chain management (SCM)
practices or integration mechanisms. These mechanisms collectively span the “supply
chain” and are evaluated using a systems perspective as opposed to a parochial frmctional
perspective. In this dissertation, a supply chain orientation is used as an organizing

philosophy to manage new product development projects.

3.2 Supply Chain Integration Framework

Figure 1 presents the conceptual framework as a basis for: investigating the
research questions of interest. The rationale for this framework encompasses the
following four areas:

Overall NPD Project Success (Bottom-line Impact)

Planning of the New Product (Product Concept Effectiveness)

Execution of New Product Strategy (Development Process Performance)
Integration Mechanisms

A detailed discussion of the key points in the conceptual framework follows.

21

 

 

Supplier H7 Product Concept Effectiveness ”0‘1““ Development
Integration H
n

 

Project Success (Business)

 

 

° Fit with Firm Capabilities A

 

 

 

 

 

 

 

 

 

 

‘3‘ - Market Share
Design-Manufacturing - Profitability
Integration ' - Breakeven Time
69
’0
Customer ‘9
, ’1 Product Develoth
Integration
0 Process Performance
'2
Concurrent H . Quality
Process '3 - Conformance
Management _ Design
- Time -to- Market
' Cost

 

 

 

Figure 1: Supply Chain Integration Framework

3.2.1 Product Development Project Success

Success in new product projects is usually evaluated along multiple criteria.
Managers are interested in knowing whether the entire product development process was
a success or not. This is usually captured as in terms of multiple measures of process
performance. Apart from this ‘product level’ analysis, managers are interested in the
overall impact of the development project on the business as measured by profitability,
break even point, and initial market penetration. The performance implications of a
faster development process is not consistent across studies. Rosenthal and Tatikonda

(1993) found in a series of seven case studies in the electronics and electrical goods

22

equipment industry that it is possible for firms to achieve market and technological
success despite the project not meeting its time target.

Support for the operationalization of the construct of product development project
success in terms of market share, profitability and break even time exists in the literature.
Initial market performance is an important indicator of project performance because
several environmental and market variables which are beyond the control of managers
influence long term market performance and the impact of time-to-market on long term
market performance will be difficult to isolate (Ali, Krapfel and La Bahn, 1995). The
profit contribution of the NPD project is also an important indicator of overall success.
Certain firms have been known to treat certain NPD projects as ‘training’ projects which
meant that the learning potential from such projects was more important than the
potential profitability of the project. But, for a large segment of NPD projects both
within a firm and across firms, realized profitability of the project is an‘ important project
related objective. Finally, break-even time (defined as the elapsed time from the end of
product launch to the start of making profit) has been proposed as an element of project

success (Graves, Carmichael, Daetz, and Wilson, 1991).

3.2.2 Product Concept Effectiveness (Planning of the new product)

A key consideration in the formulation of new product strategy is the precise
determination of product-related goals. This consideration has been reflected in the
literature by phrases such as “product definition”, ‘product. concept’ and ‘product
integrity’ (Rosenau, Griffin, Castellion, and Anschuetz, 1996). Product Concept

Effectiveness can be defined as the degree of fit between the product’s intended image
23

and performance and firm capabilities (Clark and Fujimoto, 1991; Brown and Eisenhardt,
1995). The effectiveness of capturing the product concept is evaluated on the basis of the
extent to which the development of the new product is congruent to the current
capabilities of the firm.

Fit with firm capabilities has been empirically shown to be correlated with project
success (Booz, Allen and Hamilton, Inc., 1982; de Brentani, 1989; Cooper and
Kleinschmidt, 1987, Duerr, 1986). Rosenthal and Tatikonda (1993) found in a series of
seven case studies in the electronics and electrical goods equipment industry that NPD
project planners overlooked the realities of resource availability, most notably those
relating to hiring of additional specialists or procurement of special materials and
. components. This lead to delays in the project because of inadequate appreciation of the
time required to meet design specifications with unfamiliar technologies and to
coordinate successfully the development work of separate groups including suppliers.
3.2.3 Development Process Performance (Execution of New Product Strategy)

Another key consideration is the execution of the new product strategy. This is
evaluated on the basis of an overall impact on the performance measures of the new
product development process. There is considerable agreement as to the different steps
that are needed to complete the development process. Typically, the following steps are
entailed: inception of a new idea, technical and marketing assessment of feasibility,
product design, lab testing, prototype development, trial runs, debugging, and mass

production. However, instead of focusing on the impact of all these generic activities on

24

process performance, this study examines the impact of integration mechanisms on
development process performance.

Traditionally, development process performance has been measured on the basis
of impact on lead times, productivity, and conformance quality. This study would
examine the additional measure of design quality because proactive consideration of this
measure of quality enhances market penetration via higher customer satisfaction. Instead
of productivity, this study would examine development cost performance. Thus, the
components of product development process performance included in this study are:
conformance quality, design quality, time-to-market and cost.

Conformance quality is the degree to which the product conforms to design and
operating needs (Garvin, 1987). It is largely a manufacturing prerogative and assessment
of quality performance is typically done afier the manufacture of the product or part.
Design quality or total quality includes everything about a product that is visible or
perceivable to the customer such as technical performance, styling or the match of the
product with the target customers’ tastes, rather than quality in the sense of
manufacturing conformance to specifications (Cusumano and Nabeoka, 1992). Time-to-
market is defined as the elapsed time from the beginning of idea generation when the firm
decides to develop a new product to the end of product launch when the product is
commercially available and managed in a routine manner (Wheelwright, 1988).
Synonymous terms that have been employed in the literature include cycle time (Urban
and Hauser, 1993), product development time (Lilien and Yoon, 1990), innovation time

(Mansfield, 1988), and lead time (Clark, 1989). Several authors have recommended time

25

to market as an important outcome of the new product development process (Ali, Krapfel
and LaBahn, 1995; House and Price, 1991; Emmanuelides, 1992; Datar, Jordan, Kekre,
Rajiv, and Srinivasan,l997). Product cost measures the unit development cost for the
new product comprising the NPD project. It is usually a quantitative measure and
fluctuations in actual figures can be attributed to reasons such as design complexity,
material composition and assembly costs. A subjective evaluation of actual product cost
performance versus budget is needed to make a comparison across diverse products in

different firms.

3.2.4 Supply Chain Integration Mechanisms

The product development literature has examined several integration mechanisms
or practices that directly or indirectly influence project performance. In a recent review
of the literature, Montoya-Weiss and Calantone (1994) found that these practices that
influenced project success can be grouped under four factors: strategic, development
process, market environment, and organizational. Later, Brown and Eisenhardt (1995)
found that seven broad factors affected product development success: suppliers, team
composition, team organization of work, team group process, project leader, senior
management, and customers. The robustness of findings relating these factors to project
success is varied.

In the conceptual framework of this research, selection of the appropriate
integration mechanisms or practices have been made with the following basis:
a An emphasis on supply chain management practices

0 An interfunctional and interorganizational emphasis .
o A focus on practices that have been less studied in the literature

26

Based on these criteria, the following integration mechanisms or practices are proposed.

Supplier Integration
Supplier integration is an important but often undervalued product development
activity (Hartley, Zirger, and Kamath 1996). Fast product developers have reported
benefits such as lead time reduction and improved quality more than their slower
counterparts through the use of supplier integration (De Meyer and Van Hooland, 1990).
Traditionally, supplier management in US companies was characterized by short term
arm’s length contracts with little or no role in design and engineering (Clark, 1989). In
contrast, Japanese manufacturers involved suppliers early in the development process,
assigned significant responsibility, and communicated extensively and directly with
product and process engineers. Companies like Sun, Tandem and Mips have recognized
that design and production of materials and equipment can be no longer accomplished by
a single firm and thus have resorted to supplier integration (Saxenian,199l). The Silicon
Valley firms caters to a fragmented computer market that consists of distinct markets for
super-computers, super minicomputers, engineering work stations, networked
minicomputers, personal computers, parallel and multiprocessor computers, and
specialized educational computers. This competitive context has led to the flourishing of
specialist systems producers and their networks of suppliers.
Research has associated faster development processes with early supplier
involvement (Gupta and Wilemon, 1990) and extensive supplier involvement (Clark and
Fujimoto, 1991; Imai, Ikujiro and Takeuchi, 1985). Clark and Fujirnoto (1991) found

that for Japanese car manufacturers supplier involvement accounted for about one-third of
27

the productivity advantages and four to five months of lead time advantages. In another
interesting contrast, Swink, Sandvik and Mabert (1996), found in a series of five NPD
projects in high tech companies, that early supplier involvement strategy was effective in
reducing overall development time for highly innovative products. Hartley, Meredith,
McCutcheon, and Kamath (1996), found that the timing of supplier’s involvement was
significantly related to perceived contribution to product development success. In the
same study, they also found a statistically significant relationship between NPD project
success and supplier involvement. A recent study found that suppliers direct involvement
in buyer’s NPD team was the single largest differentiator between most and least
successful supplier integration efforts (Ragatz, Handfield, and Scannell, 1997).

The evidence pointing to the benefits of early supplier’s involvement (ESI) in NPD
projects is not unanimous (Eisenhardt and Tabrizi, 1995). They found that the use of
early supplier involvement was positively related to time to market only in the mature
mainframe segment of the industry and not in the growing personal computer segment of
the industry. Zirger and Hartley (1996) did not find a statistically significant relationship
between supplier involvement and product development time.

In a survey of 109 Swedish companies, 79% of the responding companies
reported the use of supplier involvement in product development teams, however, this
factor did not significantly relate to development lead times (Trygg, 1993). Hartley,
Zirger, and Kamath (1996) found in a recent study that overall NPD project performance

was significantly related to the percent of suppliers activities that were completed on

28

time. Ittner and Larcker (1997) reported that supplier involvement had a significant
negative correlation with development time.

Bonaccorsi and Lipparini (1994), reported the following benefits of early supplier
involvement: lower development costs, standardization of components, consistency
between design and suppliers capabilities, and reduction in engineering changes, higher
quality with fewer defects, improvement in supplier’s manufacturing process, availability
of detailed process data, and reduction in time to market. Thus, supplier integration
appears to play an important role in new product development projects.
Design-Manufacturing Integration

A substantial portion of the new product’s committed cost in the early stages of the
NPD project can be traced back to the design-manufacturing interface. The integration of
the design and manufacturing functions within a firm can prove to be very important in
planning a new product and in execution of the development strategy (Rochford and
Rudelius, 1992). For example, exchange of information, and consideration of this
information in decision making can improve development process efficiency. The
literature has pointed to several approaches for coordinating design and manufacturing,
either with or without the use of computers. For example, CAD/CAM systems facilitate
easy storage, retrieval and processing of technical information which can lead to faster
design cycle times (Zirger and Hartley, 1994). Kodak was able to design and launch the
Fling Camera in 40 weeks compared to the usual 65 to 80 week development cycle time
mainly due to its reliance on a CAD system (Wheelwright and Clark, 1992). In a study of

72 products in the computer industry (Eisenhardt and Tabrizi, 1995) found that the use of

29

CAD was positively related to time to market in the mature mainframe segment of the
industry. Swink, Sandvik and Mabert (1996) cited case study evidence for the positive
influence of use of CAD tools to improve design quality performance. In a recent study
of Swedish firms, Trygg (1993) reported that the degree of usage of CAD/CAM was
significantly higher in groups that reported improvements in development lead times as
compared to the groups that did not.

Design for Manufacturability (DF M) is another technique used for integrating the
design and manufacturing interfaces in the new product development process. In this
technique manufacturing knowledge for each design decision is considered in advance in
order to avoid costly iterations and changes later on in the process (Millson, Raj and
Wilemon, 1992; Urban and Hauser, 1993). DEC, Xerox and Motorola are some
examples of companies that have reported significant reductions (35 to 75%) in the
number of components and reductions (53 to 85%) and in assembly costs due to the use
of DFM techniques (Brannan, 1990; Lewis, 1986; Welter, 1990). In a series of case
studies in the German mechanical engineering industry, Murmann (1994) found that
consideration of early manufacturability of the design improved development lead times.
Swink, Sandvik and Mabert (1996) found that design for manufacturability positively
influenced product cost performance in several case studies of companies in high tech
industries.

Quality Function Deployment (QFD) is another technique frequently used to
integrate design and manufacturing functions in the NPD process. It is a design approach

that ensures that the product meets the customer requirements when it goes into

30

production by a systematic focus on customer attributes throughout the development
process (Hauser and Clausing, 1988; Sullivan, 1986; Griffin, 1992). QFD also leads to
more fully specified designs so that changes can be made early in the design process. In a
recent study of Swedish firms, Trygg (1993) reported that the degree of usage of QFD
was significantly higher in groups that reported improvements in development lead times
as compared to the groups that did not. Thus, design-manufactming integration appears

to play an important role in new product development projects.

Customer Integgtion

Customer integration practices help in ascertaining the needs of the customers and
translate these needs into product and process parameters in the NPD process. Customer
oriented practices have been reported to have organization-wide effects. For example,
Ruekert (1992), in an empirical study of employees from different functional
backgrounds in five strategic business units of a large diversified company, found a
positive relationship between market orientation practices and firm performance. In the
NPD literature, Atuahene-Gima (1995), in a study of 275 Australian firms, found a strong
positive relationship between market orientation activities and NPD project performance.
In this study, market orientation activities were also found to influence development
process performance. Later, Gatignon and Xuereb (1997) found a contingent influence of
customer orientation on NPD performance. Under conditions of high demand
uncertainty, customer orientation practices were found to have the highest impact on
NPD performance. In this competitive situation, practices such as customer information

generation, dissemination of this information, and detecting degree of responsiveness to
31

customers is highly critical (Jaworski and Kohli, 1994). Thus, customer integration

activities have key influences in new product development projects.

3.2.5 Concurrent Process Management

Concurrent process management is a systematic approach to the integrated,
concurrent design of products and their related processes, including manufacture and
support (Winner, Pennell, Bertrand, and Slusarczuk, 1988; Nevens, and Whitney, 1989).
This approach is intended to cause the developers from the outset, to consider all
elements of the product life cycle from conception through disposal, including cost,
quality, schedule and user requirements. Case study examples of successfirl
implementation (savings of more than 60% in design cycle time) of concurrent
engineering programs include Boeing, IBM, HP and Deere and Company (Zangwill,
1993). Overlapping of parts and process design has been reported to decrease
manufacturing costs (T akeuchi and Nonaka, 1986). In a survey of 109 Swedish
companies, 89% of the responding companies reported that the intent of deploying
concurrent engineering practices was to reduce product development time (Trygg,l993).
Other motives include : 1) develop more customized products; 2) 'to reduce product
development cost; and 3) to increase suppliers involvement in NPD project.

An efficient organization of concurrent development processes have been shown
to be more effective in shortening development cycle times (Eisenhardt and Tabrizi,
1995). Paralleling new product development activities has often been recommended as
one of the approaches to reduce development time in different industries (Clark and

Fujirnoto, 1991; Cordero, 1991; Millson, Raj and Wilemon, 1992, Smith and Reinertsen,
32

1991). Munnann (1994) in a series of case studies of German companies found that
almost all companies studied tapped the potential of either total parallelism between
development activities or a certain degree of overlapping of activities. For example, he
found that almost all development activities could be started when the previous activity
was completed to the extent of 80%. Rosenthal and Tatikonda (1993) conducted a series
of seven case studies in the electronics and electrical goods equipment industry and found
that reductions in NPD cycle time were achieved through a combination of reducing
nonproductive time and increasing the overlap with which tasks are performed. All these
studies point to the importance of concurrent process management practices in NPD
contexts.

The view that concurrent process management might have a contingent influence
on development process performance, as opposed to an unqualified direct impact is
starting to evolve. For example, Eisenhardt and Tabrizi (1995) found that overlapping
development activities has a significant influence on time-to-market only for new
products developed in mature segments as opposed to new products developed in growth
segments. Similarly, the level of product innovativeness has a moderating influence on
the concurrent process management —) time-to-market link. For example, Ali et al
(1995) discovered that concurrent process management lengthens cycle time in
environments reflecting high levels of product innovativeness. The common theme
across these studies indicates that concurrent process management practices have a
differential impact on time-to market that depends on moderating influences such as

product innovativeness and stage in the product life cycle.

33

The justification for including the different constructs in Figure 1 was discussed in
the previous sections. The basis for linking the constructs as is shown in Figure 1 will be
discussed under individual hypotheses in the next section. First, the research questions

are introduced followed by the discussion of individual hypotheses.

3.3 Research Hypotheses
The conceptual framework in Figure l and the discussion in the previous section
suggest the following research questions of interest:

0 What is the impact of product concept effectiveness and components of
development process performance on success of NPD projects?

0 Which supply chain integration mechanisms have a significant influence on
components of product concept effectiveness?

0 Which supply chain integration mechanisms have a significant influence on
components of development process performance?

0 Do the variables of product innovativeness, market uncertainty, supplier
involvement, and technical complexity moderate the influence of supply chain
practices on components of development process performance?

First, representative hypotheses concerning the relationships among product
concept effectiveness, development process performance and project success are stated.
Next, hypotheses for the practices-product concept effectiveness and practices-process

performance links are explored. Finally, hypotheses relating to the moderating influences

are shown.

34

 

 

pi

is

0\

W0

 

fail

AC(

ere
SUCi
incl
has
Proj
‘0 d.
incre

freed

3.3.1 Antecedents to Project Success

In an examination of the empirical literature on the determinants of new product
performance, Lilien and Yoon (1989) cite ‘fit of the new product’ to the business context
as one of the major determinants of NPD project success. One aspect of the fit discussed
in this paper is product fit to firm capabilities. Specifically, they argue that levels of
production and marketing expertise are directly related to growth in market share, which
is one aspect of NPD project success. Other researchers have also pointed out the
importance of assessing and planning for product fit to the capabilities of the firm.
Overlooking realities of resource availability can impair the coordination of development
work among different functional groups involved in the NPD process and lead to the
failure of projects (Rosenthal and Tatikonda, 1993; Sioukas, 1995; Ali et al., 1995).
Accordingly, it is proposed that:

Hypothesis 1. Fit with firm capabilities is positively related to overall success in
NPD projects

Apart from product planning, (i.e., assessing product fit to firm capabilities),
execution of the development process is just as important in its impact on NPD project
success. In this dissertation, development process performance is operationalized to
include: conformance quality, design quality, time-to-market and cost. Previous research
has found positive relationships between development process performance and NPD
project success. For example, Smith and Reinersten (1991) report that shorter cycle times
to develop and introduce new products increases sales through extended product lives,
increases market share through pioneering, and increases profitability through pricing

freedom and economies of scale. Similarly, Rosenau (1990) found that faster
35

development was associated with project success defined in terms of higher sales and
profits.

Individual positive relationships between quality, time and cost, with project
success has also been reported in the literature (see Larson, 1988; Zipser, 1992; Flynn,
1994; Ittner and Larcker, 1997). Contrary findings in the relationship (negative) between
components of development process performance and project success is also reported
(Murmann, 1994; Eisenhardt and Tabrizi, 1995). As there is more consistency in the
pattern of positive relationships, the following are hypothesized:

Hypothesis 2. Conformance quality performance is positively related to overall
success in NPD projects

Hypothesis 3. Design quality performance is positively related to overall success
in NPD projects

Hypothesis 4. Time to market performance is positively related to overall success
in NPD projects

Hypothesis 5. Development cost performance is positively related to overall
success in NPD projects
Finally, product concept effectiveness is proposed to influence development
process performance. On a temporal scale, this relationship makes sense because an
early assessment of fit between existing capabilities of the firm and desired capability
level as dictated by the new products is necessary for obtaining superior development
process performance (see De Brentani, 1989; C00per and Kleinshmidt, 1987; Duerr,
1986). Accordingly, it is suggested that:

Hypothesis 6. Fit with firm capabilities is positively related to development
process performance in NPD projects

36

3.3.2 Antecedents to Product Concept Effectiveness

The literature has suggested a number of practices or enablers of capability
development which can be useful in managing new product development projects. In this
research, the practices of supplier integration, design-manufacturing integration, and
customer integration, which span the supply chain, are hypothesized to affect product
concept effectiveness which is measured as the product’s fit with firm capabilities.
Support for such a supply chain perspective can be found in the works of Ettlie and Reza
(1992) and Shrivastava and Souder (1987). The fit with firm capabilities construct can be
viewed as a measure of capability performance. In this sense, the next set of hypotheses
examine the capability enabler - capability performance relationships. Thus, it is
hypothesized that:

Hypothesis 7. Supplier integration practices are positively related to fit with firm
capabilities

Hypothesis 8. Design-manufacturing integration practices are positively related
to fit with firm capabilities

Hypothesis 9. Customer integration practices are positively related to fit with firm
capabilities

3.3.3 Antecedents to Development Process Performance

In this research, the supply chain practices of supplier integration, design-
manufacturing integration, customer integration, and concurrent process management are
hypothesized to affect development process performance. Support for a positive
relationship between these individual supply chain practices can be found in the literature
(see Clark and Fujimoto, 1991; Cordero, 1991; Handfield, 1994; and Hartley et al., 1996).

A smaller subset of works have found that certain supply chain practices such as supplier
37

integration and concurrent process management practices can actually impair
development process performance (Eisenhardt and Tabrizi, 1995; Crawford, 1992).
Based on the stronger support of the positive relationships, it is proposed that:

Hypothesis 10. Supplier integration practices are positively related to
development process performance

Hypothesis 1]. Design-manufacturing integration practices are positively related
to development process performance

Hypothesis 12. Customer integration practices are positively related development
process performance

Hypothesis 13. Concurrent process management practices are positively related

to development process performance
3.3.4 Hypotheses relating to Contingency Relationships

While the relationships discussed so far examine the mediating relationships
between enablers of project success, and project success, the following section
investigates potential moderating influences of supply chain practices on development
process performance.
Product Innovativeness

The role of product innovativeness in NPD projects deserves further investigation.

This is so because of inconsistent findings in the literature. Products that are high and
low in innovativeness, have been found to be more successful as compared to moderately
innovative products (Kleinschmidt and Cooper, 1991). In contrast, Ali et al (1995)
postulated that increasing product innovativeness tends to lengthen cycle time. This is so
because higher levels of innovativeness require a “trial and error” approach before the

final product concept is crystallized. In contrast, products with low levels of product
38

innovativeness are more stable and require less coordination between different functions
involved in the NPD process and thus are amenable to process concurrency. The
common theme across these studies seem to indicate that concurrent process management
practices have a differential impact on time-to market that depends on the level of product
innovativeness. Accordingly, we hypothesize that:

Hypothesis 1 4: Concurrent process management practices have a positive

influence on time to market only for products that are low on levels of product
innovativeness.

Technical Complexity
Technical complexity of a product is defined as the extent to which new
knowledge, materials, or components are embodied in the product (Capon and Glazer,
1987). Products that are high in technical complexity could introduce complications into
the development process and therefore tend to lengthen time to market. In fact, Ali et a1
(1995) found that technical complexity had the strongest delaying effect on cycle time for
developing new products. This is also consistent with the finding of C00per (1979) in a
study of new products in the chemical industry. Practices geared to accelerate product
development such as concurrent process management may have a lower effect on cycle
time for products that are highly technical. Thus, it can be argued that:
Hypothesis 15: Concurrent process management practices have a positive

influence on time to market only for products that are low on technical complexity
of the products.

Market Uncertaing

Market uncertainty or risk may also moderate the impact of concurrent process

management practices on time to market performance. Support for this view can be

39

found in Cooper and Kleinschmidt (1994) who report that in markets with higher levels
of competitiveness (and thus, uncertainty), the impact of acceleration practices on cycle
time is adverse. Customer driven processes cannot be hastened because the needs of the
market are also slowly evolving in this kind of a context. In contrast, markets with lower
levels of uncertainty may be prime candidates for employing concurrent process
management practices and competing on the basis of time to market. Thus, it is proposed
that :

Hypothesis 1 6: Concurrent process management practices have a positive

influence on time to market only for markets with lower levels of market
uncertainty.

Stage in the Product Life Cycle
Apart from concurrent process management, there is growing support for the
contingent influence of supplier integration practices. For example, stage in the product
life cycle has been shown to have an important bearing on the usefulness of the strategy
of early supplier involvement as a means to improve time to market. Early stages of the
product life cycle are characteristic of high uncertainty and complexity. In this stage,
early involvement of suppliers can actually hamper development lead times (Eisenhardt
and Tabrizi, 1995). Similar findings are reported by other researchers (Hartley et al.,
1997; Birou and Fawcett, 1994). In sharp contrast, Eisenhardt and Tabrizi (1995) found
that the use of early supplier involvement was positively related to time to market in the
mature mainframe segment of the industry. Thus, it is proposed that :
Hypothesis 1 7 : Early supplier involvement practices have a positive influence on

time to market only for products that are in the maturity stage of the product life
cycle. ‘i

40

In this chapter, the research framework of supply chain integration was
introduced, key constructs were operationalized and hypothesized relationships among
the constructs were presented. In the next chapter, the research design and methods used
to develop the survey instrument are described. Details regarding data collection

procedures are also provided.

41

CHAPTER 4

RESEARCH DESIGN AND METHODOLOGY

In the previous chapters, the research framework of supply chain integration was
introduced, key constructs were operationalized and hypothesized relationships among
the constructs were presented. The purpose of this chapter is to describe: (1) the research
design including the sampling frame and unit of analysis; (2) the methods used to develop

and pilot test the survey instrument; and (3) the data collection procedures.

4.1 Research Design

The research framework in this dissertation is based on the new product
development project as the unit of analysis. The characteristics of new products have
substantial variation across different industries. Indeed, within a single industry also,
there could be considerable variation. As the research questions in this dissertation are
exploratory with reference to the current state of knowledge, care was taken to ensure
considerable variety among the industry groups with respect to the stage in the overall
industry cycle. For the purpose of discussion, two ends of the continuum from
introduction stage to maturity stage, are worth noting. Firms in the introduction stage
were selected from industries such as electronics, electrical equipment, semiconductors
and medical instruments. On the other hand, firms in the mature segment of the product
life cycle were selected from the auto industry (including auto components industry),

chemicals industry and pharmaceutical industries.
42

The purpose of this research was to identify relationships among concepts that have
received limited attention in the literature. In the realm of the philosophy of science, this
approach is called the ‘discovery’ mode of creating knowledge (McCall and Bobko,
1991). In particular, McCall and Bobko (1991) point to the usefulness of large sample
studies for discovery. Accordingly, a multi-industry population that provides a variety of
contexts was deemed appropriate for selecting the sampling frame. As noted by Bobko,
Karren and Kerkar (1987), an orientation towards discovery facilitates the inquiry into
research questions that seek to find reasons for variability. One of the principal
objectives of this dissertation is to understand variability in success rates of new product
development projects. Especially at the new product development project level, the
variety across industries and within industries (indeed, within the same company) is
considerable. Also, a multi-industry sample facilitates testing of contingency
relationships such as the moderating roles of stage in the product life cycle, market
uncertainty and product innovativeness on concurrent process management» time-to-

market relationship.

4.1.1 Sampling Frame

The sampling frame was firms in “high tech” industries which was defined to
include firms belonging to the Standard Industrial Classification (SIC) codes: 2833 to
2836; 2873 to 2875; 2879, and all industries under general SIC 35 and 38. Examples of
industries included: Chemicals, Industrial Equipment, Computers, Electronics and
Electrical Equipment, Motor Vehicles, Scientific Instruments and Medical Devices.

Firms belonging to these SIC codes were chosen because they represent industries
43

 

 

 

invol
techn
proje
apprc

firms

projec
years
been

invest
Wont
Chalq
more
SPCCif
fTom 1
0f ana

be use.

involved in the manufacture of relatively high value-added and in most cases high
technology products. The desired range for the final sample size was between 200 to 250
projects. Based on a conservative estimate of five percent as the response rate to surveys,
approximately 5000 firms would have to be contacted. As it can be expected that several
firms, especially leading edge firms, pursue multiple NPD projects, the final number of
firms that need to be contacted was expected to be less than 5000 firms.
4.1.2 Unit of Analysis

The unit of analysis was to be a recently completed new product development
project. Respondents were asked to identify a completed project within the last three
years for which at least one year had elapsed since the launch of the new product. As has
been noted in all three recent meta-analyses on product innovation, researchers have
investigated product innovation issues using the firm or project as the unit of analysis
(Montoya-Weiss and Calantone, 1994; Brown and Eisenhardt, 1995; Kessler and
Chakrabarti, 1996). For this research, treating the project as the unit of analysis makes
more sense because practices and their influence on project success tend to be more
specific and readily identifiable than the firm as the unit of analysis. Also, the findings
from this research can be compared to similar works that have used the project level unit
of analysis. Finally, this research can contribute to the development of scales that could

be useful in future research which treat the project as the unit of analysis.

4.2 Research Methodology

4.2.1 Survey Development

To investigate the proposed research questions and associated hypotheses, a
survey instrument was developed (see Appendix 1). The items in the instrument were
developed by resorting to available scales (to the extent possible) from purchasing,
Operations, marketing and NPD literature. The questions for the constructs in the
conceptual framework were asked largely using qualitative or perceptual scales (IO-point
Likert scale). A few quantitative measures pertaining to size and cumulative person
hours spent on the project were also included in the questionnaire. While it is desirable to
collect objective performance data, in the interest of obtaining a high response rate, very
few items that asked for quantitative or “hard data” were included in the survey
instrument. This problem is especially relevant for new product development projects
which are usually considered to be highly proprietary and confidential. Therefore, most
researchers who have tried to obtain “hard data” in the past had to settle for a low sample
size. In this research, an approach of “casting a wide net” to gather data from a variety of
contexts was deemed more appropriate.

The lO-point Likert scales was used as opposed to the popular 5 or 7 point scales
in order to minimize the response time and effort to the managers. Managers can easily
relate performance on an item when it is expressed as a percentage. A similar approach
has been followed in the NEWPROD project questionnaire which was first developed by

Cooper (1979) and then used in several empirical projects since then. In addition to items

45

relating to the conceptual model, descriptive information at the project level and
information relating to the following (moderating) variables were also included:

Technological Intensity of Product (Incremental versus Breakthrough)
Product Innovations (Low/High)

Market Uncertainty (Low/High)

Stage in the Product Life Cycle (Introduction/Growth/Maturity)

Input of experienced faculty members with research interests in NPD fiom several
universities was sought for comments and suggestions for improving the survey
instrument. The instrument was then pilot tested on product development leaders in three
companies. The three companies varied in terms of the industry (Medical Devices,
Aerospace Equipment and Automotive) and type of product (ICU unit, navigation control
system and luxury car) being developed. Interviews in two of the three companies were
carried out on site. For the third company, the survey instrument and accompanying
information on the research project was mailed to the project leader who filled out the
survey and provided written feedback via fax and e-mail.

The general form of discussion for the on-site interviews progressed through a
“fimneling process” (see Handfield and Ghosh, 1994). First, the general market for the
new product as perceived by the firm were discussed. In all the cases, this led to a
discussion of a specific new product that was launched recently. This was followed by a
discussion of the key indicators of new product development project success that the firm
emphasized. The relationships among the three performance indicators of NPD projects,
i.e., cost, time-to-market and quality as perceived by the firm was then discussed. Next
the discussion centered around the role of product planning (specifically, the fit with firm

capabilities) on project success. Finally, the role of integrative mechanisms were

46

discussed. Once the project leader was focusing on a particular project, the discussion
was allowed to become less directed. However, all of the issues described above were
covered in all the three cases.

The discussion started broadly at the firm level but ended in discussions on specific
project related practices that were implemented by the firm. The on-site interviews took
approximately 90 minutes to complete. One of the site participants distributed the survey
instrument within the company to project leaders in other divisions. In all ten completed
surveys were received as part of the pilot study. All the participants in the pilot study
were specifically asked to comment on: (1) items and practices that are not currently in
the instrument but ought to be included; 2) items and practices that do not belong and
ought to be excluded; (3) understandability of terms used (and directions) in the
instrument; (4) the “doability” of the project, i.e., whether the scope of the project as
indicated by the length and complexity of the questionnaire was narrow or broad; (5) the
relevance and usefulness of the research project; and (6) the face validity of linkages
shown in the supply chain integration framework (Figure 1). Each project leader
completed the questionnaire and provided feedback on the above issues. The feedback
from the participants indicated strong interest in the project, high content validity of the
conceptual fiamework, but considerable doubt about the scope as indicated by the length
of the instrument. The pilot survey instrument was modified based on this feedback. For
example, sections that were not related to the conceptual framework of Figure 1, but were
included for post-dissertation research projects were eliminated. Long sentences and

“wordy” phrases were removed. In some cases, terms were followed by short

47

explanations to improve clarity. The revised instrument was then used for large—scale

data collection.

4.2.2 Data Collection Procedures

For the large scale mail-out, an initial contact list of approximately 5000 firrns
was generated from two groups of sources — membership lists in professional associations
and commercial mailing lists. The membership list of the following three professional
associations were used: National Association for Purchasing Management (NAPM),
Product Development Management Association (PDMA) and Institute of Electrical and
Electronics Engineers (IEEE). For the commercial mailing list, a customized list of
potential firms to be included was generated using the following criteria: 1) firms should
belong to the SIC Codes 35 to 38; 2) the number of employees should exceed 100; and
3) the annual sales of firms should exceed one million dollars. The initial contact person
for the professional organizations were high level executives within their functional areas.
The typical titles included: Director of Purchasing for NAPM; Director of Product
Development for PDMA and Director of Engineering for IEEE. For the commercial
mailing list, the initial contact persons were specified to include top level executives in
technology, R&D and product development. Typical titles of contact persons included:
Chief Technology Officer (CTO), Director of Research and Development, and Director of
Product Development. While compiling the initial list, care was taken to remove fiom
the list service firms, consulting organizations, purely research and development

organizations (i.e., no manufacturing), and academic institutions engaged in research.

48

Duplication of firms across the four sources which had the same contact location were
removed. In such cases, the contact person closest to the title of product development
was included and the other contacts within the same firm and location were removed.
From the combined contact list of 5,000 firms, 1,500 firms were selected at random
(without replacement) for the initial mailing. The initial mailing included a personalized
cover letter describing the research project, a postage-paid participation form which asked
for contact details of the respondent (which also allowed contact person to decline
participation) in case the firm agreed to participate, copy of the survey, and a postage-
paid return envelop. The cover letter and survey described the ideal respondent and
characteristics of the new product development project which qualified for inclusion in
the research (see Appendix). F inns were promised an executive summary of the results.
Firms which had not developed a new product and/or launched a new product recently
were not included in the research project. Similarly, firms which had launched products
very recently and had not been in the market for at least one year since launch were
excluded. Out of the 1,500 firms contacted, 1,396 could be reached (104 letters were
undelivered because of either change of address or because the contact person had left the
firm) of which 57 firms indicated in the participation form that they did not wish to
participate in the project. Fifteen surveys were returned because the firms had not
recently developed or launched a new product. Thus, the effective contact list was for
1,324 firms of which 346 firms responded and returned the survey, constituting a
response rate of 26. 1%. Of the 346 surveys that were returned, eight surveys could not be

used on account of excessive missing data. Therefore, responses from 338 fums (more

49

specifically new product development projects) were used to test the hypotheses in this
research.

To detect non-response bias, a test was conducted to see if there were differences
between late respondents and early respondents in terms of variables germane to the
conceptual model (Armstrong and Overton 1977; Lambert and Harrington, 1990). The
sample of respondents was split into two categories, early and late respondents on the
basis of date of receipt of the survey. The first 10% from the early respondents group and
the last 10% from the late respondents group were selected for carrying out the test. A
random sample of eight items was selected from the survey instrument and values for the
eight items for respondents in both the groups were compared. The items selected for
comparison and the results of the t-test comparing the sample means for each of the eight
items across the two groups are shown in Table 4.1. As can be seen from this table, there
was no statistical significance between the means for the eight items across the two
groups indicating that non-response bias is not a problem in this study.

This chapter began by describing the sampling frame and unit of analysis for this
research and then described the methods used to develop and pilot test the survey
instrument. Finally, the procedures used for collecting the data were discussed. Chapter
V presents the data analysis and findings of this research. In Chapter VI, the findings are

synthesized and contributions from this research are identified.

50

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 4.1 : Assessment of Non-Response Bias
Description of Item Category Mean t-value Significance
Level

Supplier Integration — Direct Early Respondent 6.85
Communication with supplier .319 .751

Late Respondent 7.03
Design-Manufacturing Integration - Early Respondent 4.83
Translating customer preferences into 1.061 .293
process parameters Late Respondent 5.56
Customer Integration - including Early Respondent 3.53
customers in decisions concerning .146 .884
manufacturing Late Respondent 3.44
Customer Integration - Changing Early Respondent 7.24
product standards to the benefit of .175 .862
customers Late Respondent 7.32
Concurrent Process Management - Early Respondent 6.18
Information made readily available to .515 .608
manufacturing as product design Late Respondent 6.47
evolved
Fit with Firm Capabilities (Planning Early Respondent 7.91
Stage) - Prior experience in designing .162 .872
similar products Late Respondent 8.00
Fit with Firm Capabilities (End of Early Respondent 7.70
Project Stage) — Availability of .496 .622
manufacturing capacity Late Respondent 7.96
Fit with Firm Capabilities (End of Early Respondent 7.15
Project Stage) — Ability to handle .475 .636
major design changes Late Respondent 7.41

 

51

 

CHAPTER 5
ANALYSIS AND FINDINGS

The purpose of this chapter is to describe the characteristics of the sample, assess
the quality of data using reliability and validity analysis, and report the findings relating
to each hypotheses. Findings from hypotheses which were not originally proposed but
are logically connected to existing hypotheses are marked as “post-hoc”. The synthesis
of findings in Chapter VI includes the findings from post-hoe analyses. Consistent with a
recent meta analysis of empirical studies in product development projects, which found
that researchers focused on two kinds of links -- between practice and performance, and
between construct and performance — interchangeably, this research also investigates both
types of linkages (see Montoya-Weiss and Calontone, 1994). Thus, both exploratory

factor analysis and regression analyses techniques are used in this research.

5.1 Introduction

In this chapter, the frequency distribution of key descriptor variables of the
research sample are reported first. This is followed by an analysis of the measurement
properties (i.e., reliability analysis and validity analysis) of the constructs germane to the
supply chain integration framework. In the final section, the findings relating to each
research question are presented. Within each research question, pertinent post-hoe
analyses that were carried out are also reported. For example, the overall fit of the
conceptual model in Figure l was tested using confirmatory factor analysis. Also, the

dimensionality of the constructs in Figure 1 was tested using exploratory factor analysis.

52

It may be noted that there are very few studies that have proposed or tested the
dimensionality of the constructs of Figure l in previous research. For example, the
dimensionality of the three integration mechanisms examined in this research - supplier
integration, design-manufacturing integration and customer integration - has not been
examined in previous research. Thus, exploratory factor analysis techniques followed by

post-hoc confirmatory factor analysis was carried out.

5.2 Description of the Sample
5.2.1. Company Background

The characteristics of the companies in the research sample in terms of age, size
(as measured by number of employees) and industry background are reported. The new
product projects that comprise the research sample were developed, for the most part, in
large and more experienced companies. As can be seen from Table 5.1, 77% of the
sample were in business for 14 or more years. Also, approximately half of the sample

consisted of firms that had 500 or more employees (see Table 5.2).

Table 5.1. Frequency Distribution of Company Age

 

 

 

 

Age in Years Frequency Percentage
Less than 6 21 6

6 to 14 57 17
more than 14 257 77
TOTAL 335 100

 

 

Three cases did not provide valid responses

53

Table 5.3 lists the industry background for the projects in the sample. As can be
seen from this table, a vast majority of the projects were in high tech industries such as
electronics, medical instruments, telecommunication equipment and computers. Projects
in more mature industries such as automotive, chemicals, machine tools, furniture, metal

alloys and food are indicative of industry representation of the rest of the sample.

Table 5.2. Frequency Distribution of Number of Employees

 

 

Number of Employees Frequency Percentage
Less than 50 36 ll

51 to 200 71 21
201 to 500 69 20
501 to 2000 58 17
more than 2000 104 31
TOTAL - 338 - 100

 

 

Table 5.3. Frequency Distribution of Industry Background of the Projects

 

 

Industry Frequency Percentage
Automotive and Automotive Components 34 1 1
Chemicals 10 3
Computers 15 5
Electronics 140 43
Machine Tools 22 7
Medical Equipment 49 15
Office Equipment 9 3
Telecommunications Equipment 26 8
Others 17 5
TOTAL 322 100

 

Sixteen cases did not provide valid responses

54

To summarize, companies in this research sample were typically: in high-value
added and short product life cycle types of industries, older companies and large sized
companies. However, since the unit of analysis of this research is a new product
development project, there could be considerable variation in the project profile within
these large companies. After discussing the profile of respondents, the profiles of the
project are discussed.

5.2.2. Respondent Profile

In the survey instrument, the ideal respondent was identified as follows:

I a person with overall responsibility for the development project

I a person who was involved in the project fiom start to end of the project

I a person who interacted with upper management and project personnel for key project
decisions

Table 5.4. Frequency Distribution of Respondent's Functional Background
I

 

 

 

 

 

— -——-_-
Primary Role Frequency Percentage
Program Management 185 55
Design Engineering 70 21
Manufacturing Engineering 24 7
Marketing 15 5
Others 42 12
TOTAL 336 100
- E

 

 

 

Two cases did not provide valid responses

The survey respondents were expected to have leadership responsibility for development
projects. Table 5.4 indicates that a majority of respondents (55%) had specific leadership
responsibilities relating to product development initiatives. Design engineering was the

next highest representative function in the sample. This suggests that design function is

55

extremely important for product development projects in the companies studied. This is
not unusual considering that the industry background of a majority of the companies in
this sample, i.e., industries with intense competition based on new products with
innovative designs. The next highest category is manufacturing engineering functional
background. The sample also consisted of respondents with titles (classified as others in
Table 5.4) that indicated specific roles in product development or technology
development. Representative titles included: Director of Product Development, Chief
Technology Officer, Senior Product Development Manager, Development Engineering
Manager and Vice-President of Product Development. In a few cases, usually medium
sized private companies, the owners or presidents of companies were the respondents.
5.2.3. Project Characteristics

In the survey instrument, guidelines for selecting the new product development
project, i.e., the unit of analysis, were identified as follows:

“Pick a new product development project completed in the last three years for

which you were a project manager. By new product we mean any product that is

new to your company or division including innovations and new product lines.”
One measure of gauging scope of the development project is to measure cumulative
person hours worked on the project. As can be seen in Table 5.5, the scope of projects
ranged from low (less than 4000 hours) to high (more than 250,000 hours), with an
approximately uniform distribution.

Another project characteristic for which data was collected was the location of the

new product in the product family stream ranging from incremental changes to new

product family platforms. In the new product development literature, the end point of

56

new product family is commonly identified as “really new products”. As can be seen
from Table 5.6, 91% of the sample comprised of really new products. Very few products

in this sample were minor extensions of existing product families.

Table 5.5. Frequency Distribution of Cumulative Person Hours Worked on New Product Development Project
(in thousands)

 

 

 

 

 

Cumulative Frequency Percentage
Person Hours
(in ‘000)
l to 4 49 15
5 to 10 49 15
11 to 30 64 20
31 to 60 41 13
61 to 100 31 10
101 to 250 30 9
more than 250 58 18
TOTAL 322 "- 100
= a m

 

Sixteen cases did not provide valid responses

Table 5.6. Frequency Distribution of Location of Product in its Product Family Stream
I

 

 

_=
Location of Frequency Percentage
this Product in its
Product Family Stream
Superficial Change I .3
Minor Extension 20 9
Major Extension 100 31
New Product Family Platform 200 60
TOTAL 331 100
w E w

 

 

Seven cases did not provide valid responses

57

Data on the overall cycle times of the new product development project as
measured by the duration from product concept to first shipment to the customer were
also collected in this research. As shown in Table 5.7, most projects had relatively low
cycle times. For 72% of the projects, the cycle times were less than two years. In the
extreme, there were a few projects that had very high cycle times (more than five years).
Table 5.8 reports the frequency distribution of the duration since product launch. As can
be seen from this table, most of the products in the sample were in the market for a

duration of less than two years since the launch of the new product.

Table 5.7. Frequency Distribution of New Product Development Project Duration
1

 

_ I

Duration of New Product Development Frequency Percentage

Project from Product Concept to
First Customer Shipment

 

 

 

(in months)

I to 12 109 32

13 to 24 133 40

25 to 36 63 I9

37 to 48 19 6

49 to 60 6 2

more than 60 6 2
TOTAL 336 .- 100

 

 

Two cases did not provide valid responses

The final descriptor of project characteristics is the unit sale price of the new product.
Table 5.9 shows the frequency distribution of the unit sale price and it is apparent from
this table that there is considerable variation in the price levels of each product. It may be

noted that for the most part, the unit sale prices reflect initial pricing and not pricing as an

58

adjustment to the long term impact of competitive conditions such as rival offerings,

diffusion of product innovation, promotion and so on.

Table 5.8. Frequency Distribution of Duration in Market since Product Launch

Duration of New Product in Frequency Percentage
Market since Launch
(in months)

 

l to 12 198 59
13 to 24 84 25
25 to 36 35 ll
37 to 48 10 3
more than 48 6 2
TOTAL ““333“- "”100.“

 

 

Five cases did not provide valid responses

 

Table 5.9. Frequency Distribution of Unit Sales Price of Product
= J ~— —— — 7v ~

 

 

 

=—

Sales Price of Frequency Percentage

A single unit

(in US 8)

1 to 101 47 14

11 to 100 61 18

101 to 1000 71 22

1001 to 10000 62 19

10001 to 100000 49 15

100001 to 1000000 28 9

more than 1000000 11 3

TOTAL "SEE” ""736”
............ ....

 

Nine cases did not provide valid responses

59

In sum, the pattern of project related characteristics of the research sample indicate that

most projects had:

I developed and introduced really new products as opposed to incremental products
I low overall cycle times (less than two years)
I products that were introduced in the market relatively recently (less than two years)

On the other hand, the pattern in terms of the scope of projects and unit sales price of the
product was less distinctive and had considerable variation across the sample.
5.2.4. Descriptive Statistics of Contingency Variables

Data were also collected on four (contingency) variables - technical complexity of
the new product, level of product innovativeness, level of market uncertainty, and stage in
the product life cycle for the targeted product — that were pertinent to the research
objectives of this dissertation. The frequency distributions for the four contingency
variables are reported below. Table 5.10 lists the frequency distribution for technical
complexity of the new product. As can be seen fiom this table, a majority of the new
products are technically complex (97% of the sample). Only in 3% of the cases, the level

of technical complexity is low.

Table 5.10. Frequency Distribution of Technical Complexity of New Product

 

 

 

 

 

Technical Complexity of Frequency Percentage
New Product
Low 10 3
Moderate 1 18 35
High 209 62
TOTAL -m337“ 100

 

 

One case did not provide valid response

60

Table 5.11 displays the frequency distribution of the level of innovativeness in the new
product. The pattern in this table suggests that the levels of innovativeness range from
moderate to high (92% of the sample), with very few of the products being of type which

are low in innovativeness (8% of the sample).

Table 5.11. Frequency Distribution of Level of Product Innovativeness for Intended Market

 

 

Level of Product Frequency Percentage

Innovativeness for

Intended Market
Low 26 8
Moderate 129 38
High 181 54
TOTAL 336 100

Two cases did not provide valid responses

Table 5.12 shows the distribution of market uncertainty in the research sample. The
pattern in this table suggests that most of the products are targeted to markets that are
highly uncertain and unstable. A relatively small percentage (24%) of the sample is

targeted to stable markets.

Table 5.12. Frequency Distribution of Level of Market Uncertainty

     

Level of Market

 

 

 

Frequency Percentage
Uncertainty
Low 79 24
Moderate 181 54
High 74 22
TOTAL “.334”. -- 100
= uuuuu I

 

Four cases did not provide valid responses

61

Finally, Table 5.13. lists the stage in the product life cycle for the new products
developed and introduced by the sample respondent firms. The early stages of the
product life cycle account for 63% of the sample. This is in part due to the fact that the
sample is representative (to a large extent) of high tech industries in which the pressures

for innovation are high.

Table 5.13. Frequency Distribution of Stage in the Product Life Cycle for Targeted Product

 

 

 

Stage in Product Life Cycle Frequency Percentage

For Targeted Product
Introduction 40 12
Early Growth 68 20
Growth 116 35
Early Maturity 51 15
Maturity 56 17
Decline 3 1
TOTAL ""33? ”ii?“

I

 

Four cases did not provide valid responses
In sum, the distributions for the four contingency variables suggest that the research
sample is comprised of projects in which :

I The technical complexity of the products is high

I The level of market uncertainty is high

I The level of product innovativeness is high

I Most new products are in the early or growth stages of the product life cycle

62

5.3 Data Quality
5.3.1 Reliability of Scales

Reliability measures the extent to which a measurement scale yields consistent
results on repeated replications of the same scale. Assessing the internal consistency of
items within a scale is the most important method of establishing reliability when only
one form of a measure is available (Flynn, Sakakibara, Schroeder, Bates and Flynn,
1990). The recommended measure for testing the internal consistency of a set of items is
provided by Cronbach’s coefficient alpha (Churchill, 1979; Cronbach, 1951). An
exploratory factor analysis (using principal components and varimax rotation technique)
was conducted on all the constructs germane to the conceptual model in Figure 1. For
each construct, the factor analysis technique was allowed to specify the number of factors
that account for the highest variance explained in the construct. The only specification
made was that the eigenvalues for each factor should exceed one. In this section, the
results of exploratory factor analysis and reliability analysis are reported. In a later
section, confirmatory factor analysis is used primarily to verify discriminant validity of
the factors and also as a post-hoe assessment of the overall fit of the model shown in
Figure 1. In all there were seven constructs in the conceptual model (See Figure 1). The
survey instrument had a total of 63 items representing the seven constructs. The results of
exploratory factor analysis are discussed next.
5.3.2. The Supplier Integration Construct

The survey instrument had twenty items relating to supplier integration that were

culled out of the literature. From this initial list, eleven items constituting three stable

63

factors emerged (see Table 5. 14). The three stable factors were labeled : Communication
and Information Sharing, Design Involvement and Infrastructure. The Communication
and Information Sharing factor represented practices that enhanced the frequency of
communication and information sharing during the different stages of the development
project between the suppliers and the project team. The Design Involvement factor
depicted the involvement of suppliers in key design activities such as defining the product
architecture and setting design specifications. The Infrastructure factor represented
programs and practices with common joint goals and objectives for key suppliers and the
firm such as joint training and education programs, access to a common information
systems and co-location of project personnel and key suppliers.

Table 5.14. Results of Exploratory Factor Analysis : The Supplier Integration

 

 

 

 

Construct
Factor 1 Factor 2 Factor 3
Items Communication & Design Infrastructure
Information Sharing Involvement

Direct communication with key suppliers 0.81 0.20 0.20
Frequency of communication with key suppliers in prototype stage 0.80 0.27 0.12
Participation of key suppliers in NPD team 0.71 0.22 0.23
Frequency of communication with key suppliers in production stage 0.70 0.15 0.11
Sharing design knowledge with key suppliers 0.59 0.36 0.38
Involvement of key suppliers in defining architecture of new products 0.19 0.85 0.16
Involvement of key suppliers in setting design specifications 0.29 0.80 0.08
Involvement of key suppliers in product design 0.30 0.72 0.27
Common linked information systems 0.20 0.01 0.74
Colocation of project personnel and key suppliers 0.12 0.17 0.73
Shared education and training programs with key suppliers 0.19 0.33 0.65
EEenvalues 5.01 1.08 1.06
Proportion of Variance Explained 45.50 9.81 9.62
Cumulative Percentage of Variance Explained 45.50 55.31 64.93
Cronbach's alpha for scale 0.85 0.82 0.63

As can be seen fiom Table 5.14, the scales comprising each of the three factors
were internally consistent and the constructs were sufficiently reliable, with Cronbach’s

alphas ranging from 0.63 to 0.85. Reliability for all three scales exceeded .60, the

64

threshold Nunnally (1978) recommends for exploratory research. There were no
significant cross loadings of an item across multiple factors.
5.3.3. The Design-Manufacturing Integration Construct

The survey instrument had seven items relating to design-manufacturing
integration that were taken from the literature. From this initial list, only one item which
had a low loading was dropped and six items constituting two stable factors were retained
(see Table 5. 15). The two stable factors were labeled : Integrative Use of Technology
and Design Proactiveness. The Integrative Use of Technology factor measured the extent
to which computers were used to link decisions and practices relating to design,
production and resource planning. The Design Proactiveness factor represented programs
that emphasized early detection of problems such as design for manufacturability (DF M)

and translation of customer’s preferences into product and process parameters.

Table 5.15. Results of Exploratory Factor Analysis : The Design-Manufacturing Integration

 

 

 

Construct
Factor 1 Factor 2
Items Integrative Use Design
of Technology Proactiveness

Use of Computers to link design and production 0.85 0.06
Use of Computers to link design and production planning 0.83 0.21
Use of Computers to link design and resource planning 0.81 0.13
Translating customer preferences into product parameters 0.00 0.88
Translating customer preferences into process parameters 0.12 0.80
Design for manufacturability 0.28 0.48
figenvalues 2.53 1 .34
Proportion of Variance Explained 42.13 22.34
Cumulative Percentage of Variance Explained 42.13 62.47
Cronbach's alpha for scale 0.80 0.60

As can be seen from Table 5.15, the scales comprising both factors were

internally consistent and the constructs were sufficiently reliable, with Cronbach’s alphas
65

of 0.80 and 0.60. There were no significant cross loadings of an item across multiple
factors.
5.3.4. The Customer Integration Construct
The survey instrument had nine items relating to customer integration that were

selected fi'om the literature. From this initial list, only one item which had a low loading
was dropped and six items constituting two stable factors were retained (see Table 5.16).
The two stable factors were labeled : Capturing the Voice of the Customer and Customer
Involvement in NPD decisions. The Capturing the Voice of the Customer factor
appeared to tap into the group of practices and procedures to gather information on
customer needs (i.e., the "voiCe" of the customer) and transmit these needs in a systematic
manner into product specifications or parameters (i.e., the translation process). The
second factor of Customer involvement in new product development decisions
represented practices that emphasized early involvement of customers in manufacturing
and other key decisions in the development process.

The scales comprising both factors were internally consistent and the constructs
were sufficiently reliable, with Cronbach’s alphas of 0.86 and 0.63 (see Table 5.16).

There were no significant cross loadings of an item across multiple factors.

66

Table 5.16. Results of Exploratory Factor Analysis : The Customer Integration

 

 

 

Construct
Factor 1 Factor 2
Items Capturing Customer
"Voice of the Customer" Involvement in NPD

Reviewing NPD efforts to ensure match with customer needs 0.83 0.1 1
Studying how customers use products 0.78 0.05
Sharing Data on customer needs across functions 0.78 0.22
Listening to customers needs while developing the product concept 0.77 0.36
Meeting with customers 0.72 0.33
Including customers in decisions concerning manufacturing -0.04 0.81
Including customers in decisions concerning NPD 0.39 0.71
Communicating key aspects of new product to customers 0.30 0.63
Eigenvalues 3 .97 1 .13
Proportion of Variance Explained 49.59 14.09
Cumulative Percentage of Variance Explained 49.59 63.68
Cronbach's alpha for scale 0.86 0.63

5.3.5. The Concurrent Process Management Construct

In the survey instrument there were nine items relating to concurrent process
management that were selected fi'om the literature. From this initial list, two items which
had low factor loadings were dropped and the remaining seven items constituting two
stable factors were retained (see Table 5.17). The two stable factors were labeled :
Dynamic Iterative Routines and Downstream Coordination. The Dynamic Iterative
Routines factor represented joint problem solving activities and information sharing on a
dynamic basis (i.e., as the different stages of the NPD process evolved). The
Downstream Coordination factor captured the coordination of product design activities to
downstream steps such as commitments to tooling, production feasibility and

manufacturing process costing.

67

Table 5.17. Results of Exploratory Factor Analysis : The Concurrent Process
Management Construct

 

 

 

Factor 1 Factor 2

Items Dynamic Downstream

Iterative Routines Coordination
Joint problem solving among design, mfg and engg in product design phase 0.86 -0.09
Joint problem solving among design, mfg and engg in process design phase 0.84 0.15
Information sharing with engg/R&D as process design evolved 0.80 0.14
Information sharing with mfg as product design evolved 0.78 -0.07
Completeness of product design when mfg made formal cost estimates -0.04 0.86
Completeness of product design when mfg provided feedback on production feasibility -0.04 0.81
Completeness of product design when mfg made purchasing commitments 0.17 0.75
Eigenvalues 2.78 1 .97
Proportion of Variance Explained 39.64 28.09
Cumulative Percentage of Variance Explained 39.64 67.73
Cronbach's alpha for scale 0.84 0.74

The scales comprising both factors were internally consistent and the constructs were
sufficiently reliable, with Cronbach’s alphas of 0.84 and 0.74 (see Table 5.17). There

were no significant cross loadings of an item across multiple factors.

5.3.6. The Product Fit with Firm Capability Construct

In the survey instrument there were nine items relating to Product Fit with Firm
Capability. All nine items were retained constituting two stable factors which were
labeled: Product Fit to Manufacturing Capability and Product Fit to Design Capability
(see Table 5.18). The Product Fit to Manufacturing Capability factor represented the fit
of the planned new product to firm capabilities in manufacturing such as capacity,
flexibility and prior experience. The Product Fit to Design Capability factor captured the
Fit of the planned product to design capabilities such as electronic data interchange,

computer aided engineering and design familiarity.

68

Table 5.18. Results of Exploratory Factor Analysis : The Product Fit with Firm

 

 

 

Capability Construct
Factor 1 Factor 2
Items Product Fit to Product Fit to
Mfg. Capability Design Capability
Product fit to the ability to handle changes in production output 0.92 0.16
Product fit to the ability to handle changes in product lines 0.90 0.17
Product fit to the ability to handle major design changes 0.90 0.1 1
Product fit to available capacity 0.83 0.27
Product fit to knowledge of process technology 0.71 0.43
Product fit to manufacturing experience 0.68 0.42
Product fit to the available expertise in computer aided engineering 0.22 0.78
Product fit to the available expertise in electronic data interchange 0.07 0.77
Product fit to design experience 0.32 0.74
Eigenvalues 5.21 1 .29
Proportion of Variance Explained 57.90 14.37
Cumulative Percentage of Variance Explained 57.90 72.27
Cronbach's alpha for scale 0.93 0.70

The scales comprising both factors were internally consistent and the constructs were
sufficiently reliable, with Cronbach’s alphas of 0.93 and 0.70 (see Table 5.18). There
were no significant cross loadings of an item across multiple factors.

5.3.7. The Product Development Process Performance Construct

In the survey instrument there were four items relating to the Product Development
Process Performance Construct. The exploratory factor analysis on these four items
yielded one stable factor suggesting that this construct was unidimensional (see Table
5.19). This factor represented the outcome measures of the development process i.e.,
quality (conformance and design), time-to-market and cost. The scale comprising this

factor was sufficiently reliable, with a Cronbach alpha of 0.78 (see Table 5.19).

69

Table 5.19. Results of Exploratory Factor Analysis : The Product DeveIOpment
Process Performance Construct

 

Items Factor 1
Unidimensional
Conformance Quality n.a.
Design Quality n.a.
Cost n.a.
Time-to-Market n.a.
Cronbach's alpha for scale 0.78

 

Note: n.a. = not applicable

5.3.8. The Product Development Project Success Construct

In the survey instrument there were three items relating to the Product Development
Project Success Construct. The exploratory factor analysis on these three items yielded
one stable factor suggesting that this construct was unidimensional (see Table 5.20). This
factor represented the outcome measures of the new product development project i.e.,
initial market share, profitability and break-even time. The scale comprising this factor
was sufficiently reliable, with a Cronbach alpha of 0.84 (see Table 5.20).

Table 5.20. Results of Exploratory Factor Analysis : The Product Development
Project Success Construct

 

Items Factor 1
Unidlmensional
Initial Market Share n.a.
Profitability n.a.
Break even time n.a.
Cronbach's alpha for scale 0.84

 

Note: n.a. = not applicable

70

In all the survey instrument contained 63 items that represented the seven
constructs. The results from the exploratory factor analysis reduced the original 63 items
to a stable and reliable list of 48 items representing the originally proposed seven factors.
It must be noted that some of the originally pr0posed factors turned out to be
multidimensional in nature and the refined list of 48 items represented 13 factors and not
7 factors as originally proposed. Taking into account the decrease in number of items
(fi'om 63 to 48) and the increase in number of factors (from 7 to 13), the refined list

constituted a significant reduction in dimensionality.

5.3.9. Validity of Constructs

The requirements for validity are two fold. The scales must truly measure what it
is supposed to measure and it must not measure anything else (Churchill, 1979). To
satisfy these two requirements several facets of validity were tested. Content validity
measures the extent to which the scale truly measures the concept that it intended to
measure, based on the domain of meaning of items comprising the scale (Churchill,
1979). It is a non-statistical assessment of validity which is ensured by expert judgment
or through an extended literature search. A high degree of content validity of the scales
was ensured by referring to the extant empirical or prescriptive literature relating to each
of the scale items followed by confirmation with experienced new product development

managers via field interviews.

71

Convergent Validity

In order to ensure that a set of items reflect a particular construct and is not an
artifact of the measurement procedure, it is desirable to include multiple items of the
same construct. Convergent validity refers to the extent to which varying approaches to
construct measurement yield the same results (Campbell and Fiske, 1959; Churchill,
1979). One method of determining convergent validity is to treat each item in a scale as a
different approach to measuring the construct (Ahire, Golhar and Waller, 1996).
Convergent validity is assessed by verifying whether each indicator’s estimated pattern
coefficient on its posited underlying construct factor is significant i.e., greater than twice
its standard error (Anderson and Gerbing, 1988). Following this recommendation, a
confirmatory factor analysis (CFA) was conducted on the 48 items (comprising 13
factors) as identified in exploratory factor analysis. The confirmatory factor structure was
tested using EQS Windows (version 5.6) structural equation modeling software package.
Maximum likelihood method of estimation was used. The results are reported in Table
5 .21. As can be seen from this table, all factor loadings were statistically significant
indicating a high degree of convergent validity.
Discriminant Validity

Discriminant Validity is the extent to which the constituent items of a scale

measure only one distinct construct (Churchill, 1979; Bagozzi, Yi, and Phillips, 1991;
Ahire et al., 1996). The Lagrange Multiplier (LM) test was used to verify if there was
any significant cross loadings of any item on more than one factor. The results as a

whole indicated a reasonably high degree of divergence across factors as indicated only a

72

few items that had cross loadings on to a different factor than hypothesized. The fact that
all of these factor loadings were not significant as indicated by the multivariate LM test
output and that the overall model was significant (as is discussed in the next section)
provided reasonable assurance of discriminant validity of the constructs. It may be noted
that the factor structure that is being tested in this CF A is post-facto and consequently the
fact that a few items had cross loadings in a 48-item 13-factor structure is not unusual.
Nevertheless, an assessment of the appropriateness of the proposed factor structure
should also include an evaluation of overall fit of the model which is reported next.

The overall validity of the confirmatory factor model was tested using multiple fit
criteria. The results of these criteria are reported in Table 5.21. The Chi-squared value
for the CFA model is 703.66 for a degree of fieedom of 665. This chi-squared value was
statistically significant with a p-value of 0.14. Another way of assessing the fit of the
overall model is by computing the ratio of Chi-squared to the degree of freedom.
According to Matsueda (1982), a ratio of x 2 to (If of no more than four-to-one is
considered a good fit. Our value of 1.06 is indicative of a good fit of the model. The
EQS output provided two goodness of fit indices - Comparitive Fit Index (CPI) and
Bender-Bonnet Non-Normed Fit Index (BBNFI). Both the fit indices were high (.99 and
.99 respectively) indicating a good fit. When these results are considered together, they

lend support to the overall validity of the pr0posed factor structure model.

73

Table 5.21. Results of Confirmatory Factor Analysis

 

 

===== E
Construct/ Standardized t-values
Items Loadings
Supplier Integration - Communication
And Information Sharing (F1)
Participation of key suppliers in NPD team .71 12.1 1
Direct communication with key suppliers .68 l 1.56
Sharing design knowledge with key suppliers .80 14.17
Supplier Integration - Design Involvement (F2)
Involvement of key suppliers in defining architecture of new products .79 14.76
Involvement of key suppliers in product design .81 15.13
Involvement of key suppliers in setting design specifications .74 13.77
Supplier Integration - Infrastructure (F3)
Shared education and training programs with key suppliers .72 1 1.33
Co-location of project personnel and key suppliers .60 9.49
Design-Manufacturing Integration - Integrative Use of Technology (F4)
Use of computers to link design and production planning .79 14.12
Use of computers to link design and production .74 13.07
Use of computers to link design and resource planning .71 12.25
Design-Manufacturing - Design Proactiveness (F5)
Translating customer preferences into product parameters .81 13.22
Translating customer preferences into process parameters .62 10.28
Customer Integration - Capturing Voice of
The Customer (F6)
Listening to customers needs while developing the product concept .86 17.08
Studying how customers use products .70 12.84
Meeting with customers .73 13.44
Sharing data on customer needs across functions .71 13.00
Reviewing NPD efforts to ensure match with customer needs .71 13.32
Customer Integration - Customer Involvement in NPD decisions (F7)
Including customers in decisions concerning NPD .78 12.69
Communicating key aspects of new product to customers .58 9.68

 

74

Table 5.21. Results of Confirmatory Factor Analysis (Cont’d)

 

 

 

Construct/ Standardized t-values
Items Loadings *

Concurrent Process Management — Dynamic Iterative Routines (F8)
Information sharing with mfg. as product design evolved .67 12.36
Information sharing with engg/R&D as process design evolved .80 14.38
Joint problem solving among design, mfg. and engg. in product design phase .88 16.94
Joint problem solving among design, mfg. and engg. in process design phase .81 15.56
Concurrent Process Management — Downstream Coordination (F9)
Completeness of product design when mfg. gave feedback on prodn. feasibility .71 12.24
Completeness of product design when mfg. made formal cost estimates .87 15.00
Completeness of product design when mfg. made purchasing commitments .62 10.26
Product Fit with Firm Capability - Manufacturing (F10)
Product fit to knowledge of process technology .63 1 1.34
Product fit to available capacity .70 12.83
Product fit to the ability to handle changes in production output .85 17.49
Product fit to the ability to handle changes in product lines .93 19.80
Product fit to the ability to handle major design changes .84 16.81
Product Fit with Firm Capability - Design (F11)
Product fit to available expertise in computer aided engineering .84 12.27
Product fit to available expertise in electronic data interchange .58 9.04
Product Development Process Performance (F12)
Conformance Quality .61 9.89
Design quality .65 10.92
Cost .68 1 1.40
Time-to-market .72 12. 19
Product Development Project Success (F13)
Market Share .79 14.12
Profitability .90 16.28
Break Even Time .85 15.92

x’Value 703 .66

df 665

p-value 0.14

x2 to df ratio 1.06

Comparitive Fit Index .99

Bentler-Bonnett Non-Normed Fit Index .99
_ =—

 

 

° all factor loadings are significant at p<.01 level

75

Predictive Validity

Predictive or Criterion-related Validity investigates the empirical relationship
between the scores on a summated scale (predictor) and an objective outcome (the
criterion). The most commonly used measure of predictive validity is a validity
coefficient which is the correlation between predictor and criterion scores (Flynn et al.,
1990). A validity coefficient is an index of how well criterion scores can be predicted
from the scores on the summated scale. In our context, in order to demonstrate predictive
validity, all of the exogenous factors constituting the three integration mechanisms and
concurrent process management must be correlated with a performance measure such as
project success. The results of correlations of the exogenous factors with project success
are shown in Table 5.22. With the exception of one factor, all factors were significantly
correlated with project success. This validates the inclusion of these exogenous factors in

the conceptual scheme that seeks to predict project success.

Table 5.22. Results of Predictive Validity of Constructs : Correlations of Exogenous Factors
with Project Success
Pearson Product Moment Coefficients (p-value)

 

 

 

 

__
Criterion Variable

Predictor Variables New Product Development Project Success
Supplier Integration -

Communication and Information Sharing .11 (.04)

Design Involvement .09 (.09)

Infrastructure .04 (.45)
Design-Manufacturing Integration -

Integrative Use of Technology .12 (.03)

Design Proactiveness .22 (.00)
Customer Integration —

Capturing “Voice of the Customer .26 (.00)

Customer Involvement in NPD decisions .09 (.08)
Concurrent Process Management -

Dynamic Iterative Routines .18 (.00)

Downstream Coordination .12 (.03)

 

The previous sections demonstrated the desirable measurement properties of the
constructs and items in the conceptual model of Figure 1. In the next section, the

findings relating to each of the research questions are reported.

5.4 Findings

5.4.1 Antecedents to Project Success

H1: Product Concept Effectiveness and New Product Development Project Succeg:
Hypothesis 1 proposed a positive relationship between product concept effectiveness and
project success. To test this relationship, a regression of the two product fit variables on
project success was conducted. The results of the regression analyses of Product fit to
manufacturing capability and Product Fit to design capability as independent variables

and Project Success as the dependent variable are reported in Table 5.23.

Table 5.23. Results of Stepwise Regression Analysis of the Effect of Product Concept Effectiveness
on New Product Development Project Success
Standardized Coefficients (t-value)

 

 

 

 

 

 

 

:— __
Criterion Variable
Predictor Variables New Product Development
Project Success
Product Fit with Manufacturing
Capabilities .19
(3.18)‘
Product Fit with Design
Capabilities .13
(2.09)b
F-Value 14. 15‘
R2 .08
Adjusted R2 .07
‘p<.01; bp<.05 1

77

In table 5.23, the F-value, model R2 and adjusted R2, standardized regression
coefficients of the independent variables and associated t-values are listed. Both
indicators of Fit with firm capability were significantly related (p<.05) to Project Success.
The explanatory power of the model as indicated by model adjusted R2 was 7%. H1 was
supported.

H2 toH4: Development Process Performance and Project Success:

Hypotheses 2 to 5 proposed positive relationships between individual components
of development process performance - conformance quality (HZ), design quality (H3),
time-to-market (H4) and cost (H5) - and project success. To test these relationships,
regressions of individual components of development process performance on project
success were conducted separately. The results of the regression analyses are shown in
Tables 5.24 to 5.27.

As can be seen from tables 5.24 to 5.27, all the four individual components of
development process performance - conformance quality, design quality, time-to-market
and cost - were significantly related to project success. The explanatory power of the

models were between 14 to 20%. H2 to H5 were all supported.

Table 5.24. Results of Regression Analysis of the Effect of Conformance Quality
on New Product Development Project Success
Standardized Coefficients (t-value)

Criterion Variable

 

 

 

 

Predictor Variables New Product Development Project Success
Conformance Quality .40
(8.06)‘
F-Value 64.93'
R2 lg : I I====
‘p<.01

78

Table 5.25. Results of Regression Analysis of the Effect of Design Quality
on New Product Development Project Success
Standardized Coefficients (t-value)

17
-

Criterion Variable

 

Predictor Variables New Product Development
Project Success

 

 

Design Quality .40
(8.02)‘
F -Value 64.34‘

R2 16

 

Table 5.26. Results of Regression Analysis of the Effect of Time-to-Market
on New Product Development Project Success
Standardized Coefficients (t—value)

 

 

 

 

 

 

 

-— -
Criterion Variable
Predictor Variables New Product Development Project Success
Tirne-to-Market .45
(9.21)‘
F-Value 84.78'
R2 .20
E E
‘p<.01

Table 5.27. Results of Regression Analysis of the Effect of Cost Performance
on New Product Development Project Success
Standardized Coefficients (t-value)

Criterion Variable

 

 

 

 

 

Predictor Variables New Product Development Project Success
Cost Performance .38
(7.50)‘
F-Value 56.31‘
R2 .14
= =
‘p<.01

79

Post-Hoe Analyses: Development Process Performance and Project Success:

In order to further analyze the link between development process performance and
project success, several post-hoe analyses were carried out. A stepwise regression
analyses of the four components of development process performance was conducted to
see the relative importance of each of these components on project success. The results
shown in Table 5.28 indicate: (1) all four components are significant predictors (p<.10) of
project success; and (2) time-to—market appears to be the most significant predictor of all
the four components.

In Table 5.29, the results of regressing a composite index of development process
performance (simple average of all four components) on project success are shown. The
index is a significant predictor of project success and it accounts for 27% of the variation

in project success.

Table 5.28. Results of Stepwise Regression Analysis of the Effect of Components of Development
Process Performance on New Product Development Project Success
Standardized Coefficients (t-value)

Criterion Variable

 

 

 

Predictor Variables New-Product DevelopmentProject Success
Conformance Quality .15
(2.38)b
Design Quality .11
( I .70)°
Time-to-Market .26
(4.49)‘
Cost Performance .15
(2.61)"
F-Value 3 1.99'
R2 .28
Adjusted R2 .27

‘p<.01; 5p<.05; ‘p<. 10

80

Table 5.29. Results of Regression Analysis of the Effect of the Index of Development Process
Performance on New Product Development Project Success
Standardized Coefficients (t-value)

 

Criterion Variable

 

Predictor Variables New Product Development
Project Success

 

Index of Development Process

 

Performance .53
(1 1 .39)‘
F-Value 129.81‘
R2 .28

 

'p<.01

The previous analyses examined the effects of the four components of development
process performance and its composite index on the overall construct of project success.
In the next analysis, the impact of the four components of development process
performance on the individual indicators of project success (i.e., market share,
profitability and break even time) is examined. The results are shown in Table 5.30.

As far as market share is concerned, the most important predictor was time-to
market followed by design quality. For profitability, cost performance was most
important followed by conformance quality and time-to-market. Finally, for break even
time, time-to-market was most important followed by conformance quality. All models
were significant as indicated by the high F-value. Explanatory power was around 20%.
It is interesting to note that different development process measures appear to influence

different project success measures.

81

Table 5.30. Results of Stepwise Regression Analysis of the Effect of Components of Development

Process Performance on Components of Project Success
Standardized Coefficients (t-value)

 

Criterion Variables

 

 

 

Predictor Variables Market Profitability Break Even
Share Time
Conformance Quality .12 .17 .18
(l .68)° (3.15)‘ (3.20)‘
Design Quality .19 .07 .04
(2.78)” (1.02) (.53)
Time-to-Market .24 . 1 6 .28
(4.27)‘ (2.75)b (4.77)‘I
Cost Performance -.02 .30 .13
(-.3 1) (5.32)‘ (2.22)”
F—Value 28.54' 36.59‘ 3 1.05'
R2 .21 .25 .22
Adjusted R2 .20 .24 .21

 

‘p<.0 1; bp<.05; cp<.10

In order to investigate the possibility of multicollinearity among the four

components of development process performance, an analyses of the correlation matrix

(see Table 5.31) was done. As expected, the four measures were inter-correlated. It is

interesting to note that correlations within similar items in content (conformance and

design quality) were higher than the correlations that were dissimilar in content.

82

Table 5.31. Descriptive Statistics of the Components of Development
Process Performance

 

 

Product Moment Correlations

 

 

Variable Mean SD. 1 2 3
1. Conformance
Quality 6.37 2.13 1.00
2. Design Quality 6.86 1.92 .68‘ 1.00
3. Cost Performance 5.71 2.37 .38' .45' 1.00

4. Time-to-Market 5.1 l 2.69 .44' .46' .49‘

     

-p<.01

While the zero order correlations in Table 5.31. suggest that all components are
highly inter-correlated, a clearer picture emerges in examining the partial correlations
among the four components of development process performance (see Table 5.32). As
can be seen from this table, all partial correlations are lower in magnitude as compared to
its zero order counter parts. It is also interesting to note that some correlations that were
significant in zero order analyses are no longer significant in partial correlational analyses
suggesting an investigation as to whether multicollinearity really exists and if so to what
extent are the previous results invalid. For this purpose, a partial correlation analyses of
the impact of each component (controlled for all other components) on project success,

was conducted (see Table 5.33).

83

Table 5.32. Partial Correlations of the Components of Development
Process Performance

Partial Correlations

 

 

Variable 1 2 3
1. Conformance
Quality 1.00
2. Design Quality .58' 1.00
3. Cost Performance .05 .20' 1.00
4. Tirne-to-Market . 1 5" .17‘ .34‘
-==

 

‘p<.0 1; hp<.05; cp<.10

Recall that in a previous regression analyses of all components on market share
(Table 5.30), the most important predictor was time-to market followed by design quality.
The same pattern of time-to market followed by design quality is obtained in the partial
correlation analysis of the four components on market share (see Table 5.33). The
regression analyses results for profitability indicated that cost, conformance quality and
time-to-market was the order of importance in terms of strength of effect. In the partial
correlation results, cost was followed by time-to-market and then conformance quality.
As the difference between the last two in magnitude is not so high, it may be assumed
that time and conformance quality are equal in importance after cost performance in

terms of their impact on profitability of a new product development project.

84

Table 5.33. Partial Correlations between Components of Development
Process Performance and Market Share

 

Partial Correlations between: Controlling For:

 

Variables

 

Conformance Quality

 

 

 
 

 

 

 

 

 

 

 

 

and Market Share .11b Design Quality, Cost and Time-to market
Design Quality and

Market Share .14' Conformance Quality, Cost and Time-to market
Cost Performance and

Market Share -.02 Conformance Quality, Design Quality and

Time-to market

Time-to-Market and

Market Share .22' Conformance Quality, Design Quality and Cost
‘p<.01; IDp<.05;

Table 5.34. Partial Correlations between Components of Development
Process Performance and Profitability
Partial Correlations between: Controlling For:
Variables

Conformance Quality

and Profitability .l 1" Design Quality, Cost and Time-to market
Design Quality and

Profitability .06 Conformance Quality, Cost and Time-to market
Cost Performance and

Profitability .27‘I Conformance Quality, Design Quality and Time-to market
Time-to-Market and

Profitability .14' Conformance Quality, Design Quality and Cost

-

'p<.01; hp<.05

For break-even time, regression analyses results suggested that time-to-market
followed by conformance quality was the order of importance. The partial correlation

analysis results confirms the same pattern (see Table 5.35). In sum, the pattern of results

85

in regression and partial correlation analyses are robust enough to discount the threat of

multicollinearity among the components of development process performance.

Table 5.35. Partial Correlations between Components of Development
Process Performance and Break Even Time

 

 

 

Partial Correlations between: Controlling For:
Variables

Conformance Quality

and Break Even Time .14" Design Quality, Cost and Time-to market
Design Quality and

Break Even Time .02 Conformance Quality, Cost andTime-to market
Cost Performance and

Break Even Time .1 1" Conformance Quality, Design Quality and Time-to market
Time-to-Market and

Break Even Time .24' Conformance Quality, Design Quality and Cost

 
 

 
    

 

'p<.01 ; "p<.05;

Post Hoc Analyses: Relative Impact of Product Concept Effectiveness and Development
Process Performance on Project Success

It may be recalled that the testing of H1 and H2 to H5, found support for the
separate impact of product concept effectiveness and development process performance
on project success. A logical extension would be to investigate the relative impact of
these two constructs on project success. Accordingly, a stepwise regression analyses of
the three constructs - product fit with manufacturing capabilities, product fit with design
capabilities and development process performance — on project success was conducted.
The results indicate that development process performance and product fit with
manufacturing capabilities (and in this order) are significant predictors of project success.

The three predictors accounted for 29% of variation in project success. In the next series

86

of post-hoe analyses, interaction terms are also included as predictor variables so as to
investigate the possibility that the fit constructs and development process performance

might complement one another in their joint impact on project success.

Table 5.36. Results of Stepwise Regression Analysis of the Effect of Product Concept Effectiveness
and Development Process Performance on New Product Development Project Success
Standardized Coefficients (t-value)

Criterion Variable

 

 

A
A.

 

Predictor Variables New Product Development
Project Success
Product Fit with Manufacturing
Capabilities . 14
(3.07)‘
Product Fit with Design
Capabilities .08
(1.58)
Index of Development Process
Performance .50
(10.54)‘
F-Value 71 .27'
R2 .30
Adjusted R2 .29

 

‘p<.01

In Table 5.37, the results of hierarchical regression analyses of product concept
effectiveness, development process performance and their interaction terms on the three
components of project success are reported. In hierarchical regression, independent
variables are entered in a particular order in order to minimize the effect of
multicollinearity among the independent variables (Cronbach, 1987, Jaccard, Turrisi and
Wan, 1990). First, main effects are entered into the model, following which interaction

terms that explain additional variance over and above the main effects enter the model.
87

An interaction term that enters as a significant independent variable ‘over rides’ or

dominates a significant main effect.

Table 5.37. Results of Hierarchical Regression Analysis of the Effect of Product Concept
Effectiveness, Development Process Performance and their Interactions on Components of
Project Success

Standardized Coefficients (t-value)

 

Criterion Variables

 

 

 

Predictor Variables Market Profitability Break Even
Share Time
Main Effects:
Product Fit with Manufacturing
Capabilities (PF M) -.04 .1 1 .07
(-.25) (.63) (1.29)
Product Fit with Design
Capabilities (PFD) .08 .06 -.04
(1.43) (1.17) (-.29)
Development Process .17 .35 .34
Performance (DPP) (2.01 )" (4. 12)‘ (4.82)‘
Interaction Effects:
PFMXDPP .32 .17 .13
(3.77)‘ (1.96)" (1.41)
PFD X DPP .12 .08 .17
(1 .46) (.96) (2.44)"
F-Value 47.94' 53.28' 49.91'
R2 .22 .24 .23
Adjusted R2 .22 .24 .22

 

‘p<.0 1; "p<.05; ‘p<. 10

In all three regression analyses at least one interaction term enters as a significant
predictor. The regression models are all significant with a reasonably high degree of
explanatory power (around 22%). For both the market share and profitability regressions,
the interaction between product fit with manufacturing capability and development

88

process performance enters as a significant predictor. However, for break even time it is
the interaction between product fit with design capability and development process
performance that enters as a significant predictor. This result suggests that product
concept effectiveness and development process performance have a complementary role
in their impact on project success. Also, the performance consequences of this
complementary interaction are different depending on the type of project success measure
being considered. For near term performance effects such as initial market share and
profitability, an emphasis on manufacturing capabilities complements development
process performance, while design capability does not. For quick capital recovery, the
match of appropriate design skills and good execution appears to have a significant

impact.

5.4.2 Antecedents to Product Concept Effectiveness

H6: Supplier Integration and Product Concept Effectiveness: Hypothesis 6 proposed a
positive relationship between supplier integration and fit with firm capabilities. To test
this relationship, a regression of the three sub-dimensions of supplier integration on the
two product fit variables were conducted. The results of the regression analyses of the
three dimensions of supplier integration on Product fit to manufactruing capability is
shown in the second column of Table 5.38 and the results for Product Fit to design

capability is shown in the third column of this table.

89

Table 5.38. Results of Stepwise Regression Analysis of the Effect of Supplier Integration on Product
Concept Effectiveness
Standardized Coefficients (t-value)

 

Criterion Variable

 

Predictor Variables Product Fit with Product Fit with
Manufacturing Design
Capabilities Capabilities

 

Supplier Integration -
Communication and

 

Information Sharing .08 .02
(1.26) (.22)
Supplier Integration -
Design Involvement .08 .10
(1.37) (1.73)c
Supplier Integration -
Infrastructure . 16 .25
(3.01)" (4.30)‘
F-Value 9.04‘ 18.13'
R2 .03 .10
Adjusted R2 .02 .09

 

'p<.01; "p<.05; ‘p<. 10

The Infrastructure component of supplier integration appeared as a significant
predictor of product fit to manufacturing capability as well as product fit to design
capability. This suggests that practices such as co-location of project personnel and
suppliers, and joint training programs facilitate smoother integration of key suppliers in
the product planning effort. These practices improves the planning for the new product
by considering suppliers design and manufacturing skills while planning for a new

product. The results also show that actual design participation in the NPD project (as

90

indicated by the factor — Design Involvement) is a significant predictor of Product fit with
design capability which makes logical sense. H6 was supported.

Although, Supplier Integration - Communication and Information Sharing
dimension was expected to influence both dimensions of product fit to firm capabilities, it
did not affect either dimension. This is somewhat surprising considering that product
planning effectiveness could conceivably be enhanced by improved commrmication
between suppliers and project personnel. Post-hoe analysis of the correlations of the two
product fit factors to individual practices constituting the communication and information
sharing dimension of supplier integration point to some possibilities for this result.
Indeed, the practice of sharing design knowledge with suppliers was significantly
correlated to both dimensions of product fit to firm capability (p<.01), suggesting that
this form of communication enhances planning effectiveness. Similarly, the practices of
participation of key suppliers in NPD team and direct communication with key suppliers
was significantly correlated to product fit to design capability (p<.01). Thus, the
relationship that was weak at the construct level, is actually significant at the practice
level.

H7: Desigp-Manufacturing Integration and Product Concept Effectiveness: Hypothesis 7
predicted a positive relationship between design-manufacturing integration and fit with
firm capabilities. To test this relationship, a regression of the two sub-dimensions of
design-manufacturing integration on the two product fit variables were conducted. The

results of the regression analyses for Product fit to manufacturing capability is shown in

91

the second column of Table 5.39 and for Product Fit to design capability in the third

column of this table.

Table 5.39. Results of Stepwise Regression Analysis of the Effect of Design- Manufacturing
Integration on Product Concept Effectiveness
Standardized Coefficients (t-value)

 

Criterion Variable

 

 

 

Predictor Variables Product Fit with Product Fit with
Manufacturing Design
Capabilities Capabilities
Design-Manufacturing Integration -
Integrative Use of
Technology . 15 .40
(2.72)‘ (7.98)ll
Design-Manufacturing Integration —
Design Proactiveness .25 .17
(4.50)‘ (3 .44)‘I
F-Value 19.94' 52.22"
R2 .1 1 .24
Adjusted R2 .10 .23

 

-p<.o1

Both components of design-manufacturing integration — Integrative Use of
Technology and Design Proactiveness were significant predictors of product fit to
manufacturing capability as well as product fit to design capability. This suggests that
considering the long term impact of design decisions enhances product planning. Also
the use of computers in linking design and manufacturing facilitates product planning.

H7 was strongly supported.

H8: Customer Integrption and Product Concept Effectiveness: Hypothesis 8 predicted a

positive relationship between customer integration and fit with firm capabilities. To test

this relationship, a regression of the two sub-dimensions of customer integration on the

92

 

two product fit variables were conducted. The results of the regression analyses for
Product fit to manufacturing capability is shown in the second column of Table 5.40 and
for Product Fit to design capability in the third column of this table.

The results in Table 5.40 show that the factor of customer involvement in key NPD
decisions is critical for both types of capabilities - design and manufacturing - but the
factor of ‘Capturing the Voice of the Customer’ is critical only for the design capability.
In this research, customer involvement is defined to include consideration of customer
needs while making key NPD decisions, thus constituting a generic factor for customer
integration. Having a procedure for automatically considering customer needs during the

NPD project seems to impact product planning effectiveness. H8 was also supported.

Table 5.40. Results of Stepwise Regression Analysis of the Effect of
Customer Integration on Product Concept Effectiveness
Standardized Coefficients (t-value)

 

 

 

 

 

 

 

=— ——=
Criterion Variable
Predictor Variables Product Fit with Product Fit with
Manufacturing Design
Capabilities Capabilities
Customer Integration -
Capturing “Voice of the Customer .21 .13
(3 .87)‘ (2.16)‘
Customer Integration -
Customer Involvement in NPD decisions .09 .19 ,
(1.49) (3.10)‘
F-Value 1 5 .00' 14.22'
R2 .04 .08
Adjusted R2 .04 .07
= =—===
‘p<.01

93

Post Hoc An_algses: Global Test of Impact of Integration Mecllanisms on Product
Concept Effectiveness

In order to mitigate the artificial inflation of Type 11 error in testing the
hypotheses 6 to' 8, a global test with em factors of integration mechanisms as independent
variables and product fit to firm capability as dependent variable was carried out. Factor
score regression analyses was conducted for this purpose and the results for fit with
manufacturing capability and fit with design capability as dependent variables are

reported in first and second columns respectively of Table 5.41.

Table 5.41. Results of Factor Score Regression Analyses of the Effects of Supplier,
Design-Manufacturing and Customer Integration on Product Concept Effectiveness
Standardized Coefficients (t-value)

   

 

 

 

Criterion Variable
Predictor Variables Product Fit with Product Fit with
Manufacturing Design
Capabilities Capabilities
Supplier Integration —
Communication and lnforrnation Sharing -.05 .07
(-.81) (1.35)
Supplier Integration -
Design Involvement .00 -.03
(0.01) (-.55)
Supplier Integration - ~
Infrastructure -.02 .10
(-.24) (1.78)c
Design-Manufacturing Integration -
Integrative Use of Technology .19 .35
(3.32)‘ (6.28)’
Design-Manufacturing Integration -
Design Proactiveness .13 .17
(2.39)" (3.34)‘
Customer Integration -
Capturing “Voice of the Customer” .-.05 .04
' (-.85) (.76)
Customer Integration —
Customer Involvement in NPD decisions -.03 -.02
(-.54) (-.28)
F-Valuc 8.32' 24.02'
1?.2 .05 .19
Adjusted It2 .05 .18

 

'p<.01; "p<.05; ‘p<.10

94

The results in Table 5.41 show that both dimensions of design-manufacturing
integration are critical for both types of capabilities - design and manufacturing. This
suggests that design-manufacturing integration is the most important integration
mechanism for enhancing fit with firm capabilities in the areas of design and
manufacturing. Also, the infrastructure dimension of supplier integration was a
significant predictor of product fit to design capability.

5.4.3 Antecedents to Development Process Performance

H9: Product Concept Effectiveness and Develop_r_n_ent Process Performance: Hypothesis 9
proposed a positive relationship between fit with frrrn capabilities and development
process performance. To test this relationship, a regression of the two product fit
variables on development process performance was conducted. The results of the
regression analyses of Product fit to manufacturing capability and Product Fit to design

capability as independent variables and Development Process Performance as the

dependent variable are reported in Table 5.42.

Table 5.42. Results of Stepwise Regression Analysis of the Effect of Product Concept Effectiveness
on the Index of Development Process Performance
Standardized Coefficients (t-value)

Criterion Variable

 

 

 

 

Predictor Variables Tum of Development Process Performance
Product Fit with Manufacturing Capabilities .23 (4.25)‘
Product Fit with Design Capabilities .09 (1.44)
F-Value 18.06'
R2 .05
Adjusted R2 .05
m l u
‘p<.01

95

The results in Table 5.42 suggest that product planning is a necessary first step for
better execution of the development process. In particular, the results point to the
criticality of considering the fit of the planned product to existing manufacturing
capabilities. H9 was supported. It is interesting to note that product fit to design
capabilities did not appear as a significant predictor of development process performance.
This does not imply that this factor is unimportant. In fact an examination of the raw
correlations of each indicator of product fit to design capability with conformance quality
(one of the dimensions of development process performance) revealed significant
correlations of all three items at p<.05 level. This means that the relationship at the
construct-to- construct level was not significant but the relationship at the construct to

variable level was significant.

H10: Supplier Integgtion and Development Process Performance: Hypothesis 10

proposed a positive relationship between supplier integration and development process
performance. To test this relationship, a regression of the three sub-dimensions of
supplier integration on development process performance was conducted. The results are
reported in Table 5.43.

Two of the three dimensions of supplier integration significantly affected
development process performance. As per the results, communication and information
sharing with key suppliers and infrastructure related practices facilitates proper execution

of NPD projects. H10 was supported.

96

Table 5.43. Results of Stepwise Regression Analysis of the Effect of
Supplier Integration on Development Process Performance
Standardized Coefficients (t-value)

 

Criterion Variable

 

Predictor Variables Index of Development Process Performance

 

Supplier Integration —

 

Communication and Information Sharing .11(1.67)c
Design Involvement .03 (.35)
Infrastructure .14 (2.28)"
F-Value 8.59‘
R2 .05
Adjusted R2 .04

 
 

 
  
   

‘p<.01; "p<.05; cp<.10

H11: Design-Manufacturing Integration and Development Process Performance:

Hypothesis 11 proposed a positive relationship between design-manufacturing integration
and development process performance. To test this relationship, a regression of the two
sub-dimensions of design-manufacturing integration on development process
performance was conducted. The results are reported in Table 5.44.

Only the Design Proactiveness dimension of design-manufacturing integration
appeared as a significant predictor of development process performance. The fact that
Integrative use of technology did not appear as a significant predictor is surprising. It is
conceivable that this is an artifact of the way this factor was measured. The scale items
for this factor measured the extent of use of computers to integrate design, manufacturing

and resource planning. It might be that most companies in this sample have already

97

reaped the benefits of these technologies as they might be a “given” in today’s

manufacturing environment as opposed to a source of advantage. On the whole, H11 was

 

 

supported.
Table 5.44. Results of Stepwise Regression Analysis of the Effect of
Design-Manufacturing Integration on Development Process Performance
Standardized Coefficients (t-value)
Criterion Variable
Predictor Variables Index of Development

Process Performance

. I.

 

. A

Design-Manufacturing Integration -

 

 

Integrative Use of Technology .05 (.88)
Design Proactiveness .24 (4.42)‘
F-Value 19.57'
R2 .06
Adjusted R2 .05
‘p<.01 —

H12: Customer Integr_ation and Development Process Performance: Hypothesis 12

predicted a positive relationship between customer integration and development process
performance. To test this relationship, a regression of the two sub-dimensions of
customer integration on development process performance was conducted. The results are
reported in Table 5.45.

Only the ‘Capturing the Voice of the Customer’ dimension of customer
integration appeared as a significant predictor of development process performance
implying that this factor is useful for both product planning and development process

execution. Also, the second dimension of Customer Involvement in NPD decisions
98

which was a strong predictor of product planning did not appear to influence
development process execution. This suggests the limited role of the customer after

major decisions relating to planning have been effectively carried out. H12 was also

 

 

 

 

 

 

supported.
Table 5.45. Results of Stepwise Regression Analysis of the Effect of
Customer Integration on Development Process Performance
Standardized Coefficients (t-value)
Criterion Variable
Predictor Variables Index of Devélopment Process Wormance
Customer Integration —
Capturing “Voice of the Customer" .24 (4.51)‘
Customer Involvement in NPD decisions .03 (.50)
F —Value 20.31'
R2 .06
Adjusted R2 .05
'p<.01

H13: Concurrent Process Management and Development Process Performance:

Hypothesis 13 predicted a positive relationship between concurrent process management
and development process performance. To test this relationship, a regression of the two
sub-dimensions of concurrent process management on development process performance
was conducted. The results are reported in Table 5.46.

Only the Dynamic Iterative Routines dimension of concurrent process
management appeared as a significant predictor of development process performance.

The fact that Downstream Coordination did not appear as a significant predictor is
99

surprising. The results suggest that an ongoing real-time upstream and downstream
coordination is preferred to static downstream coordination. Also, it is possible that
measurement could have played a role on the pattern of this result. This factor largely
measured the early ‘freezing of design’ as a method for effective downstream
coordination. It is plausible that other mechanisms for effective downstream coordination
may have been practiced by firms which were not captured in this research. This issue is
worthy of further exploration. On the other hand, information sharing and joint problem
solving which constituted the Dynamic Iterative Routines Factor seems to be effective for

enhancing development process execution. H13 was supported.

Table 5.46. Results of Stepwise Regression Analysis of the Effect of
Concurrent Process Management on Development Process Performance
Standardized Coefficients (t-value)

 

 

 

 

 

=—
Criterion Variable
Predictor Variables Index of Development P?ocess Performance
Concurrent Process Management —
Dynamic Iterative Routines .27 (5.12)‘
Downstream Coordination .07 (1.28)
F-Value 26.20'
R2 .07
Adjusted R2 .07

 

‘p<.01

100

Post Hoc Analyses: Global Test of Impact of Integration Mechanisms and Concurrent

Process Management on Development Process Performance

In order to mitigate the artificial inflation of Type 11 error in testing the
hypotheses 10 to 13, a global test with gfl factors of integration mechanisms and
concurrent process management as independent variables and development process
performance as dependent variable was carried out. Factor score regression analyses was
conducted for this purpose and the results are reported in Table 5.47.

As can be seen from Table 5.47, Supplier Integration-Infrastructure, Customer
Integration-Capturing VOC and Concurrent Process Management - Dynamic Iterative
entered as significant predictors of Development Process Performance. It may be noted
that design manufactming integration did not enter as a significant factor, possibly
because the impact of such integration is more on up-front activities such as product
planning as opposed to downstream activities relating to development process execution.
A sirrrilar result was found in the case of Concurrent Process Management. While,
Dynamic iterative routines was found to have a significant impact on development

process performance, downstream coordination did not.

101

Table 5.47. Results of Factor Score Regression Analyses of the Effects of Integration Mechanisms
and Concurrent Process Management on Development Process Performance
Standardized Coefficients (t-value)

 

 

 

 

= =
Criterion Variable
Predictor Variables Development Process Performance
Supplier Integration - Communication and Information Sharing .03
(.58)
Supplier Integration - Design Involvement .09
(1.64)
Supplier Integration — Infrastructure .1 l"
(2.07)
Design-Manufacturing Integration - Integrative Use of Technology -.01
(-.10)
DesignoManufacturing Integration - Design Proactiveness .06
(.96)
Customer Integration — Capturing “Voice of the Customer" .12"
(2.11)
Customer Integration — Customer Involvement in NPD decisions .02
(.32)
Concurrent Process Management - Dynamic Iterative Routines .19‘
(3.24)
Concurrent Process Management - Downstream Coordination .08
(1.40)
F-Value 9.68‘
R2 .09
Adjusted R2 .08

 

'p<.01; "p<.05; cp<.ro

5.4.4. Contingency Relationships

Four contingency relationships were also tested. The contingencies related to
Product Innovativeness, Technical Complexity, Market Uncertainty and Stage in the
Product Life Cycle. For all contingency relationships, the dependent variable was time-
to-market. For the first three relationships, the independent factor was Concurrent

Process Management and for the last relationship, it was Supplier Integration.

102

H14: Moderating Influence of Product Innovativeness: H14 stated products that were
low in innovativeness are more likely to reap the benefits of concurrent process
management in terms of lower time to market. In operational terms, this meant that
regression slopes for the high and low innovativeness group were statistically and
significantly different from one another. Another way of testing this relationship is to
include an interaction term along with the main effects of concurrent process
management and product innovativeness. H14 will be supported if the interaction term
enters as a significant predictor of time-to market.

Tables 5.48 a to c display the results of testing H14. No interaction term entered
the equation for both factors of concurrent process management (see Tables 5.48 a and b).
However, when a single practice from the factor, i.e., Frequency of Cross functional
Communication was considered, the interaction term entered as a significant predictor.

H14 was not supported at the construct level, but was supported at the measured variable

 

 

 

 

 

level.
Table 5.48a. Results of the Moderating Effect of Product Innovation on
the Concurrent Process Management (Factor1)-Time to Market Relationship
Standardized Coefficients (t-value)
1 E
Criterion Variable
Predictor Variables Time-to—Market
Concurrent Process Management — Dynamic Iterative Routines .16 (2.91)‘
Product Innovation .00 (.05)
Concurrent Process Management X Product Innovation .01 (.15)
F-Value 8.45'
R2 .03
Adjusted R2 .02
fit u

103

Table 5.48b. Results of the Moderating Effect of Product Innovation on
the Concurrent Process Management (Factor 2) -Time to Market Relationship
Standardized Coefficients (t-value)

Criterion Variable

 

 

 

 

Predictor Variables Time-to-Market
Concurrent Process Management - Downstream Coordination .13 (2.34)"
Product Innovation .02 (.31)
Concurrent Process Management X Product Innovation .03 (.31)
F -Value 5.48"
R2 .02
Adjusted R2 .01
p .

Table 5.48c. Results of the Moderating Effect of Product Innovation on
the Cross Functional Communication -Time to Market Relationship
Standardized Coefficients (t-value)

Criterion Variable
Predictor Variables Time-to-Market

 

Cross Functional Communication — Engg./R&D and manufacturing .05 (0.56)

 

 

Product Innovation -.08 (-1.09)
Cross Functional Communication X Product Innovation .11 (2.05)"
F -Value 4.22"
R2 .01
Adjusted R2 .01
p .

H15: Moderating Influence of Technical Complexity:

H15 stated products that were low in technical complexity are more amenable to
the use of concurrent process management to speed up development as measured by time

to market. The logic being that prior familiarity with the technical know-how could

104

avoid design lead time delays as in the case of new technologies. Table 5.49 a and b state
the results of testing this hypothesis. The second factor of Downstream Coordination did
have a differential impact on time-to-market depending on level of technical complexity.

Thus, H15 was supported.

Table 5.49s. Results of the Moderating Effect of Technical Complexity on
the Concurrent Process Management (Factor 1) -Time to Market Relationship
Standardized Coefficients (t-value)

Criterion Variable

 

 

 

Predictor Variables Time-to-Market
Concurrent Process Management - Dynamic Iterative Routines .16 (2.91)‘
Technical Complexity -.01 (-.12)
Concurrent Process Management X Technical Complexity .02 (.19)
F-Value 8.45"
R2 .02
Adjusted R2 .02

 

Table 5.4%. Results of the Moderating Effect of Technical Complexity on
the Concurrent Process Management (Factor 2) -Time to Market Relationship
Standardized Coefficients (t-value)

 

 

 

 

Criterion Variable

Predictor Variables Time-to-Market
Concurrent Process Management - Downstream Coordination .07 (0.72)
Technical Complexity -.08 (-1.21)
Concurrent Process Management X Technical Complexity .21 (3.94)‘

F-Value 15.50‘

R1 .04
_ Adjusted R2 .04

 

saw
105

H16: Moderating Influence of Market Uncertainty: While H15 dealt with product

characteristics, H16 deals with the characteristics of the target market for the new
product. H16 states that products that are targeted to markets that are less uncertain can
be speeded up more rapidly using concurrent process management techniques as opposed
to products introduced in more uncertain markets. Table 5.50 a and b display the results.
As can be seen from these tables, both types of concurrent process management
techniques appear to work well for speeding up products targeted to less uncertain

markets. H16 was strongly supported.

Table 5.50a. Results of the Moderating Effect of Market Uncertainty on
the Concurrent Process Management (Factorl) -Time to Market Relationship
Standardized Coefficients (t-value)

 

Criterion Variable

 

 

 

 

Predictor Variables Time-to—Market
Concurrent Process Management — Dynamic Iterative Routines .04 (.63)
Market Uncertainty -.07 (-.70)
Concurrent Process Management X Market Uncertainty .20 (3.67)‘
F-Value 13.45'
R2 .04
Adjusted R2 .04
$5"?

106

Table 5.50b. Results of the Moderating Effect of Market Uncertainty on
the Concurrent Process Management (Factor 2) -Time to Market Relationship
Standardized Coefficients (t-value)

 

Criterion Variable

 

 

 

 

Predictor Variables Time-to—Market
Concurrent Process Management - Downstream Coordination -.01 (-.07)
Market Uncertainty -.O3 (-.34)
Concurrent Process Management X Market Uncertainty .21 (3.94)"
F-Value 15.50"
R2 - .04
Adjusted R2 .04
"p<.01 I

H17: Moderating Influence of Stage in the Product Life Cycle: The impact of supplier

integration on reducing time-to-market goals is a subject of controversy because of

inconsistent results. H17 stated that supplier integration facilitates faster time to market

only for products that are further along the product life cycle, i.e., mature products.

According to this hypothesis, supplier integration has no or less impact for markets in the

introduction or grth stage. The results are shown in Tables 5.51 a to c, corresponding

to each dimension of supplier integration. As can be seen from these tables, there was no

support for this hypothesis. This could be because of the composition of the sample. As

shown earlier, the sample was biased towards projects that introduced products that were

in the early stage of the product life cycle.

107

Table 5.51a. Results of the Moderating Effect of Product Life Cycle on
the Supplier Integration (Factorl) ~Time to Market Relationship
Standardized Coefficients (t-value)

 

 

 

 

 

 

' Criterion Variable
Predictor Variables Time-to-Market
Supplier Integration - Communication and Information Sharing .14 (2.59)"
Product Life Cycle ' .06 (1.03)
Supplier Integration X Product Life Cycle .05 (.92)
F-Value 6.73"
R2 .02

Adjusted R2 .02

     

 

"p<.05

Table 5.51b. Results of the Moderating Effect of Product Life Cycle on
the Supplier Integration (Factor 2) -Time to Market Relationship
Standardized Coefficients (t-value)

   
   
   

 

Criterion Variable
Predictor Variables Time-to-Market

 

Supplier Integration — Design Involvement .14 (2.54)"
Product Life Cycle .06 (1.07)

Supplier Integration X Product Life Cycle .08 (1.36)

 

 

 

F-Value 6.45"

R2 .02

Adjusted R2 .02
"p<.05

108

Table 5.51c. Results of the Moderating Effect of Product Life Cycle on
the Supplier Integration (Factor 3) -Time to Market Relationship
Standardized Coefficients (t—value)

 

Criterion Variable

 

 

Predictor Variables Time-to—Market
Supplier Integration - Infrastructure .08 (1.53)
Product Life Cycle .05 (.98)

Supplier Integration X Product Life Cycle .06 (.94)

 

F-Value 2.33 ‘
R2 .01
Adjusted R2 .00

   

'not significant (p>.10)

5.4.5 Summary of Major Findings

The major findings of this research are summarized below:

I All three integration mechanisms examined in this research — supplier integration,
design-manufacturing integration and customer integration - were multidimensional
in nature. Their measurement properties were tested and found to be robust

I Development Process Performance was an unidimensional construct.

I Conformance Quality was more important than design quality

I The three integration mechanisms had a differential impact on different dimensions of
performance

I Components of Development Process Performance had a differential performance
impact

I Product Concept Effectiveness and Development Process Performance were
complementary in nature

109

I Performance impact of the complements were different
I Moderating Effect of Product Innovativeness was found
I Moderating Effect of Market Uncertainty was found

I Moderating Effect of Technical Uncertainty was found

I Moderating Effect of PLC was not found

This chapter presented the descriptive statistics of the research sample, measurement
properties of the different constructs and findings relating to each research question. In
the next chapter, a synthesis of the research findings is conducted. Implications from the
synthesis to theory building and actionable practice are identified. Limitations of this

research are acknowledged and avenues for extensions to this research are offered.

110

CHAPTER 6

DISCUSSION AND CONCLUSIONS

6.1 Introduction

This dissertation had three research objectives: (1) to understand the relative
impact of “product concept effectiveness” and “development process performance” on
new product development project success; (2) to understand the relative impact of
components of development process performance (time-to-market, cost and quality) on
new product development project success; and (3) to identify specific practices that
influence product concept effectiveness, development process performance and project
success. The results relating to each research objective are first summarized and then

synthesized in a subsequent section.

6.2 Determinants of Project Success

The results relating to the first two research objectives are discussed together as
they both examine determinants of project success. Both product concept effectiveness
and development process performance were found to be significantly related to new
product development project success. A more detailed examination revealed that:

I development process performance was more important than product concept
effectiveness, although both factors were significant

I within components of product concept effectiveness, product fit to
manufacturing capability was critical

I within components of development process performance, time-to-market was
consistently the most important predictor

111

I the interaction of product concept effectiveness and development process
performance had a differential impact on individual measures of project
success

I for market share and profitability, the interaction of product fit to
manufacturing capability and development process performance was a
significant predictor

I for break even time, the interaction of product fit to design capability and
development process performance was a significant predictor
6.3. Supply Chain Integration Mechanisms
Research Objective Three related to identifying specific integration mechanisms
that influenced product concept effectiveness, development process performance and
project success. This objective has a special practical relevance because selecting
appropriate sets of antecedent project level practices that affect success in new product
development projects is troublesome and many efforts have been known to fail. This
research explicated a supply chain framework for grouping similar practices along three
types of integration mechanisms in a supply chain - supplier integration, design-
manufacturing integration and customer integration. The first and the last type of
integration mechanisms represent significant inter-organizational types of coordination
that is required for effective management of new product development projects. The
second type of integration mechanism is intra-organizational and arguably the most
critical type of integration with far-reaching consequences on whether or not the project is
a success.
As far as the impact of supply chain integration mechanisms on product concept
effectiveness is concerned, the results reveal that all the three types of integration

mechanisms that were examined had a significant impact on at least one dimension of

112

product concept effectiveness. Also, the results suggest that product fit to manufacturing
capability is influenced by supplier integration (infrastructure dimension), design-
manufacturing integration (Design Proactiveness and Integrated Use of Technology
dimensions) and customer integration (Capturing Voice of the Customer dimension).
Product fit to design capability was also influenced by supplier integration
(communication and information sharing, and design involvement dimensions), design-
manufacturing integration (Design Proactiveness and Integrated Use of Technology
dimensions) and customer integration (Capturing Voice of the Customer and Customer
Involvement in NPD decisions dimensions).

Supply chain integration mechanisms and concurrent process management also
had a significant impact on development process performance. The results imply that
development process performance is influenced by supplier integration (communication
and information sharing, and infi'astructure dimensions), design-manufacturing
integration (design proactiveness dimension) and customer integration (Capturing Voice

of the Customer dimension).

6.4 Contingency Relationships

Support for the four contingency relationships tested were mixed. Technical and
Market Complexity moderated the influence of concurrent process management on time-
to-market as predicted. There was no support for the moderating influence of product
innovativeness on the concurrent process management -) time to market relationship at

the construct level. However, a practice within the concurrent process management

113

factor (frequency of cross functional communication) was found to have a significant
moderating influence on the concurrent process management -) time to market
relationship. Also, there was no support for the moderating influence of stage in the

product life cycle on the supplier integration -) time to market relationship.

6.5 Synthesis

The three measures of project success - market share, profitability and break even
time could be construed as three different (but not necessarily mutually exclusive) 'dials'
for monitoring performance of new product development projects. The 'strategic levers'
that are required to influence these three measures may be different. Accordingly, it is
beneficial to synthesize the results with respect to the influence on these three aspects of
project success.

A strong emphasis on the performance measure of market share at the project level
is consistent with a market pioneering (called "MP" for brevity) strategy which seeks to
penetrate the new market by capturing a high market share by capitulating on the fact that
there is an unmet need that can be fulfilled through the new product. The pioneering
effect is not long lasting and may be as small as a couple of months in the case of product
with low life cycles such as electronics. A strong emphasis on profitability is indicative
of an efficiency and effectiveness strategy (called "EE" for brevity) in which projects vie
with one another in a portfolio such that profitable projects are retained and less
profitable ones are weeded out. This takes place on a dynanric basis via the " go, no-go"
decision. Underlying each decision to pursue or abandon a project are questions relating

to efficient use of scarce resources and effective choice of products and/or markets.
114

While the prevalence of EB strategy has precedence, the enabling factors that constitute
this strategy is not well understood. In an empirical context, the findings from this
dissertation seeks to contribute to this understanding. The last indicator of break even
time is pursued in contexts in which huge investments belie individual development
projects. The strategy to be used in this case called "window of opportunity" (or WOU)
seeks to reduce the time window in which investments are recouped. Both pricing related
practices (such as penetration pricing) and timing related practices (reengineering the
product delivery process) might be deployed in the WOU strategy. All the three
strategies listed above - MP, EE and WOU are not meant to be mutually exclusive.
However, the issue of recommending an unqualified selection of one or more of these
strategies under varying contexts is beyond the scope of this study.
In Table 6.1, the findings relating to the influence of components of product

concept effectiveness on market share, profitability and break even time are displayed.
Very Strong effects (p<.01) are marked as "l J I ", moderately strong effects (p<.05) are

marked as "I I ", and strong effects are marked as "I ".

Table 6.1. Strength of Effects for the Product Fit to Capability Components to Project Success

 

 

 

 

Relationships
Item (IV) Market Share Profitability Break Even Time
(DVl) (DV2) (9V3)
Fit-Mfg w w w
Fit-Des ‘1 \l \I

 

 

 

 

 

As can be seen fi'om the above table, product fit to manufacturing capability had a

moderately strong but consistent impact on all three components of project success,

115

suggesting that ensuring manufacturing capability at the time of product planning is a
viable strategy to be used for influencing all the three measures. The desired capability
are in the following areas: knowledge of process technology, availability of capacity,
experience of manufacturing similar products, ability to handle changes relating to
volume, rrrix and design. The effect of product fit to design capability on components of
project success is also consistent, albeit, not as strong as the product fit to manufacturing
capability. The results appear to imply that an emphasis on planning with
manufacturability issues in mind has performance implications relating to MP, EE and
MOU strategies.

In Table 6.2, the findings relating to the influence of components of development
process performance on market share, profitability and break even time are shown. As
can be seen from this table, using time-to market as a measure of quality of execution of a
development process is invaluable in terms of its eventual impact on project success.
Practices that enhance time-to-market performance appear to be effective across the board

whether the intention is market penetration, pioneering or efiiciency and effectiveness.

Table 6.2. Strength of Effects for Development Process Performance to Components of Project
Success Relationships

 

 

 

 

 

 

Item (IV) Market Share Profitability Break Even Time
(DVD (0V2) (9V3)

Conformance \l W W

quality

Design Quality W

Prodirct Cost \NV \N

Time to market WV W \/\N

 

 

 

 

 

Conformance quality also appears to be a consistent and significant predictor of
all three measures of project success suggesting that conformance quality is an order

116

qualifier for excelling in new product development projects. But projects win orders
primarily through aggressive achievement of time-to market goals. The impact of
lowering time to market appears to be on not only improving project success but also on
improving other dimensions of development process performance such as conformance
quality. This is also evident form the fact that a factor analysis on the four items of
development process performance revealed urridimensionality.

The fact that market share and break even time had the same influence factor with
the strongest effect, i.e., time-to market implies that this measure is effective to monitor
in products that compete on a MP or WOU strategy. A further refinement of these
strategies to include an EE strategy requires an emphasis of lowering unit product costs.

In Table 6.3, the findings relating to the influence of integration mechanisms on
product concept effectiveness, development process performance and project success are
shown. As can be seen from this table, design-manufacturing integration and customer
integration are the two key sources of integration mechanisms that impact ultimate
project success as well as the mediating constructs of product concept‘effectiveness, and
development process performance. Specifically, the Capturing Voice of the Customer
and Design Proactiveness dimensions seem to be the most significant integration
mechanisms as far as impact on success is concerned. In other words, the message is that
being proactive in design by considering manufacturability issues and also transmitting
customer needs related information within the organization is most critical. This is
especially interesting because these two dimensions suggest proactiveness as a driving

force for success, although one involves inter-organizational communication and the

117

other is intra-organizational communication. The implication of this finding on project
structure is that success depends on formal or informal means of tapping into the voice of
the customer, Capturing customer needs and being proactive of design commitments at an

early stage.

Table 6.3. Strength of Effects for Integration Mechanisms to Components of Product Concept
Effectiveness, Development Process Performance and Project Success Relationships

 

 

 

 

 

Fit with Fit with Development Project
Mfg Design Process Success
capability Capability Performance
Supplier Integration:
Commn & Info Sharing ‘1 ‘1
Design Involvement ‘1
Infrastructure \N \NN/ ‘N
Design Manufacturing
Integration:
Integrative Use of Technology \I \/ \IW
Design Proactiveness W‘l W \NV \liN
Customer Integration:
Capturing “Voice of the \/\N \l \l \/ \N‘J \l‘N
Customer” ‘N
Customer involvement in NPD
decisions

 

 

 

 

 

 

In Table 6.4, the findings relating to the influence of concurrent process
management on development process performance and project success are shown. As
can be seen from this table, the downstream coordination dimension of Concurrent
Process Management is the strongest predictor of both project success and development
process performance. Once again, the proactiveness dimension appears to have the
strongest influence on success. This time, the proactiveness shows up in the form of
design freezing at the stage of committing to process choice, tooling and production

feasibility considerations.

118

Table 6.4. Strength of Effects for Concurrent Process Management to Development Process
Performance and Project Success Relationships

 

 

 

Development Project
Process Success
Performance
Concurrent Process
Management:
Dynamic Iterative Routines \/ NI
Downstream Coordination \NV \N

 

 

 

 

In Table 6.5, the findings relating to the influence of integration mechanisms on
components of project success are shown. As can be seen fiom this table, the same two
dimensions of design manufacturing and customer integration as in the case of overall
construct of project success appear as significant predictors. An additional insight
relating to customer integration might be noted. After the dominant factor of Capturing
the Voice of the Customer enters, the other factor of Customer Involvement in NPD
decisions actually impedes project success. An initial speculation for this result suggests
diminishing returns to success after a "distant" data collection and dissemination exercise.
This might be due to a rapid change in customer wants over time that might send
confusing signals after the initial crystallization of the product concept. Further research
is needed to explain this result.

In Table 6.6, the findings relating to the influence of concurrent process
management on components of project success are shown. Sirrrilar to the earlier situation
at the construct level, the dynamic iterative routines dimension is important for all three
measures of project success. The downstream coordination dimension of Concurrent

Process Management does not seem to be important for market share.

119

Table 6.5. Strength of Effects for Integration Mechanisms to Components of Project Success
Relationships

 

Market Profitability Break even time
share

 

Supplier Integration:
Commn & Info Sharing \N
Design Involvement
Infrastructure

 

Design Manufacturing
Integration:
Integrative Use of Technology
Design Proactiveness \N \N \INN

 

 

Customer Integration:
Capturing “Voice of the

Customer” VxN V ‘1 \/ ‘NV
Customer involvement in NPD W (-ve)

decisions

 

 

 

 

 

Table 6.6. Strength of Effects for Concurrent Process Management to Components of Project
Success Relationships

 

 

 

Market Profitability Break even
share time
Concurrent Process
Management:
Dynamic Iterative Routines W VW \/ \/
Downstream Coordination \/ ‘1

 

 

 

 

 

6.5.1 . Pattern of Influences
The summary of pattern of influences is given in Figure 2. The key findings as
can be seen from this figure are:

' Product fit to manufacturing capability was important for all three types of
new product development strategies

I Time-to-market was important for market pioneering and for window of
opportunity strategies

I Reducing product costs was important for the efficiency and effectiveness
strategy

120

I The integration mechanisms of design-manufacturing (proactive dimension)
and customer integration (capturing “voice of the customer” dimension) was
the most consistent predictor. They affected two dimensions of development
process performance (time and cost) and one dimension of product planning
(i.e., fit to manufacturing capability)

I Supplier integration only affected development process performance (both
time and cost) not product planning. The infrastructure dimension of supplier

integration influenced time-to-market and the design involvement dimension
affected product cost

I Concurrent process management not only speeds up time to market, but also
reduces product costs. Only the dynamic iterative routines dimension
influenced time-to-market, but both the dynamic iterative routines dimension
and downstream coordination dimension influenced product cost

The implications of this pattern of findings are worth noting. Fit to manufacturing
capability appears to be a key consideration in managing NPD projects. This dissertation
found that two integrating mechanisms - proactive dimension of design-manufacturing
integration and capturing “voice of the customer” dimension of customer integration —-
could be effective for improving fit to manufacturing capability. Concurrent process
management facilitated the reduction in unit costs of the product. However, its impact on
time-to-market was less uniform. Only the dynamic iterative routines dimension
influenced time-to-market. The role of supplier integration in NPD projects still appears
to be controversial. In this dissertation, supplier integration affected development process
performance, but not product planning. Moreover, the sub-dimensions of supplier

integration had a differential impact on components of development process performance.

Supplier involvement in design influenced product costs but not time-to-market.

121

Investing in infrastructure programs with suppliers appears to affect time-to-market

performance.

 

 

 

 

 

 

 

 

 

 

Concurrent 03mm. I - H ‘
Process w TMarket MN“! Share
Management Bo”,
9M
Supplier Product
Integration Design Cost

 

 

 

 

 

 

 

 

 

 

Product Fit . .
Profitabil
(Mfg Cap.) “3'

 

 

 

 

 

 

Bleak Even
Time

 

 

 

Figure 2 : Pattern of Influences

6.6. Managerial Implications

From a managerial perspective, this research provides product development
managers an understanding of specific integration mechanisms that impact mediators of
success in new product development projects. In particular, critical supply chain
practices that influence product concept effectiveness and development process

performance were identified. For example, a manager may be interested in knowing

122

which practices improve conformance quality and time-to-market could benefit from
examining customer and design manufacturing integration practices.

At the generic level, this research focused on three common objectives of new
product development projects that interest managers, i.e., market share, profitability and
break even time. The results from this research offer normative guidelines to managers
who are interested in achieving high performance among one or more of these objectives.
For example, this research identified “product fit to manufacturing capability of firms” as
an essential part of product planning regardless of the type of new product development
project objective being pursued. Also, the results of this research show that design
manufacturing integration and customer integration are the most crucial integration
mechanisms that enhance product fit to manufacturing capability. Specifically, practices
within the proactive group of design-manufacturing integration and practices within
“capturing the voice of customer’ group of customer integration, are most influential in
enhancing product fit to manufacturing capability of firms.

The conceptual model can be useful to product development managers entrusted
with the audit of the process of product development within their firms. As described in
an earlier chapter, previous research on the determinants of project success have failed to
capture the integrative role of linking mechanisms as a driver of project success. In fact,
this research hypothesized that the impact of such linking mechanisms is via the
mediating constructs of product planning and development process execution. This
division is of particular interest to NPD managers who are interested in understanding the

drivers of the “fuzzy front end” of the development process, as well as the downstream

123

execution of the development process. The results of this research largely supported the
conceptual model, thereby offering a new perspective of managing new product
development projects. One implication of this result is that a supply chain orientation to
managing NPD projects appears to be useful. Although, managing multiple NPD
projects simultaneously, which 'is more reflective of managerial practice, was not
explicitly examined in this research, even at the single project level, a supply chain
emphasis on integration mechanisms appear to matter. A relatively simple model of
supply chain integration, using only three groups of integration, proved to be useful in
demonstrating the importance of integration mechanisms in NPD projects.

While the choice of integration mechanisms in a NPD project is an important
decision given the limitations of resource commitments, managers are equally interested
in the inter-relationships among the different objectives of development process. That is,
managers are interested in knowing whether the development process objective of time-
to-market can be obtained without sacrificing other objectives such as quality or cost. As
discussed in an earlier chapter, this issue is of special interest given the inconsistent
findings in the literature. The findings of this study point in favor of the synergy school.
Specifically, the message to managers is that achievement of time-to-market objectives
has the associative benefits of achieving objective relative to product cost and product
quality. Removing “time out of the system” through cycle time reduction techniques
appear to have simultaneous impact of lowering product costs and improving design and

conformance quality.

124

Through the results of the contingency relationships investigated in this research,
the message to managers is that the impact of concurrent process management on
performance is not unequivocal. Specific conditions under which process concurrency
might be a viable strategy were identified. For example, this research supported the
notion that process concurrency, in general, is more advantageous when deployed in
environments that are suggestive of low technical complexity (product factor) or less

uncertain markets (market factor).

6.7 Academic Contributions

This research contributes to the literature in several disciplines. To the new product
development literature, several insights were found. First, this dissertation took a supply
chain view of integration mechanisms and demonstrated its usefulness in the NPD
context. Interdepartmental integration has been a focus of many research initiatives. In
the product deveIOpment literature, some researchers have investigated the role of
interdepartmental integration as a predictor of success. Hitt, Hoskisson and Nixon (1993)
argue that inter-functional integration can positively influence timely new product
development if integration mechanisms are managed effectively. However, they do not
test this proposition empirically. Empirically, some researchers have found positive
relationships between inter-functional integration and NPD performance. For example,
Kahn and McDonough (1997) found a direct link between interdepartmental integration
and NPD performance. Moenaert, Souder, De Meyer, and Deschoolmeester (1994)

examined the interaction between marketing and research and development (R&D) in 40

125

technologically innovative Belgian companies and found that the quality of the inter-
fimctional climate had a significant effect on project success. This dissertation builds on
this work by proposing that supply chain integration positively influence both project
success and mediators of project success, i.e., product planning and development process
performance. I

Second, the conceptual model of this dissertation proposed and tested an integrated
schema of constructs from two established streams (rational planning and disciplined 1
problem solving) streams, and found that a new level of analysis merits further
investigation, i.e., that of integration mechanisms. This is especially so, considering that
the positive impact of several integration mechanisms were consistently shown to impact
not only the mediating constructs of product planning and development process
performance, but also the ultimate construct of NPD project success. Within integration
mechanisms also, several of the underlying practices were common to the third popular
stream within NPD research, i.e., the communication web stream. Thus, findings from
this dissertation appear to suggest that the three established schools within NPD research
actually converge at a higher level of integration mechanisms.

Third, this research demonstrated the differential impact of integration mechanisms
on individual measures of product planning and development process performance. For
example, supplier integration was found to impact time—to-market and product costs but
not product planning effectiveness. On the other hand, design-manufacturing integration
and customer integration were found to positively influence both product planning as

well as development process execution.

126

Fourth, this research contributes to the contingency school of NPD research which
has been plagued with controversial and inconsistent results. Specifically, concurrent
process management was shown to have contingent benefits on time-to-market goals
depending on product and market conditions. Support for the limited role of concurrency
process management in accelerating NPD projects in less technical, less innovative and
less uncertain market was found.

To the business strategy literature, this dissertation makes a contribution by
empirically supporting the complementary role of planning and execution at the new
product development project level. Prior work in business strategy literature has found
the complementary relationship at the project level. For example, Bryson and Bromiley
(1993) in an exploratory study of 68 case descriptions of major projects found that a
number of contextual variables strongly influenced aspects of the project planning and
implementation process and also indirectly influenced project outcomes through the
planning and implementation process. In this dissertation, integration mechanisms were
found to facilitate the complementary role of planning and execution on NPD project
success.

To the operations management literature, this dissertation makes a contribution by
empirically supporting the synergy school of thought which argues that superior cost,
quality and time performance can be achieved simultaneously. While, the existing
literature in operations management have focused on the plant level as the unit of

analysis, this dissertation found support for the synergy school at the NPD project level.

127

Specifically, improvements in time-to-market were associated with improvements in cost
and quality performance.

Finally, this dissertation makes contributions to research methods by isolating
reliable and valid scales for new constructs (for example, three types of integration

mechanisms, and fit to firm capability) at the NPD project level.

6.8 Limitations of the Research

Some limitations of this research are worth noting. This paper addresses the
impact of integration mechanisms on a ‘static’ basis. Hence, the evolution and learning
associated with usage of integration mechanisms could not be captured. This would
require the use of multiple longitudinal case studies. Also, the impact of integration
mechanisms on performance is subject to the confounding of extraneous effects that
could not be controlled in this study. Field studies lack the rigor of control for extraneous
effects as is possible in lab experiments. This continues to be a pragmatic challenge as
the use of lab experiments for studying quality tool usage is prohibitively expensive,
especially if one is using objective data from companies.

In this research, we focused on project level (not firm) variables and on firms
from several industries. Many other factors can impact new product development project
success. The importance of different dimensions of project success itself may vary across
and within industries, and practices that positively influence success in one industry may
be more or less effective in another. The complexity, hostility, and dynamism of the

environment, industry concentration and market dominance can influence all aspects of

128

innovation. For example, the automotive industry context is not as volatile as that of
software or electronics, in which product life cycles may be as low as a couple of months.

Overall, the results must be interpreted with caution about unwarranted generalizations.

6.9 Directions for Future Research

This research examined the association between integration mechanisms and
mediators of project success, their causal relationships is still open to question. A logical
extension of this research could consider the causal directions among the constructs. Of
particular interest is the integration mechanisms —> product concept effectiveness link. In
this research, there is an implicit assumption that selection of appropriate integration
mechanisms can enhance the effectiveness of planning of the new product. It is
conceivable to hypothesize that the causal direction is the reverse, i.e., effective planning
helps determine appropriate integration mechanisms. Future research could investigate
this possibility.

In this research, cross-sectional research design was used to study the role of
integration mechanisms. Future efforts could involve longitudinal research designs that
capture the evolutionary pattern of adoption of integration mechanisms. Also, the
longitudinal performance impact of integration mechanisms could be captured using
dynamic modeling.

This research mainly focused on new product development projects that were
managed in North America (US and Canada). Recent research has suggested that the
manner in which new product development projects are managed might differ fiom one

country to another (Calantone, Schmidt and Song, 1996; Nakata and Sivakumar, 1996).
129

While the research cited did not specifically study integration mechanisms, future
research could address the issue of whether national culture plays a role in the selection
and deployment of integration mechanisms. If so, are the performance consequences
different from that obtained in this research?

This dissertation was a first step toward generating systematic assessment of the
effects of “supply chain” integration mechanisms on new product development project
success. Judging from the results of this research, it is clear that future research could
identify other sources of integration mechanisms and relate these to project success. For
example, technology integration may be an important component of managing new
product development projects. Similarly, human resource integration could be another
component that can be investigated.

This research opens some interesting new lines of research by shifting the focus
from the new product development project level to the level of integration mechanisms.
Such a move would constitute a ‘finer grain’ of analyses in new product development
research. In this unit of analysis, the focus is on how (and why) integration mechanisms
affect project success across firms. Similarly, the focus can be shifted at a higher level to
the strategic business unit in which multiple products (albeit homogeneous to one
another) are developed and introduced simultaneously. It might be that the role of
integration mechanisms is more complex in such an environment. Finally, the focus can
be shifted to the firm level, i.e., multiple heterogeneous products. While there has been
considerable research at the firm level (for a critical review, see Damanpour, 1991), the

precise role of integration mechanisms in an environment that includes diverse product

130

and customer markets has not been examined. This could be a fruitful research

opportunity to pursue in the future.

131

APPENDIX A : SURVEY INSTRUMENT

LABOUTIHIS sprayer; .;.

u-Iu Iv-A'a

This survey is designed to provide new insights into the management of new product development process by assessing the impact
of supply chain practices on success in new product development projects(NPD). The survey is being administered nation-wide to
firms in multiple industries. The survey asks about project objectives, enabling practices, product concept effectiveness and project
success. This research is supported by a Doctoral Dissertation Research Grant from National Association for Purchasing
Management (NAPM).

amen: -.. .....-r....“......-...-........“l“.'
A report on the survey findings will be sent exclusively to participants in Summer 1998. The report will help new product
development managers to identify

0 the relative importance of product concept effectiveness and development process performance in NPD projects

0 the relative importance of cost, quality and lead time objectives in NPD projects

0 enabling strategies that influence the attainment of cost, quality and lead time objectives

- enabling strategies that strengthen the fit of the planned product to existing firm capabilities

o determinants of overall NPD project success

All responses will be kept confidential. No individual companies, projects, or individuals will be identified. All data will be
analyzed and presented on aggregate basis only.

  

- . . wia'wm.wfivverr :r:::: ‘1'?“ '9' v 2 NW """ ”if 7 WW"- “11'1"“?!'""."‘-'-"=‘<ih"~.’¥':'"IT-~‘Ju‘~*‘£"’.'.3"'."‘-IE}Ewmf‘W-‘YH'WJF‘CTFifi!!!
_ . , '7 ‘ , ,_, > ' " . _ _ .. . _ " ' _ _ . ‘ _ _ : . _ _: f.
:4 .nbi")LIIM‘.q'iEET 3269;312:91- i:. r s i';¥:‘M‘UJ:MJ- I"- '.-_- Fit—In} 3‘15" '48— “; .‘-‘*.I1-"MF.'-i ~.-- .3 In . urn-Zn.“ nh‘f: ll; - uh” -‘O'°vfl'.l~-:\ -r,'.si‘dd¢ulfl'-Vl‘1\ '2': Clvfil‘ his": rain-nit

  

 

 

 

_. ., , . : - .: : w: n :. .. .— 7'..s'.'--.~.-.‘--..-:1I -:n-.1~er's'v.-‘~~r~-nu s-‘-‘7l.'=’-’ Never 7 '1'. .W'T:"."W-P'.“'I"."I.""'. -r~:1y-wen~ V;"PF‘J"_". .‘-'-‘L-':*EI!£‘.' , ' , ._.,.. ‘7,

,. _ . ‘ ' : ‘ ,‘ , ..

.7 ' . I

“. ,,. 2' , :frv,~"' -.-: .i*-: - ,' ' '_: ' , - ' ~ . ., : ,, - . ." .~ . ..- .. , ~-
g- 5"“. n‘.TJ.:9fi«MM,fixv,-fi :1 {dAufiVP- new—"’- : -- u: - v7 3:. '.- ,. mp ..:,: -- .. ..,. .42..» 4.. : -'~ a. . a. :0 u ,.,: . - If _ 3. r’ 'ps“, .I~1.v. nu.5‘;‘fx‘1&.’fi’{Mkzng-L‘mh‘g.3:1J3an-‘Juv 3;“

1. This survey takes 25 to 30 minutes to complete.
2. This survey asks about a recently completed ( in the last 3 years ) product development project.
Please select a project where:
0 first customer shipment has occurred, and
o the product is of either a manufactured or assembled nature
3. Due to the nature of the questions, for most companies the appropriate person to complete this survey is a person with overall
responsibility for the development project. The respondent must have been involved in the project from start to end, and
should have interacted with both upper management and project personnel for key project decisions.
4. Please return the survey using the enclosed envelope.
5. If you have any questions regarding this survey, please feel free to call (517) 353-6381, extn. 275 (Jay) or e-mail me at

jayaramm@pilot.msu.edu.

We are committed to providing finely and actionable malts.
Thank you for your involvemnl.

132

INSTRUCTIONS
0 Please answer all the questions.
0 If you do not know the answer to any question, or understand the question, mark DK by the question.

W --....,_.. -_....--.... “ 5119mm.fieTBQDUCIPEYEEOPMEWPRMUYT:.4...‘7'1.. If. j‘:‘.li§..".;.;

IL‘-Ao.a~¢M~‘.wAW~M—a. .a-_-.~~ “Ham-4“

Pick a new product development project completed in the last 3 years for which you were a project manager. By new product we
mean any product that is new to your company or division. This would also include

o innovations

0 new product lines

Name of the product for which you are responding:

 

Description of this product:
n”%‘¢¢mv?~?rfifli riff—pt‘ ::“i‘eu‘ ‘ ’5‘: '3'vr' :A‘.‘ - - . . . T ‘1‘, ‘ ‘ * ' " 'm .. ‘ mr-"‘sw?;fi-kfigfif, ,v._:-.::-‘_,: “9‘- - jr'sl. .‘j' ‘v‘. 9”"-"~""."3:‘:“""‘.'3","‘._"
‘ I g .‘ud'Mb ‘nfiQ‘LQEWB’m‘ikly swan: .. “ manna-an 1- , 1 ‘ ‘ ' -" ' .u' :::-an. ‘ ‘~ ‘3‘

Wmmmu'mmns «HI-“ll;

 

I
f

The following questions ask about the different practices used by your company in supplier integration, design-manufacturing

integration, customer integration, and concurrent process management.

INSTRUCTIONS
I. On a 0 to 10 scale. indicate the extent to which each of the following practices was used by circling the appropriate
nunrber, and

2. In the final column (far right), indicate the effectiveness of the practice to the successful outcome of this specific project.
Put a number from O to IO (where IO - highly effective and O - not effective).

 

 

fir #9,... "WWW“"“” ”wwmw’w”MW”?‘§;7T33‘T‘."T“'.’TT "W 3‘" ..‘-..‘...Z'.I‘;Z:
Effectiveness
Not Some Great Scale
Used Extent Extent I-IO
Participation of key suppliers in NPD team 0 l 2 3 4 5 6 7 8 9 10
Direct communication with key suppliers 0 l 2 3 4 S 6 7 8 9 IO
Shared education & training programs with key suppliers 0 I ' 2 3 4 5 6 7 8 9 10

Common linked information systems(EDI,CAD/CAM,e—mail) O l 2 3 4 5 6 7 8 9 IO

Colocation of project personnel and key suppliers 0 l 2 3 4 5 6 7 8 9 IO _
Sharing design knowledge with key suppliers 0 l 2 3 4 5 6 7 8 9 IO _
Sharing manufacturing knowledge with key suppliers 0 I 2 3 4 5 6 7 8 9 IO _
Sharing customer requirements with key suppliers 0 l 2 3 4 5 6 7 8 9 IO _
Frequency of communication with key suppliers during: Effectiveness
“As Very Scale
Seldom Needed” Frequently l-lO
Concept stage 0 I 2 3 4 5 6 7 8 9 IO _
First prototype stage 0 l 2 3 4 5 6 7 8 9 IO _
Full production stage 0 l 2 3 4 5 6 7 8 9 IO _

133

Involvement of key suppliers in:
Effectiveness

Not Some what Highly Scale
Involved Involved Involved l- l 0
Defining the architecture of' new products 0 l 2 3 4 5 6 7 8 9 IO _
Setting design specifications 0 I 2 3 4 5 6 7 8 9 IO _
Product design 0 I 2 3 4 5 6 7 8 9 IO _
Prototype building and small scale testing 0 l 2 3 4 5 6 7 8 9 IO _
Extent to which input and suggestions were solicited from key suppliers on:
Effectiveness
Some Great Scale
Never Extent Extent I-IO
Design modifications 0 l 2 3 4 5 6 7 8 9 IO _
Problem solving 0 I 2 3 4 5 6 7 8 9 IO _
Reduction in number of parts and critical components 0 I 2 3 4 5 6 7 8 9 IO _
To what extent are the following activities typical of your firm’s orientation with Its key suppliers :
Effectiveness
Not Some Very Scale
Typical Extent Typical l-IO
Our key suppliers assume financial
risks in our NPD projects 0 l 2 3 4 5 6 7 8 9 IO _
NPD project goals are agreed upon
by our key suppliers prior to start of the project 0 I 2 3 4 S 6 7 8 9 IO _
mm Mawmxtwwflawleéi * _; 12?.” I - Tsjsis‘wf.
To what extent were the following practices used? Effectiveness
Not Some Great Scale
Used Extent Extent l-IO
Design for manufacturability O l 2 3 4 5 6 7 8 9 IO _
Job rotation between design and manufacturing personnel 0 I 2 3 4 5 6 7 8 9 IO __
Translating customer preferences into product parameters 0 I 2 3 4 5 6 7 8 9 IO _
Translating customer preferences into process parameters 0 l 2 3 4 5 6 7 8 9 IO _
To what extent were computers or computerized equipment used for linking the following pairs of activities in NPD projects?
Effectiveness
Not Some Great Scale
Used Extent Extent l-lO
Design and production planning
(i.e., CAD data directly related to CAPP) O I 2 3 4 5 6 7 8 9 IO _

Design and production (i.e., CAD data directly controlling

production equipment such as CNC machines, robots or FMS) O l 2 3 4 5 6 7 8 9 IO _
Design and resource planning (i.e., parts data fiom CAD directly

linked to materials requirement planning (MRP) software) 0 I 2 3 4 5 6 7 8 9 10

134

Womlmezrwan‘l’rmca . ' '

To what degree were formal or informal processes used for : Effectiveness
Not Some Great Scale
Used Extent Extent I-IO

including customers in decisions concerning NPD 0 I 2 3 4 5 6 7 8 9 IO _

including customers in decisions concerning manufacturing 0 I 2 3 4 5 6 7 8 9 IO _

communicating key technology aspects of new product to customers 0 I 2 3 4 5 6 7 8 9 IO _

changing product standardstothe benefit of customers 0 I 2 3 4 S 6 7 8 9 IO __

listeningtocustomer’s needs while developingthcproduct concept 0 l 2 3 4 5 6 7 8 9 l0

studying how customers use the firm’s products 0 I 2 3 4 5 6 7 8 9 IO _
meeting with customers 0 I 2 3 4 5 6 7 8 9 IO _
sharing data on customer needs across functions 0 I 2 3 4 5 6 7 8 9 IO _

reviewing NPD efforts to ensure that they match customer needs 0 I 2 3 4 5 6 7 8 9 IO _

W “mlfi Ei'i‘l'i-iér'; x1!“ "' 3.9-1". 15".“ .. 1'31". ‘ "'.'T‘. s 3' ’3' $29“? .a'I‘E-‘Fi ”mm-2&6. .‘Ir‘O‘D'f'a'L'. -1‘;'._"5‘ 2Mfi‘; “Ni-:53”,
‘ I’ I I ‘
w . .r -1 .H-- 4.». ma». 1.33119. 1 1.! 4«¥‘-.'=h’u2£ 5%? - i. ._ £0- -‘---'.~1‘ .4“; vac-:4. : Iv.|.A?-.-.‘..:.'-.. ' ' J. J-i-k°\;&i&M'WTWMWM‘§-m

The following questions address infonnatr'on sharing, communication and joint problem solving among desim, engineeringlMD,
and manufacturing.

Effectiveness
To what degree was: Not Some Great Scale
at all Extent Extent I-IO
Information made readily available to manufacturing
as the product design evolved? O l 2 3 4 5 6 7 8 9 IO _
Information made readily available to engineering/R&D
asthemsdesign evolved? O I 2 3 4 S 6 7 8 9 IO _
Joint problem solving used in product desig phase
oftheproject O I 2 3 4 5 6 7 8 9 IO _
Joint problem solving used in process desigp phase
oftheproject 0 I 2 3 4 5 6 7 8 9 IO _
Effectiveness
0% 50% 100% Scale
How complete was the: Complete Complete Complete I-IO
mss desim when engineering/R&D ended their active
involvement in product design? [A group is actively involved
if they are consulted for key ensuing NPD project decisions] 0 I 2 3 4 5 6 7 8 9 IO _
product desim when manufacturing provided feedback
of production feasibility? O I 2 3 4 5 6 7 8 9 IO _

product desimwhenmanufacturingmadeformal cost estimates? 0 I 2 3 4 5 6 7 8 9 IO

product design when manufacturing made

commitments to purchase materials, tools and equipment? 0 I 2 3 4 5 6 7 8 9 IO _
Effectiveness
Once it Once a Atleast once Scale
Month (or longer) Week a day I-IO

How often did engineering/R&D and manufacturing
communicatebetweenthemselves? O I 2 3 4 5 6 7 8 9 10—

135

SECTION C. PRODUCT FIT
The following questions ask about the effectiveness of the new product in terms of matching the capabilities of the firm.

— “3.3.3.2.; ... ”2-..... .;.~..-,, .. - ' a-.—. ' ...,«_. .3 r

At the time of project planning. to what extent did the firm have the following skills and resources needed for the targeted
product?

No Expertise Moderate Substantial
Expertise Expertise
Design
Computer Aided Engineering 0 I 2 3 4 5 6 7 8 9 IO
Electronic Data Interchange O l 2 3 4 5 6 7 8 9 10
Prior Experience (of designing
similar products) 0 I 2 3 4 S 6 7 8 9 IO
Manufacturing
Knowledge of Process Technology 0 I 2 3 4 5 6 7 8 9 IO
Availability of Capacity 0 I 2 3 4 5 6 7 8 9 IO
Prior Experience (of manufacturing 3
similar products) 0 I 2 3 4 S 6 7 8 9 IO
Volume Flexibility (Ability
to handle changes in production
output) 0 l 2 3 4 5 6 7 8 9 IO
Mix Flexibility (Ability to handle
changes in product lines) 0 I 2 3 4 5 6 7 8 9 IO
Changeover Flexibility (Ability to
handle major design changes) 0 I 2 3 4 5 6 7 8 9 IO

At the end of the project, to what extent did the firm achieve the originally planned objectives on acquiring the following skills
and resources?

Low Moderate Excellent
Achievement Achievement Achievement
Design
Computer Aided Engineering 0 I 2 3 4 5 6 7 8 9 IO
Electronic Data Interchange O I 2 3 4 5 6 7 8 9 IO
Prior Experience
(of designing similar products) 0 I 2 3 4 5 6 7 8 9 IO
Manufacturing
Knowledge of Process Technology 0 I 2 3 4 5 6 7 8 9 IO
Availability of Capacity 0 I 2 3 4 5 6 7 8 9 IO
Prior Experience
(of manufacturing similar products) 0 I 2 3 4 5 6 7 8 9 IO
Volume Flexibility
(Ability to handle changes in production output) 0 l 2 3 4 5 6 7 8 9 IO
Mix Flexibility
(Ability to handle changes in product lines) 0 I 2 3 4 5 6 7 8 9 IO
Changeover Flexibility
(Ability to handle major design changes) 0 I 2 3 4 S 6 7 8 9 IO

136

.- ' SECTION Dz PROJECT. OBJECHVES , , .
The questions below address the achievement of the original objectives of the product development process performance.

259.“ What—JEEPWWWWQFfifihai ' ' ‘

. -.-... -‘ ..... ... ”we. . .- . -.. .. ..- ..,,.__. .. .. .--4 . -. ... 3,,“ .9..;.-,.._.w 7“

To what extent did the product

Worse On Target Exceeded Targets
Conform to quality objectives set? 0 I 2 3 4 5 6 7 8 9 IO
Conform to design objectives set? 0 I 2 3 4 5 6 7 8 9 IO
Conform to cost objectives set? 0 I 2 3 4 5 6 7 8 9 IO
Conform to time-to-market
objectives set? 0 I 2 3 4 5 6 7 8 9 IO

 

. . :mmuimarnne~wrvms~ "’ fismmlrnmm --1rrnn~”

 

.h. .-a'.‘»- __.- .- ‘. - ‘ __ ‘.:,' ~._ '_ " -~'_ , a ‘r r l -' ".fi‘z“ J "" ' .. :' "".‘ ' ' " K K, #- MI“ ‘
kim8_wtm_fi§~'iui-'L¢--'.fnp._l.(".-.anI"; ... . ..:.hI,-a-_f’s:¢\.w3-fiu.‘ .I-. .1. run; 4 . u.‘ .1- .1». .‘ - - L, ,4.‘ 3.... .lu‘mv mus-km be. '. - IE.» «.411 “heal.- '2’, . .'~.&-‘

The questions below address the overall success of the new product development project.

hm ml?” :1 awfiWristfi- .::-x .nu'. a r w- -.‘!"'2' r; rr-s'm'rn.’ tiara-rise: .3 7r- :.' Mme-tsp: -.~-: arenas 2r»? - . f ' ‘ r.“—

' err-2:.-

    

Saégfsixmuac; ..-....'a, .:. .:.'.'.-..‘ zu‘l“4—K.h-h\“ - an!» “a. \‘u‘ ..s.». 22.2». ...:w-12.Js¢..:.~..t:4u.‘ :7. -:;.._.r._ ..1.. _.....';.:,.lr;;n.--.-;Aw

To what extent was the project successful In terms of

Unsuccessful Successful Highly
Successful
Initial market share? 0 l 2 3 4 5 6 7 8 9 IO
Profitability? O I 2 3 4 5 6 7 8 9 IO
Break-even time? 0 I 2 3 4 5 6 7 8 9 IO
overwrite"area-1v mm WW' W WWW . = .. ~ ~ -. : -. '. x .‘ - - a. - ~ ..-.. ' ‘ > a . WW1?“ WWWWWWWW W‘rWWWWWQ
“131m2w4. a.1«‘.¢.ru.'a-as-wevnuu. s‘k‘ 33:” -'a."*¢.ar§1..'.'.- ..:u. .6. -ma—I-sum A; A a - - ‘ . . ‘1. '\ 1 ' a ‘ '1 '3':- .: ‘ t 1‘ "u' ' '3 a. 'A-Ir .‘ .-.v _; lufliIt—lmyfl-l‘“; nu.) ssh-:6.

 

Number of employees in your company (circle one): <50 5 I-200 201-500 50I -2000 2000 +
Age of company (circle one): <6 years 6-I4 yrs. IS + yrs.
Cumulative person hours worked on this project (in thousands) :

I-4 5-IO II-30 31-60 6I-I00 101-250 251+

Project duration from product concept to first customer shipment was (circle one):
l-12 mos I3-24 mos 25-36 mos 37-48 mos 49-60 mos 6I+ mos
The product has been in the market for: (circle one):
I-l2 mos I3-24 mos 25-36 mos 37-48 mos 49+ mos
The product was targeted for this market (circle one):
Consumer Industrial Other

 

The approximate price of a single unit in US. dollars is (circle one):
51-10 I I-IOO IOI-IOOO 1001-10,000 I0,00I-IO0,000 IO0,00I-I,000,000 I I,000,000 +
What is the location of this product in its product family stream (circle one):

Superficial Minor Major New Product
Change Extension Extension Family platform

0 I 2 3 4 5 6 7 8 9 IO

137

With respect to the new product, please rate: (circle one):

Low Moderate High
Technical know-how embodied in the new product 0 I 2 3 4 5 6 7 8 9 IO
Innovativeness of product to the intended market 0 I 2 3 4 5 6 7 8 9 10
Very Moderately Very Even
unstable stable and stable
Predictability of market demand 0 l 2 3 4 5 6 7 8 9 IO

At what stage of the product life cycle was your main market for the targeted product? (circle one):

Introduction Early Growth Early Maturity Decline
Growth Maturity
What was your primary role on this project? (circle one):
Design Manufacturing Marketing Program Other
Engineering Engineering Management

THANK YOU FOR COMPLETING THIS SURVEY.
PLEASE RETURN YOUR COMPLETED QUESTIONNAIRE IN THE ENCLOSED POSTAGE-PAID ENVELOPE

 

To receive the summary report, please attach your
business card here or write your address in the
space at left.

 

 

 

 

 

 

 

138

APPENDIX B: COVER LETTER

<<Date>>

Dear <<Title>> <<Last Name>>z

The Eli Broad Graduate School of Business at Michigan State University, with support from the National
Association of Purchasing Management (NAPM) is conducting research with over 5,000 companies
nationally regarding Supply Chain Integration in New Product Development (NPD) Projects. We
would like your firm’s participation in this critical benchmarking research beginning in February, 1998.

Launching successful new products continue to be a challenge to most companies. Yet, when done right,
new products have a significant impact on a firm’s competitiveness through improvements in market share,
profitability and overall innovative capability. Through the findings of this research project, we are
developing detailed information about:

I Strategies and practices that are most effective to achieve success in NPD projects
'- Relative role of cost, quality and time-to—market on NPD project success

'- Practices that influence capabilities of firms to develop new products

- Impact of capabilities on NPD project success

In return for your participation, you will receive a benchmarking summary report of your score on
supply chain practices relative to the sample of respondents in August 1998. This summary report
will provide your firm with valuable strategic and tactical benchmarking information for achieving
supply chain integration in NPD projects. All company specific information will be kept strictly
confidential, and no companies or individuals will be identified in the research findings.

A survey has been developed which will take approximately 25 to 30 minutes to complete. The
.appropriate respondent for this survey is the person with overall responsibility for a recently completed
new product development project. You are encouraged to involve multiple participants of the NPD project
team so that we can provide you with the best and most comprehensive benchmarking information.

We sincerely hope that your firm would participate in this important benchmarking research and believe
that the resulting information will be valuable in enhancing your firm’s competitiveness. We look forward
to your response.

Sincerely,
Jayanth Jayaram Ram Narasimhan Roger Calantone
Doctoral Candidate Professor Professor and Associate Dean

139

APPENDIX C: REMINDER POST CARD

<<Date>>

Dear <<Title>> <<Last Name>>:

About 10 days ago, I mailed to you a survey regarding supply chain integration in new
product development projects. If you have already responded, thank you very much. If
not, is there any chance you could fill it out and mail it now? I would be most grateful.
If you need another copy of the survey, please call me at (517) 355-1237 or (517) 353-
6381 Ext. 275, and leave your name and address. I will send a copy immediately. Thank
you!

Sincerely,

Jayanth Jayaram
Doctoral Candidate - Michigan State University

140

APPENDIX D: FOLLOW-UP LETTER

<<Date>>

Dear <<Title>> <<Last Name>>:

Recently, I mailed you a survey as part of my dissertation project which I believe will provide insight into
the impact of supply chain integration on the success of new product development (NPD) projects. If you
have already returned the survey, thank you for your participation. If you have not yet had a chance to
complete the survey, could you please take a few minutes and fill out the enclosed survey now? A
completed survey from your company is invaluable to the success of my dissertation.

In return for your participation, I will provide you with a customized report which benchmarks your firm
with the sampled firms on detailed information such as:

0 Strategies and practices that are most effective to achieve success in NPD projects
0 Relative role of cost, quality and time-to-market on NPD project success
0 Practices that influence capabilities of firms to develop new products

All company specific information will be kept strictly confidential, and no companies or individuals
will be identified in the research findings. To receive the customized report please staple your business
card to the completed survey and return in the enclosed stamped envelop. I expect to mail you the
customized report no later than August 1998.

The research proposal for this dissertation project won a national award in a competition hosted by
National Association of Purchasing Management in 1997. Both the pretest of my dissertation survey and
feedback from early respondents indicated that the survey typically took less than 20 minutes to complete
and most found it an enjoyable and informative experience. To the best of my knowledge and bmed on
feedback from initial respondents, there are no questions on my survey that require sharing any proprietary
information. However, you be the judge of this.

If you have any questions, please e-mail me at jayaramm@pilot.msu.edu or call me at (517) 355-1237 or
(517) 353-6381 ext. 275. Again, I believe this study will be useful to you and look forward to your
favorable response.

Sincerely,

Jayanth Jayaram
Doctoral Candidate

141

REFERENCES

Ahire, S.L., Golhar, D.Y., and Waller, MA. (1996). Developing and validation of TQM
implementation constructs. Decision Sciences, 27(1), Winter, 23-56.

Ali, A., Krapfel, R., and LaBahn, D. (1995). Product innovativeness and entry strategy:
Impact on cycle time and break even time. Journal of Product Innovation Management,
12(1 January), 54-69.

Anderson, James C., and Gerbing, David, W. (1988). Structural equation modeling in
practice: A review and recommended two-step approach. Psychological Bulletin, 103(3),
411-423.

Andrews, KR. (1987). The Concept of Corporate Strategy. IrwinzHomewood, IL.

Armstrong, J. Scott and Terry S. Overton. (1977). Estimating Nonresponse Bias in Mail
Surveys. Journal of Marketing Research, 14 (August), 396-402.

Allen, T. J. (1971). Communications, technology transfer, and the role of technical
gatekeeper. R and D Management, 1, 14-21.

Atuahene-Gima, K. (1995). An exploratory analysis of the impact of market orientation
on new product performance: A contingency approach. Journal of Product Innovation
Management, 12, 275-293.

Bagozzi, R., Yi, Y., and Phillips, L. (1991). Assessing construct validity in
organizational research. Administrative Science Quarterly, 36, 421-458.

Bobko, P., Karren, R., and Kerkar, S. (1987). Systematic research needs for
understanding supervisory-based estimates of SDy in utility analysis. Organizational
Behavior and Human Decision Processes, 40, 69-95.

Birou, L. M., and Fawcett, S. E. (1994). Supplier involvement in integrated product
development: A comparison of US and European practices. International Journal of
Physical Distribution and Logistics Management, 24(5), 4-14.

Bollen, K. (1989). Structural Equations with Latent Variables. New York: John Wiley
and Sons.

Bonaccorsi, A., and Lipparini, A. (1994). Strategic Partnerships in New Product
Development: An Italian Case Study. Journal of Product Innovation Management, I 1,
1 34-145.

142

Booz, A. a. H., Inc. (1982). New Product Management for the 1980's. New York: Booz
Allen and Hamilton.

Boyer, K. K., Ward, Peter, Leong, Keong. (1996). Approaches to the factory of the
future: An empirical taxonomy. Journal of Operations Management, 14, 297-313.

Brannan, B. (1990, ). Why design for assembly? Cost and quality improvements at
Motorola. Paper presented at the Proceedings of the 1990 International Forum on Design
for Manufacturing and Assembly, New Port, Rhode Island.

Brown, S. L., and Eisenhardt, K. M. (1995). Product Development: Past Research,
Present Findings, and Future Directions. Academy of Management Review, 20(2), 343-
378.

Bryson, J .M. and Bromiley, P. (1993). Critical factors affecting the planning and
implementation of projects. Strategic Management Journal, 14(5 July), 319-337.

Calantone, R. J ., and di Benedetto, A. (1996). Performance and T ime-to-Market:
Accelerating Cycle Time with Cross-Functional Teams (Working Paper). East Lansing,
MI: Michigan State University.

Calantone, R. J ., Schmidt, J. B., & Song, M. X. (1996). Controllable factors of new
product success: A Cross-National Comparison. Marketing Science, 15(4), 341-358.

Campbell, D.T., and F iske, D.W. (1959). Convergent and discriminant validation by the
multitrait-multimethod matrix. Psychological Bulletin, 56(1), 81-105.

Capon, N., and Glazer, R. (1987). Marketing and Technology: A Strategic Coalignment.
Journal of Marketing, 51 (July), 1-14.

Churchill, GA. (1979). A paradigm for developing better measures of marketing
constructs. Journal of Marketing Research, 16, February, 64-73.

Clark, K. B. (1989). Project Scope and Project Performance: The effect of parts strategy
and supplier involvement on product development. Management Science, 35(10 (
October)), 1247-1263.

Cohen, M. A., Eliashbcrg, J ., and Ho, T.-H. (1996). New Product Development: The
performance and Time-to-Market tradeoff. Management Science, 42(2 (F ebruary)), 173-
186.

Cooper, R. G. (1979). The Dimensions of Industrial New Product Success and Failure.
Journal of Marketing, 43(3 Summer), 93-103.

143

Cooper, R. G. (1990). Stage-gate systems: A new tool for managing new products.
Business Horizons(May-June), 44-54.

Cooper, R. G., and Kleinschmidt, E. J. (1987). New Products : What Separates Winners
from Losers ? Journal of Product Innovation Management, 4, 169-184.

Cooper, R. G., and Kleinschmidt, E. J. (1994). Determinants of Timeliness in Product
Development. Journal of Product Innovation Management, 1 I , 38 l -3 96.

Cordero, R. (1991). Managing for Speed to Avoid Product Obsolescence : A Survey of
Techniques. Journal of Product Innovation Management, 8, 283-294.

Crawford, M. C. (1992). The Hidden Costs of Accelerated Product Development. Journal
of Product Innovation Management, 9, 188-199.

Cronbach, L.J. (1951). Coefiicient alpha and the internal structure of tests.
Psychometrika, 16, 297-334.

Cronbach, L.J. (1987). Statistical tests for moderator variables: Flaws in analysis
recently proposed. Psychological Bulletin, 102, 414-417.

Cusumano, A. M., and Nobeoka, K. (1992). Strategy, Structure and Performance in
Product Development : Observations from the Auto Industry. Research Policy, 21, 265-
293.

Damanpour, F. (1991). Organizational Innovation: A meta analysis of effects of
determinants and moderators. Academy of Management Journal, 34(3), 533-590.

Datar, S., Jordan, C., Kekre, S., Rajiv, S., and Srinivasan, K. (1996). New product
development structures: The effect of customer overload on post-concept time to market.
Journal of Product Innovation Management, I 3, 325-333.

De Brentani, U. (1989). Success and failures in new industrial services. Journal of
Product Innovation Management, 6(4), 239-258.

De Meyer, A. and Van Hooland, B. (1990). The contribution of manufacturing to
shortening design cycle times. R&D Management, 20(3), 229-239.

Duerr, M. G. (1986). The Commercial Development of New Products. New York:
Conference Board.

Dumaine, B. (1991). Earning more by moving faster. F ortune(0ctober 7), 89-90.

144

Eisenhardt, K. M., and Tabrizi, B., N. (1995). Accelerating Adaptive Processes: Product
Innovation in the Global Computer Industry. Administrative Science Quarterly, 40, 84-
110.

Emmanuelides, A. P. (1991). Determinants of product development time: A framework
for analysis. Academy of Management Best Paper Proceedings, 342-346.

Ettlie, J. E., and Reza, E. M. (1992). Organizational Integration and Process Innovation.
Academy of Management Journal, 35(4), 795-827.

Flynn,B.B., Sakakibara, s., Schroeder, R.G., Bates, K.A., and Flynn, 1.13., (1990).
Empirical research methods in operations management. Journal of Operations
Management, 9(2), 250-284.

Garvin, D. A. (1987). Competing on the eight dimensions of quality. Harvard Business
Review(November-December), 101 -1 O9.

Gatignon, H., and Xuereb, J .-M. (1997). Strategic orientation of the firm and new product
performance. Journal of Marketing Research, 34(February), 77-90.

Graves, S. B. (1989). Why costs increase when projects accelerate. Research Technology
Management, 32(2), 16-18.

Graves, S. B., Carmichael, W. P., Daetz, D., and Wilson, E. (1991). Improving the
product development process. Hewlett Packard Journal, 42(3 June), 71.

Griffin, A. (1992). Evaluating QFD's use in US firms as a process for developing
products. Journal of Product Innovation Management, 9, 171-187.

Griffin, A. (1993). Metrics for Measuring Product development Cycle Time. Journal of
Product Innovation Management, 10, 112-125.

Gupta, A. K., and Wilemon, D. L. (1990). Accelerating the development of technology-
based new products. California Management Review(Winter), 24-44.

Handfield, R. B. (1994). Effects of Concurrent Engineering on Make-to-Order Products.
IEEE Transactions on Engineering Management, 41 (4 (N ovember)), 384-393.

Handfield, R.B., and S. Ghosh. (1994). Creating a quality culture through organizational
change: A case analysis. Journal of International Marketing, 2(3), 7-36.

Hartley, J. L., Meredith, J. R., McCutcheon, D., and Kamath, R. R. (1997). Suppliers'
contributions to product development: An exploratory survey. IEEE Transactions on
Engineering Management, 44(3), 258-267.

145

Hartley, J. L., Zirger, B. J ., and Kamath, R. R. (1997). Managing the Buyer-Supplier
Interface for On-time Performance in Product Development. Journal of Operations
Management, 15(1), 57-70.

Hauptrnan, 0., and Hirji, K. K. (1996). The influence of process concurrency on project
outcomes in product development: An empirical study of cross-functional teams. IEEE
Transactions on Engineering Management, 43(2), 153-164.

Hauser, J. R., and Clausing, D. (1988). The House of Quality. Harvard Business
Review(May-June), 63-73.

Himmerfarb, P. ( 1 992). Survival of the F ittest: New Product Development During the
90's. Englewood, NJ: Prentice Hall.

Hise, R. T., O'Neal, L. 0., McNeal, J. U., and Parasuraman, A. (1989). The Effect of
Product Design Activities on Commercial Levels of New Industrial Products. Journal of
Product Innovation Management, 6, 43-50.

Hof, R. D. (1992). Suddenly, Hewlett Packard is doing everything right. Business
Week(March 23), 88-89.

House, C. H., and Price, R. L. (1991). The return map: Tracking product teams. Harvard
Business Review(January-February), 92-100.

Iansiti, M. (1995). Science-based product development: An empirical study of the
mainframe computer industry. Production and Operations Management Journal, 4(4
Fall), 335-359.

Imai, K., Ikujiro, N., and Takeuchi, H. (1985). Managing the new product development
process: How Japanese companies learn and unlearn. In R. H. Hayes, K. Clark, and A.
Lorenz (Eds), The Uneasy AlliancesManaging the Productivity- Technology Dilemma
(pp. 33 7-375). Boston: Harvard Business School Press.

Ittner, C. D., and Larcker, D. F. (1997). Product development cycle time and
organizational performance. Journal of Marketing Research, 34(February), 522-534.

Jaccard, J .J ., Turrisi, R., and Wan, CK. (1990). Interaction effects in multiple regression.
Newbury Park : Sage Publications.

Jaworski, B. J ., and Kohli, A. K. (1993). Market Orientation: Antecedents and
Consequences. Journal of Marketing, 5 7(July), 53-70.

146

Kahn, K. B., and McDonough, ER (1997) An empirical study of the relationships
among co-location, integration, performance, and satisfaction, Journal of Product

Innovation Management, 14(3, May), 161-178.

Kessler, E. H., and Chakrabarti, A. K. (1996). Innovation speed : A conceptual model of
context, antecedents, and outcomes. Academy of Management Review, 21(4), 1143-1191.

Kleinschmidt, E. J ., and Cooper, R. G. (1991). The impact of product innovativeness on
performance. Journal of Product Innovation Management, 8, 240-251.

Koufieros, X. (1995). Time-based competitionsDeveloping a nomological network of
constructs and instrument development. Unpublished Unpublished Ph.D. Dissertation,
University of Toledo, Toledo, OH.

Krishnan, V. (1996). Managing the simultaneous execution of coupled phases in
concurrent product development. IEEE Transactions on Engineering Management, 43(2
May), 210-217.

LaBahn, D. W., and Krapfel, R. (1994). Early supplier involvement in new product
development: A model of the suppliers' perspective (Research Report 12-1994).
University Park, PA: Pennsylvania State University. ‘

Lambert, Douglas M. and Harrington, Thomas C. (1990). Measuring Nonresponse Bias
in Customer Service Mail Surveys. Journal of Business Logistics, 11 (2), 5-25.

Larson, C. (1988). Team tactics can cut product development costs. Journal of Business
Strategy, 9(5 September-October), 22-25.

Lewis, G. (1986). Designing to reduce assembly costs. Plastics Design Forum, 11(1
January-February), 54-60.

Liker, J. K., Sobek, D. K., Ward, A. C., and Cristiano, J. J. (1996). Involving suppliers in
product development in the United States and Japan: Evidence for set-based concurrent
engineering. IEEE Transactions on Engineering Management, 43(2 May), 165-178.

Lilien, G. L., and Yoon, E. (1989). Determinants of new industrial product performance:
A strategic reexamination of the empirical literature. IEEE Transactions on Engineering
Management, 36(1 (February)), 3-10.

Mansfield, E. (1988). The speed and cost of industrial innovation in Japan and the United
States: External and internal technology. Management Science, 34(October), 1157-1168.

Matsueda, KL. (1982). Testing control theory and differential association : A causal
modeling approach. American Sociological Review, 47, 489-504.

147

McCall, M.W., and Bobko, P. (1991). Research Methods in the Service of Discovery. In
Dunnette & Hough (Eds) Handbook of Industrial and Organization Psychology, Palo
Alto, CA: Consulting Psychologists Press.

Millson, M. R., Raj, S. P., and Wilemon, D. (1992). A survey of major approaches for
accelerating new product development. Journal of Product Innovation Management, 9,
53-69.

Moenaert , R.K., Souder, W.E., De Meyer, A. and Deschoolrneester, D. (1994). R&D -
Marketing integration mechanisms, communication flows, and innovation success,
Journal of Product Innovation Management, 11 (1 , January), 31-45.

Montoya-Weiss, M., and Calantone, R. (1994). Determinants of New Product
Performance: A review and meta-analysis. Journal of Product Innovation Management,
11, 397-417.

Myers, 8., and Marquis, D. G. (1969). Successful industrial innovations. (V 0]. NSF 69-
17). Washington DC: National Science Foundation.

Nakata, C., And Sivakumar, K . (1996), National culture and new product development :
An integrative review, Journal of Marketing, 60 (I , January), 61-72.

Nevens, R., and Whitney, D. (1989). Concurrent Design of Products and Processes. New
York: McGraw Hill.

Nijssen, E. J ., Arbouw, A. R. L., and Commaneur, H. R. (1995). Accelerating new
product development: A preliminary empirical test of a hierarchy of implementation.
Journal of Product Innovation Management, 12, 99-109.

Nunnally, J .C. (1978). Psychometric Theory. New York: McGraw-Hill 'Book Company.
Ragatz, G. L., Handfield, R. L., and Scannell, T. V. (1997). Success factors for
integrating suppliers into new product development. Journal of Product Innovation
Management, 14(3), 190-202.

Raia, E. (1991). Taking time out of product design. Purchasing, [10(6 April 4), 36-39.

Rochford, L., and Rudelius, W. (1992). How involving more functional areas within a
firm affects the product process. Journal of Product Innovation Process, 9, 287-299.

Rosenau, M. D. (1990). Faster New Product Development: Getting the Right Product to
the Market Quickly. New York: AMACOM.

148

Rosenau, M. D., Griffin, A., Castellion, G. A., and Anschuetz, N. F. (1996). The PDMA
Handbook of New Product Development. New York: John Wiley and Sons.

Rosenthal, S. R., and Tatikonda, M. V. (1993). Time management in new product
development: Case findings. Engineering Management Review(F all), 13-20.

Ruekert, R.W., (1992). Developing a market orientation: An organizational strategy
perspective. International Journal of Research in Marketing, 9, 225-245.

Saxenian, A. L. (1991). The origins and dynamics of production networks in Silicon
Valley. Research Policy, 20, 423-437.

Shrivastava, P., and Souder, W. E. (1987). The strategic management of technological
innovations: A review and a model. Journal of Management Studies, 24(1 January), 25-
41.

Smith, P. G., and Reinersten, D. G. (1991). Developing Product in Half the Time. New
York: Van Nostrand Reinhold.

Song, M. X., & Parry, M. (1996). Product innovativeness and New Product Success in
Japanese Firms (Working Paper). E.Lansing, MI: Michigan State University.

Song, M. X., & Montoya-Weiss, M. M. (1995). A contingency model of the Japanese new
product development process: The moderating effect of technological uncertainty
(Working Paper). E.Lansing, MI: Michigan State University.

Sullivan, L. P. (1986). Quality Function Deployment. Quality Progress, 19(6 June), 39-
50.

Swink, M. L., Sandvik, C. J ., and Mabert, V. A. (1996). Customizing concurrent
engineering processes: Five case studies. Journal of Product Innovation Management, 13,
229-244.

Takeuchi, H., and Nonaka, I. (1986). The new new product development game. Harvard
Business Review, 64(1), 137-146.

Tatikonda, M. V. (1995). Technology planning and implementation in product
development projects : An empirical study of innovation (we, organization and

performance. Unpublished Ph.D. Dissertation, Boston University, Boston.

Teece, D. J. (1988). Capturing value from technological innovationzlntegration, strategic
planning and licensing decisions. Interfaces, 18(3), 46-61.

149

Teresko, J. (1991). Hewlett-Packard keeps reinventing itself. Industry Week, 240(16
August 19), 44-52.

Trygg, L. (1993). Concurrent engineering practices in selected Swedish companies: A
Movement or an activity of the few? Journal of Product Innovation Management, 10,
403-415.

Ulrich, K., Sartorius, D., Pearson, S., and Jakiela, M. (1993). Including the value of time
in design for manufacturing decision making. Management Science, 39(4), 429-447.

Urban, G. L., and Hauser, J. R. (1993). Design and Marketing of New Products.
Englewoods Cliffs, NJ: Prentice-Hall.

Valentino, D., and Christ, W. (1989). Teaming up for market : Cheaper, better, faster.
Management Review, 78(11 November), 46-49. -

von Braun, C.-F. (1990). The acceleration trap. Sloan Management Review(F all), 49-58.

Welter, T. R. (1990). How to build and operate a product design team. Industry
Week(April 16), 35-58.

Wheelwright, S. C. (1988). Product development and manufacturing set-up. In M.
Tushman and W. Moore (Eds), Readings in the Management of Innovation (V 01. 2nd
edition, pp. 444-453). Cambridge, MA: Ballinger.

Wheelwright, S. C., and Clark, K. B. (1992). Revolutionizing Product Development. New
York: Free Press.

Wind, J ., Mahajan, V., and Bayless, J. L. (1990). The role of new product models in
supporting and improving the new product development process:Some preliminary
results. Cambridge, MA: The Marketing Science Institute.

Winner, R. I., Pennell, J. P., Bertrand, H. E., and Slusarczuk, M. M. G. (1988). The Role
of Concurrent Engineering in Weapons Systems Acquisition. (V 01. IDA Report R-338).
Alexandra, VI: Institute of Defense Analysis.

Zangwill, W. I. (1993). Lightning strategies for innovation .' How the world's best firms
create new products. New York: Lexington Books.

Zipser, A. (1992). Chipping away at Intel - the super semiconductor maker loses its
monopoly. Barrons, 72(12 march 23), 16-17.

Zirger, B. J ., and Hartley, J. (1994). A conceptual model of product development cycle
time. Journal of Engineering and Technology Management, I 1, 229-251.

150

Zirger, B. J ., and Hartley, J. L. (1996). The effect of acceleration techniques on product
development. IEEE Transactions on Engineering Management, 43 (May), 143-152.

151

"‘111111111111111“