Internal and External Integration for Product Development: The Contingency Effects of Uncertainty,...

June, 2015

Paper’s summary

Internal and External Integration for Product Development: The Contingency Effects of Uncertainty, Equivocality, and Platform Strategy

Based on the paper “Koufteros, X., Vonderembse, M., Jayaram, J., 2005, Internal and External Integration for Product Development: The Contingency Effects of Uncertainty, Equivocality, and Platform Strategy, Decision Sciences, Vol. 36 No. 1 pp 97-133”

Leonardo Laranjeira GomesPhD StudentMIT-Zaragoza International Logistics ProgramZaragoza Logistics Center

Page - 2 -

Overview: a contingent SEM approach on the relationships between NPD integration, competitive capabilities, and profitability

Highlights• The relationship between internal integration, external integration, and competitive capabilities

• Internal integration: concurrent engineering practices• External integration: customer integration and supplier integration• Competitive capabilities: product innovation and quality performances

• The contingency effects of:• Uncertainty and equivocality in the environment• Platform development strategy

• The approach:• Survey with 224 manufacturing firms• Structural equation modeling (SEM)• Test for moderation

• Key findings:• NPD integration positively influences product innovation and quality performances (and

ultimately profitability)• Equivocality moderates these relationships

Source: Koufteros et al., 2005, Leo Gomes’ elaboration

Page - 3 -

Contents

1. Introduction

2. Theory development

3. Methods

4. Results

5. Conclusions, contributions and discussion

6. Personal opinion

Page - 4 -

Contents

1. Introduction


3. Methods

4. Results


6. Personal opinion

Page - 5 -

NPD is a key strategic activity, yet it often fails; moreover, findings from previous research in the field are paradoxical...

Previous literature on new product development (NPD) vs. performance


A consistent finding: the lack of a robust theory that explains the patters in the findings

Author(s) Year(s) Topic General Idea

CooperCooper and Edgett

19902003

NPD failure

• Average failure rate: 35-40%- Top performers: 20.5%

Montoya-Weiss & CalantoneBrown & EisenhardtKessler & Chakarabarti

199419951996

Meta-analytic

NPD

• Interdisciplinary studies• Paradoxical findings on NPD performance

- Relationships and impacts

Henard & Szymanzki 2001Directions for future research

• Call for the seek for the key drivers of NPD performance

• Suggest for examining the contingency effects

Gerwin & Barrowman 2002Integrated

NPD• Points out contingency variables that could

explain the paradoxical findings

Page - 6 -

… this research seeks to address this gap by offering a contingency view of the effects of integration in NPD performance

Research objectives:• Test how integration strategy influences performance in the context of NPD• Evaluate the moderating impact of uncertainty, equivocality, and platform strategies

Research questions:• Does a high level of internal integration lead to a higher level of external integration?• Do certain contextual variables moderate the linkages between integration and performance?

Key variables:• Integration:

- Internal integration: concurrent engineering practices- External integration: customer integration and supplier (product and process) integration

• Performance:- Competitive capabilities: product innovation and quality performances- Profitability, as an ultimate goal

• Contextual variables:• Uncertainty and equivocality in the environment• Platform development strategy


Page - 7 -

Contents

1. Introduction


3. Methods

4. Results


6. Personal opinion

Page - 8 -

Successful firms structure cross-functional and/or boundary spanning teams early in the product development efforts…


Literature on NPD integration and uncertainty

Author(s) Year(s) Topic General Idea

Liker et al.Hartley et al.

19961997

Integrated NPD

• Successful firms involve important constituents early in the product development effort

Droge et al. 2000NPD

automotive• Cross- functional and boundary spanning improves

time-to-market and performance

Burns & StalkerLawrence & LorchHuber, et al.Huber & Daft

1961196719751987

Adaptation to

uncertainty

• Environmental uncertainty and ambiguity have an important influence in structuring an organization

Daft and Lengel 1986Uncertainty

reduction

• To reduce uncertainty, the organization needs to process more information

• Information processing involves cross-functional activities

Gupta et al. 1986Uncertainty

NPD• Uncertainty increases the need of interconnected NPD

…and the importance of integration seems to be increasing on uncertainty

Page - 9 -

The study posits internal integration as a precursor of external integration, which is expected to influence competitive capabilities…


… and suggests that there may be contingency variables that affect these relationships

• Concurrent engineering represents internal integration (Langowitz, 1988; Barkan, 1992; Millson et al., 1992)

• Internal integration facilitates external integration

• External integration can impact innovation speed and frequency by facilitating coordination with boundary groups (Parthasarthy and Hammond, 2002)

• Supplier integration involves two separate constructs: product and process integration

• Contingency variables may affect these relationships (Galbraith, 1973)

Hypothesized structure model (adapted from the authors)The view behind the model

Concurrent eng.

Customer integration

Product integration

Process integration

Product innovation

Profitab.

Quality

Supplier

Internal integration

External integration

Competitive capabilities

1

2

3

4

5

Contingencies

• Uncertainty and Equivocality

• Platform strategy 7

6

Page - 10 -

Contents

1. Introduction


3. Methods• Research design and sample characteristics• Measurement and structural model methods

4. Results


6. Personal opinion

Page - 11 -

The development of the survey measurement model involved multiple rounds of interviews with practitioners and a pre-test

Steps on the development of the measurement model

1. Structured interviews with 10 practitioners for the generation of the construct’s items

2. Evaluation of the items by 14 practitioners (engineering and PD) and faculty from 3 universities (manufacturing management, engineering, marketing, and IT)

3. Preliminary assessment test of measurement properties with 34 firms• All Cronbach’s alpha > 0.8• EFA for testing unidimensionality

4. Development of survey items (32 items, mixed through the instrument):• Concurrent engineering: 10 items• Customer integration: 4 items• Supplier product integration: 3 items• Supplier process integration: 3 items• Product innovation: 4 items• Quality: 7 items• Profitability: 1 item


Five-point Likert-scale measuring the extent to which a practice is used by the firm

(1= not at all,...,5=a great deal)

Seven-point Likert-scale comparing the firm with the average of the industry

(1= much below,...,7=much above)

Page - 12 -

List of survey itemsLatent variables Items

Concurrent engineering

X1. Much of process design is done concurrently with product designX2. Product development activities are concurrentX3. Product development group members share informationX4. Product development group members trust each otherX5. Product development employees work as a teamX6. Product development group members seek integrative solutionsX7. Purchasing managers are involved from the early stages of product developmentX8. Process engineers are involved from the early stages of product developmentX9. Manufacturing is involved from the early stages of product developmentX10. Various disciplines are involved in product development from the early stages


X11. In developing the product concept, we listen to our customer needsX12. We visit our customers to discuss product development issuesX13. We study how our customers use our productsX14. Our product development people meet with customers

Supplier product integrationX15. Our suppliers do the product engineering of component parts for usX16. Our suppliers develop component parts for usX17. Our suppliers develop whole subassemblies for us

Supplier process integrationX18. Our suppliers are involved in the early stages of product developmentX19. We ask our suppliers for their input on the design of component partsX20. We make use of supplier expertise in the development of our products

Product innovation

X21. Our capability of developing unique features isX22. Our capability of developing new product and features isX23. Our capability of developing a number of new features isX24. Our capability of developing a number of new products is

Quality

X25. Our capability of offering products that function according to customer needs over a reasonable lifetime isX26. Our capability of offering a high value product to the customers isX27. Our capability of offering safe-to-use products that meet customer needs isX28. Our capability of offering reliable products that meet customer needs isX29. Our capability of offering durable products that meet customer needs isX30. Our capability of offering quality products that meet customer expectations isX31. Our capability of offering high performance products that meet customer needs is

Profitability X32. What is your profitability relative to the average in the industry

Source: adapted from Koufteros et al., 2005

Page - 13 -

Survey application and responses’ profile

The Society of Manufacturing Engineers (SME) supported the survey, which target 2,5k manufacturing execs. and obtained 10% response rate


• Mailing list and logistical support from SME• Executives from 2,500 discrete-part manufacturing firms

(>100 employees)• Four SIC codes (key segments in much of the reported

manufacturing research)• 2-weeks pre-notification (traditional mail, letter from SME)• 253 responses (244 usable): 10% response rate• Chi-square test shows good fit with population (p>.527)• Most companies < 500 employees: even CEOs would have

knowledge on NPD practices

Survey application

SIC 34: Fab-ricated metal

35%

SIC 35: Machinery30%

SIC 37: Transport.

equip.15%

SIC 36: Elec-tronics

12%

Misc.8%

By SIC code

VP31%

Manager19%

President/CEO12%

Director11%

Misc.27%

By position

100-49969%

500-99916%

1000+15%By company size

(employees)

Page - 14 -

The researchers tested the measurement model prior to the structural model. CFA and several fit measures were also employed

Fit tests employed for the measurement and structural model• Confirmatory Factor Analysis (CFA): maximum likelihood estimation on a covariance matrix,

using the entire set simultaneously• Convergent validity: significance of individual item loading though t tests• Overall fit: chi-square to degrees of freedom

• Misspecification analysis: completely standardized expected change in Λx (potential cross loading) with cut-off value 0.4

• Discriminant validity: comparing average variance extracted (AVE) with squared correlation between constructs (or confidence intervals)

• Other methods employed: • Nonnormed fit index (NNFI)• Comparative fit index (CFI)• Parsimony goodness-of-fit index (PGFI)• Standardized RMR• Root mean square error of approximation (RMSEA)


The hypothesized relationships was evaluated by the structural model. If a model fits the data, t-values of structural coefficients (γ and β) can be used to test the hypothesis

Page - 15 -

To assess the role of uncertainty, equivocality, and platform strategy, the study tested the model for moderator effects

Method employed for testing the model for moderator effects: Test measurement invariance, in terms of loadings matrices, then assess invariance for the path coefficients

1. Establish two groups for each of the three moderators: low and high uncertainty- Mean score on the respective moderator scale

2. Test for measurement and structural coefficient invariance- Baseline model with two groups (model 1)

- Impose equality constraints on both the λx and λy matrices (model 2)

- Impose equality constraints on θε and θδ (model 3)- Impose equality constraints on γ and β (model 4)

Evaluation• Non-significant difference in chi-square between models can indicate that the structural models

do not differ• If there is a significant difference in chi-square, then a search for identifying which particular

coefficients differ takes place

Note: measures of uncertainty, equivocality, and platform strategy appear to be unidimensional and have satisfactory Chonbach’s alpha


Page - 16 -

Contents

1. Introduction


3. Methods

4. Results• Measurement model• Structural model• Contingency effects


6. Personal opinion

Page - 17 -

The measurement model appears to be supported by various fit indices, which supports the study to move on to the structural model and HT

Results for the fit indices on the measurement model• Chi-square: 802.75 (444 df)• Chi-square / df = 1.81• CFI: 0.94• NNFI: 0.93• PGFI: 0.70• PNFI: 0.79• Standardized RMR: 0.48• RMSEA: 0.58• All of the items have significant relationships with their factors: all factor loading above 0.63

and most above 0.8• t-values significant at 0.001

• None of the completely standardized expected changes in Λx were greater than 0.40

• Composite reliabilities and AVE estimates for each construct exceed customary acceptable levels


Page - 18 -

Completely standardized loading and t-values of survey items (n=244)Latent

variables Items Completely std. loadings t-values

Concurrent engineering

X1. Much of process design is done concurrently with product design .82 –∗ ∗X2. Product development activities are concurrent .89 17.57X3. Product development group members share information .72 12.71X4. Product development group members trust each other .74 13.28X5. Product development employees work as a team .87 17.00X6. Product development group members seek integrative solutions .83 15.73X7. Purchasing managers are involved from the early stages of product development .67 11.58X8. Process engineers are involved from the early stages of product development .72 12.82X9. Manufacturing is involved from the early stages of product development .86 16.66X10. Various disciplines are involved in product development from the early stages .82 15.55


X11. In developing the product concept, we listen to our customer needs .82 –∗ ∗X12. We visit our customers to discuss product development issues .78 13.62X13. We study how our customers use our products .78 13.62X14. Our product development people meet with customers .81 14.43

Supplier product

integration

X15. Our suppliers do the product engineering of component parts for us .82 –∗X16. Our suppliers develop component parts for us .88 14.90X17. Our suppliers develop whole subassemblies for us .68 11.09

Supplier process

integration

X18. Our suppliers are involved in the early stages of product development .85 –∗X19. We ask our suppliers for their input on the design of component parts .87 17.24X20. We make use of supplier expertise in the development of our products .84 16.16

Product innovation

X21. Our capability of developing unique features is .67 -X22. Our capability of developing new product and features is .82 11.33X23. Our capability of developing a number of new features is .91 12.24X24. Our capability of developing a number of new products is .84 11.58

Quality

X25. Our capability of offering products that function according to customer needs over a reasonable lifetime is .63 –∗X26. Our capability of offering a high value product to the customers is .71 9.44X27. Our capability of offering safe-to-use products that meet customer needs is .79 10.30X28. Our capability of offering reliable products that meet customer needs is .87 11.00X29. Our capability of offering durable products that meet customer needs is .81 10.50X30. Our capability of offering quality products that meet customer expectations is .85 10.82X31. Our capability of offering high performance products that meet customer needs is .86 10.93

Profitability X32. What is your profitability relative to the average in the industry 1.0 –∗

Note: * indicates a parameter fixed at 1 in the original solution. Fit indices: Chi-square = 802.75 (p=.00), 444 df, chi-square/df = 1.81, NNFI=.93, CFI=.94, PGFI=.70, PNFI=.79, std. RMR=.048, RMSEA=.058Source: adapted from Koufteros et al., 2005

Page - 19 -

Descriptive statistics, correlations, composite reliability, and discriminant validity (n = 244)

Source: Koufteros et al., 2005

Page - 20 -

Internal integration positively affects external integration and, indirectly, profitability; yet, questions arise from non-signif. and neg. relationships

Source: Koufteros et al., 2005; Leo Gomes’ elaboration

Negative relationships suggest that assigning product development responsibilities to suppliers may affect the firm’s ability to innovate

• Concurrent engineering is positively associated with external integration, but is weaker on prod. Integration

• Significant impacts of customer integration and supplier product integration on product innovation

• Product integration leads to process integration

• Product innovation impacts profitability, mediated by quality

• Supplier product integration negatively impacts innovation and, ultimately, profits

Hypothesized structural model results (adapted from the authors) Comments

Concurrent eng.


Supplier product

integration

Supplier process

integration

Product innovation

Profitab.

Quality


External integration Competitive capabilities

H1,t=12.1

H2, t=6.5

H3, t=8.21

H4, t=4.43

H5, t=1.94

H7, t=-2.36

H8, t=1.31

H9, t=1.44

H10, t=-1.48

H11, t=7.21

H12, t=4.86

H6, t=9.49

Page - 21 -

Summary table of structural model across environments (standardized coefficients)


Page - 22 -

Contingency effects: uncertainty does not play a significant moderating role on the model hypothesized…

Invariance tests across uncertainty environments: hypothesis 13

Analysis of Δchi-square to Δdf:• Model 2: no significant difference on chi-square. This implies loadings equivalence• Model 3: significant difference. Errors are not invariant between high and low uncertainty• Model 4: no significant difference in the overall model (including path coefficients)


…therefore, hypothesis 13 was rejected

Page - 23 -

Contingency effects: equivocality plays a moderating role on some of the constructs in the model hypothesized



Page - 24 -

Graphical representation of the moderating effects of equivocality


High equivocalityLow equivocality

Concurrent eng.


Product integration

Process integration

Product innovation

Profitab.

Quality




Concurrent eng.


Product integration

Process integration

Product innovation

Profitab.

Quality




t=9.37

5.45

t=5.44

t=1.57

t=1.65

t=-2.57

t=-0.9

t=2.32t=0.48

t=5.85

t=3.96

t=10.07

3.44

t=6.79

t=4.88

t=1.93

t=-0.6

t=3.31

t=-.08t=-2.95

t=4.84

t=2.89

t=7.76 t=6.7

Impact of supplier

integration

Equivocality level

Low High

Product

Process

I Q

I Q

Q

Q

I

I

Page - 25 -

Contingency effects: platform strategy does not play a significant moderating role on the model hypothesized…


Analysis of Δchi-square to Δdf:• Model 2: significant difference on chi-square. This implies that loadings are not equivalent

between the two groups• Model 3: significant difference. Errors terms are different between the two groups• Model 4: no significant difference in the overall model (including path coefficients)


…therefore, hypothesis 15 is not supported

Page - 26 -

Contents

1. Introduction


3. Methods

4. Results


6. Personal opinion

Page - 27 -

Conclusion: firms should pursue product integration in high equivocality environments and process integration in low equivocality environments

• Internal integration is an important enabler of external integration (customer and supplier)- External integration probably would not be realized in the absence of internal integration

• Internal integration also leads to higher levels of competitive capabilities• Significant indirect effects of internal integration on innovation (t=5.35) and quality (t=4.63) • Customer integration appears to be vital for product innovation, especially in high equivocality• Contrary to expectations, customer integration is not directly associated with quality (but it is

mediated by innovation)• The effects of suppliers integration on competitive capabilities are mixed:

- Supplier integration may have a negative impact on product development, particularly in uncertain environments (Eisenhardt and Tabrizi’s, 1995)

- The use of outside suppliers increases the organizational complexity of coordinating design decisions (Liker et al., 1996)

- There might be regional differences. “Japanese companies have a comparatively long history of assigning greater responsibility for product development” (Ibid)

• While equivocality may be important in understanding the relationship between integration and performance, uncertainty and platform strategy does not appear to moderate the relationships


Page - 28 -

The research also contributes to literature and methodology. Its major limitations lie on the sample source and risk of common method bias

Contributions:

1. Includes internal and external integration variables in the same study (a rare feature)- How internal and external integration strategies affect each other as well as performance

2. Includes both customer and supplier integration as a manifestation of external integration- Previous literature focused on either supplier or customer

3. Examines the impact of contextual environmental variables as moderators

4. Employs SEM to accommodate moderators- Goes beyond the prototypical assessment of a measurement and structural model via SEM

Limitations:• Single respondents per company (but uses the ‘most knowledgeable’ respondent)• Uses perceptual measures for both endogenous and exogenous variables (by the same

individual) can lead to common method bias


Page - 29 -

Future research: include other variables and moderators

• Include other variables: - The employment of IT as a tool for NPD activities (communication and integration)- Assess the level of proficiency with integration practices are carried out

• Include other moderators:- Small firms: more informal approaches and practices in their NPD efforts?- The employment of IT as a tool for NPD activities (e.g. communication, integration, virtual

prototyping)- Assess the level of proficiency with integration practices are carried out- Life cycle and country of origin effects (e.g. US vs Europe vs Japan)


Page - 30 -

Contents

1. Introduction


3. Methods

4. Results


6. Personal opinion

Page - 31 -

Personal opinion

• Very insightful study

• Well structured research method:

• Well constructed a priori theory

• Combination of internal and external, customer and supplier integration

• SEM with moderation

• However:

• Survey items seems subjective

• Sample: 27% of respondents from “miscellaneous” levels

• Only members of the Society of Manufacturing Engineers

• Suggestion of extension:

• Combine survey with secondary data (e.g. profitability)

• Multi-continent international project, moderating for countries

• Control for company sizes and position in the supply chain


Page - 32 -

Backup slides

Page - 33 -

Internal integration: concurrent engineering

• Early involvement of a cross functional team in a process to plan product design, process design, and manufacturing activities simultaneously

• It may afford a firm a stream of integrative innovations that improve the value of products to customers, enhance quality, shorten time-to-market, and reduce cost

• With early release of information, engineers can begin working on different phases of the projet• Early detection of problems• Enacts a shared team vision and improves product development success• The logic for internal integration is equally relevant for integrating activities with external entities• Internal integration may affect external integration• Hypotheses:

- H1: concurrent engineering is positively associated with customer integration- H2: concurrent engineering is positively associated with supplier product integration- H3: concurrent engineering is positively associated with supplier process integration


1

Page - 34 -


• Involves determining customer requirements and tailoring internal activities to meet these requirements

• Customer have a vested interest in product development• Ensures that customers’ voice will be heard and their recommendations and suggestions are

incorporated in the design of new products• Useful to customers for planning purposes• Hypotheses:

- H4: customer integration is positively associated with product innovation performance- H5: customer integration is positively associated with product quality performance


2

Page - 35 -

Supplier integration

• Lead suppliers to operate as strategic collaborators• Black box: suppliers carry out product engineering activities on behalf of their customers and

even develop components of entire subassemblies. This is called supplier product integration• Gray box: supplier’s engineers work alongside the customer’s engineers to jointly design the

product so the supplier’s process can be effectively integrated with the design. This is called supplier process integration

• Hypotheses:- H6: supplier product integration is positively associates with supplier process integration- H7: supplier product integration is positively associates with product innovation

performance- H8: supplier product integration is positively associates with product quality performance- H9: supplier process integration is positively associates with product innovation

performance- H10: supplier process integration is positively associates with product quality performance


3

Page - 36 -

Product innovation

• The capability of organizations to introduce new products and features• Fosters organizational learning and enables time-to-market to be shortened even further• Because of frequent innovation, products closely match current customer demands• Design or quality deficiency can be overcome quickly, resulting in more satisfied customers• Hypothesis:

- H11: product innovation is positively associated with quality performance


4

Page - 37 -

Quality

• Quality from the customers’ perspective• The capability of the firm to design and produce products that would fulfill customer expectations• Quality is posited to affect profitability, which is used as a firm performance measure• Hypothesis:

- H12: quality performance is positively associated with profitability


5

Page - 38 -

Impact of uncertainty and equivocality on the adoption of integrated product development practices

Uncertainty• Rapid change in the external environment promotes uncertainty in product development• The presence of environmental hostility magnifies the positive influence of NPD activities on new

product success (Cantalone et al. 1997)

Equivocality• The presence of multiple and conflicting interpretations about a phenomenon• Confusion stems from the presence of complexity• Firms that can cope with higher levels of equivocality are in a sense creating structural

mechanisms internally and externally to provide consistency in interpretations

Hypothesis• H13: uncertainty in the environment will have a differential impact on the path coefficients of the

model• H14: equivocality in the environment will have a differential impact on the path coefficients of the

model


6

Page - 39 -

Impact of platform strategy on the structural relationships

• A dominant product design that forms the basis for all future extensions of the product within the same product family

• Leads to efficiency, lower costs, higher quality, and faster time-to market• Planning multiple generations• Product changes can be made quickly because technical and marketing uncertainties are lower• Technical learning that is transferable to new design cycles• Reduce typical start-up uncertainty and confusion• Hypothesis:

- H15: the use of platform strategies will have a differential impact on the path coefficients of the model


7

Page - 40 -

Measures of uncertainty, equivocality, and platform strategy(1/2)


Page - 41 -

Measures of uncertainty, equivocality, and platform strategy(2/2)


Internal and External Integration for Product Development: The Contingency Effects of Uncertainty,...

Business

Transcript of Internal and External Integration for Product Development: The Contingency Effects of Uncertainty,...