Introduction Manufacturing capability Integrated Problem Solving LeadershipPM Organization

Introduction 기업 내부 역량 § Manufacturing capability § Integrated Problem Solving § Leadership(PM) & Organization § Flexible Development Process § Development productivity : overlapping, § Lead time § Product Quality § Cost § Etc : reward, CAD, … NPD 성과 기업 외부와의 NW § Supplier Involvement § Market Feedback 2

Product development performance : strategy, organization, and management in the world auto industry Clark, K. B. and T. Fujimoto (1991).

1. Manufacturing Capability • Elements of successful NPD : Focus on Design process, Solution of technical problems : Manufacturing capability(ability to make things rapidly & efficiently) is a critical sources of advantage in product development • Manufacturing Productivity & Development Productivity : positive correlation 4

1. Manufacturing Capability • Hidden manufacturing activities in the development process • 1. Prototype Fabrication Short (6 mo) Test engineers can begin prototype evaluation earlier Reasonably high More engineering changes resulting from tests can be made in time Mostly production suppliers Prototype as problem detector Reduce the negative impact of late engineering change 5

1. Manufacturing Capability • Hidden manufacturing activities in the development process • 2. Die development 1) Lead time(Performance of die development) JIT Philosophy : Japanese firms have less WIP dies than US Supplier Network -US : arm’s-length contract with separate companies 2) Cost(Performance of die development) - Japan : at most 20 % of the cost of a die - US: 30~50 % of the cost of a die -Japan : close & long term relationship with major suppliers 6

1. Manufacturing Capability • Hidden manufacturing activities in the development process • 3. Pilot Run & Production Start-up 3. 1. Pilot run - A full-scale rehearsal of the commercial production system - Objective: to find and solve problems undetected during the prototype production & testing Japan U. S. & Europe Problem solving cycle Fast Slow Location of pilot production Unseparated (empty hanger approach) Separated ü Reducing the opportunity cost for the pilot run by minimizing lost production. ü Future workers can get early training in a very realistic setting. ü Workers tend to get excited about a new model they see on the line. ü Workers become motivated to learn more about a new model. 7

1. Manufacturing Capability • Hidden manufacturing activities in the development process • 3. Pilot Run & Production Start-up 3. 2. Production Start-up(Ramp-up in assembly) - The start-up of commercial production system - Objective: same as pilot run - Effective production ramp-up depends on manufacturing capability and its fit with choice of ramp-up curve ü Complexity ↓ ü Task Continuity ↓ ü Stable operating condition ↓ ü Stable operating ∵ manufacturing capability condition ↓ U. S. & Europe (process control) Japan 8

2. Integrating Problem Solving Cycle • Successful product development can be achieved by : Effective integration of different functional group and engineering discipline - cross functional team, simultaneous engineering • 5 Dimensions of Integrated Problem Solving Limited communication Reduce lead time Rich, dense, early communication 9

2. Integrating Problem Solving Cycle • Successful integration problem solving can be achieved by : Stage overlapping (high degree of simultaneous activity) : Intensive Communication (rich, frequent, bidirectional information flow) 10

2. Integrating Problem Solving Cycle • Integrated Problem Solving & Development Performance 1. Integrated Problem Solving - “ideal profile index” composed by 9 characteristics (ex: high simultaneity ratio, early feedback from manufacturing, etc) 2. Development Performance - lead time, development productivity, total product quality Japan Effect Positive correlation 11

3. Leadership and Organization • 조직의 형태와 리더십은 제품개발성과에 영향 미치는 중요한 요소임 • 3 Dimensions of Product Development Organizational dimension Specialized Expected function • Accumulate and preserve technological expertise at the level of individual components and activities • execute individual tasks with speed and efficiency Internal Integration External Integration • Achieve high internal integrity of total product • Achieve fast product development through better task coordination • Achieve high external integrity of total product • Match product concept, product design, and customer expectation 12

3. Leadership and Organization • 4 Modes of Development Organization • Broader responsibility and clout. : internal/external coordination : product planning & concept development 13

3. Leadership and Organization • Patterns of organization and leadership - Measure • Specialization: Number of people who participate in a project on a long-term basis • Internal Integration: Engineering coordination indicator (based on PM role) • External integration: Concept development & Market link indicator (Based on PM role) • Performance: Development Productivity, Adjusted lead time, TPQ Index Specialization U. S Internal Integration Japan External Integration Japan 14

3. Leadership and Organization • Organization & Performance ü Shaded diagram refer to high significant of relationships(. 03) ü External integration, in contrast, is closely associated with total product quality but not with lead time or productivity 15

3. Leadership and Organization • Organization & Performance ü Lower performers generally lack integration. ü Some Japanese firms employing the middle weight system are both fast and efficient. ü The very different circumstances of high-end specialists were acknowledged earlier, but even volume producers occasionally spawn light weight pm achieving solid results in a single dimension of performance(TQM). 16

Discussion Questions 1. Dimension of integrated problem solving에서 우측으로 갈수록 lead time을 줄일 수 있 는 좋은 problem solving form이라고 하였다. 이에 대한 반대 의견은? 2. 본 연구에서 제시한 5가지 dimension 외에 추가할 수 있는 dimension에는 무엇이 있을까 ? Limited communication Reduce lead time Rich, dense, early communication 17

Discussion Questions 3. 본 연구에서는 신제품 개발 조직 구조가 functional structure일 때보다 heavyweight product manager일 때, 더 좋은 성과를 가져온다고 하였다. Heavyweight product manager를 가져가는 조직 구조의 단점은 무엇이 있을까? 18

Developing product on “Internet Time” : The Anatomy of a Flexible Development Process Alan Mac. Cormack Roberto Verganti Marco Iansiti (2001)

1. Research Objective 1. Many literatures on the Effective Development process before this article 2. Previous authors have developed theoretical models which demonstrate the value of greater flexibility in a design process faced with uncertainty. 3. This study attempts to examine the underlying mechanisms through which firms directly influence the flexibility of their development processes 20

2. Models of the Product Development Process Figure 1 - A stage-Gate Model of Product Development (Traditional Water fall model) ü The product concept is defined and frozen prior to the start of detailed design, and the functionality this specifies in each module is completed prior to the start of system level test. ü The First challenge this model faces is that it assumes all information about potential design choices is known or can be discovered during concept development. ü Second challenge is that feedback on how the product performs as a system is not obtained until late in a project, when the functionality in each module has been fully developed 21

2. Models of the Product Development Process Figure 2 - A More Flexible Model of Product Development ü Recent studies have began to investigate more flexible models of development characterized by the overlapping of development stages (e. g. Krishnan et al. 1997). ü In this model, development becomes an “evolutionary” process of learning and adaptation (Tushman and O’Reilly 1997). ü This model commonly used for software development. ü These models have been proposed to address flaws in the traditional waterfall model of development. 22

3. Hypotheses ü These two objectives (i. e. maximizing product performance and facilitating process flexibility) are often incompatible from a design standpoint; hence, the selection of the “optimal” product architecture becomes a more complex problem (Ulrich 1995). Hypothesis 1. In uncertain and dynamic environments, greater investments in architectural design will be associated with better performing projects. ü Where information about the user environment is tacit or “sticky” (i. e. , not easily captured by traditional market research techniques) there is value in mechanisms which facilitate the release of early product versions to users when a project retains the ability to change the design. Hypothesis 2. In uncertain and dynamic environments, earlier feedback on a product’s system-level performance will be associated with better performing projects. ü Lower-level learning results in knowledge that can be directly applied to a specific context, whereas higher-level learning results in a deeper knowledge of the process of problem solving, especially with respect to analyzing new frames of reference. Hypothesis 3. In uncertain and dynamic environments, development teams with greater amounts of generational experience will be associated with better-performing projects. 23

4. Measures • Sample: 29 completed projects(unit of analysis) from 17 firms in internet software industry. • Performance(Dependent variable): Product Quality. Overall quality was defined as a combination of product features. Use the assessments of a panel of experts gathered using a two-round Delphi process. • Control variable: Resources allocated to each project to control for the potential impact of resources on product quality. • Development Process(Independent variables) • Investments in Architectural Design: The ratio of architecture design resources to development and test resources. • Early Market and Technical Feedback: Percentage of the product’s functionality that has been developed when each milestone(prototype during the concept development stage, system integration, the first beta release point) is reached. • Generational Experience: the proportion of team members with greater 24 than two generations of experience.

5. Results 25

Discussion Questions Successful Customer involvement Case LDD = Lego Digital Designer Design by customer. 신제품 개발의 혁신 동력을 다양한 User로 부터 효율적으로 확보 29

Project Scope and Project Performance : The effect of parts strategy and supplier involvement on product development Kim B. Clark(1989)

1. Introduction – The importance of product development has motivated significant attention to the determinants of performance by both practitioners and academics. – This study examines one aspect of project strategy, project scope and its effect on project performance(Lead time and project manhour). 31

1. Introduction • Project Scope – The extent to which a new product is based on unique parts developed in-house. – Unique parts have been selected, the firm may rely on a supplier for engineering work. • Two elements of scope – The choice of unique versus off-the-shelf parts – Choice of supplier involvement – Once unique parts have been selected the firm may rely on a supplier for engineering work, reducing internal engineering efforts in the project. 32

2. Measure • Scope and Engineering Manhours – Decisions about scope have a direct impact on observed manhours in the project NH = 1 - b[C+S(1 -C)] ü C: Fraction of parts that are off-the-shelf ü S: Fraction of engineering effort for unique parts done by suppliers ü b: Fraction of total engineering effort that is part specific NH = 1 – (b*C) – [b*s(1 -C)] ü b*c: Fraction of off-the-shelf parts ü b*s(1 -C): Fraction of unique parts developed by supplier 33

3. Results Summary data on performance and content by region 34

3. Results • Scope and Engineering Manhours (regression analysis) • Adding measures of project scope has a dramatic effect on the regression. • Interaction between price and NH imply that the impact of scope on manhours depends on the complexity of the product. • Japanese dummy excluded from the regression: the coefficient on the supplier variable increases by a factor of 2. 7 quality of the relationship and the way that it is managed is important. 35

3. Measure • Scope and Lead time – Impact of scope on lead time is not straight forward. – Lead time: Time elapsed between start of the development project and market introduction. – Lead time is determined by the critical path in the network – Supplier involvement may reduce lead time if a supplier were more capable in executing parts engineering 36

3. Results Scope and Lead Time (regression analysis) • Strong positive impact of scope on lead time increase in scope from the Japanese level of 0. 57 to the U. S. level of 0. 66 would increase lead time by 3. 9 months • The effect of Interaction between price and NH is insignificant scope on lead time does not depend on the complexity of the product. • Japanese firms derive real advantages from their supply base. 37

3. Results • Who does the work, and how the content gets implemented (offthe-shelf vs unique parts), makes a difference in the length of the planning process. 38

Discussion Questions ü Executives, viewing manufacturing mainly as a cost center, give shot shift to the impact that outsourcing or offshoring it may have on a company’s capacity to innovate. ü Indeed, most don’t consider manufacturing to be part of a company’s innovation system at all. ü Massive migration of manufacturing from United States has seriously eroded the domestic capabilities needed to turn inventions into highquality, cost-competitive products, damaging America’s ability to retain a lead in many sectors. 40

41

Discussion Questions • The problems In managing supplier Involvement (FINN WYNSTRA et al. 2001) • The relation as a source of problems • The supplier as a source of problems • The manufacturer as a source of problems • Solution • Identifying specific processes and tasks that need to be carried out, aimed at the integration of product development and sourcing processes; • Forming an organization that supports the execution of such tasks; and, finally, • Staffing the organization with people that have 42

Accelerating Adaptive Processes : Product Innovation in the Global Computer Industry Kathleen M. Eisenhardt Behnam N. Tabrizi(1995)

1. Comparison of compression and experiential model Compression Model Experiential Model Key Assumption Predictable or Certainty Unpredictable or Uncertainty Image of product innovation Predictable series of well-defined steps Uncertain path through foggy & shifting markets & technologies Strategy for speed Rationalize & then squeeze the process Quickly build understanding & options while maintaining focus & motivation Bureaucratic Flexible Organization features Hypothesis (Independent Variables) Dependent Variables Data and Method Routine (H 1) (H 2) (H 3) (H 4) (H 5) (H 6) Planning Supplier Involvement Using the CAD Overlap Multifunctional team Greater Award Improvisational Experimental (H 7) Multiple Iteration (H 8) Extensive Testing (H 9) Frequent Milestones (H 10) Power Leader Development Time Survey and deep interview results from 72 projects(36 Computer Companies) 44

2. Compression model H 1: More time spent in planning is associated with shorter development time. H 2: More supplier involvement is associated with shorter development time. H 3: More designers using computer-aided design(CAD) is associated with shorter development time. H 4: A higher degree of project overlap is associated with shorter development time. H 5: Multifunctional teams are associated with shorter development time. H 6: Greater reward for schedule attainment is associated with shorter development time. 45

3. Experiential model H 7: More design iterations are associated with shorter development time. H 8: More time spent in testing throughout the development process is associated with shorter development time. H 9: Less time between milestones is associated with shorter development time. H 10: Greater power of the project leader is associated with shorter development time. 46

4. Methods - Data from 72 product development projects drawn from European, Asian, and U. S. computer firms. - Computer companies that compete in the personal computer(PC), minicomputer, mainframe, and peripherals segments of the industry. - Unit of analysis: product development project - Within each company, two product development projects within a single product group were used. 47

5. Result Opposite Compression Model Opposite Experiential Model Compression Model Experiential Model Compression + Experiential Parsimonious model 48

5. Result • Split-sample Analyses Certain products 49

5. Result • Split-sample Analyses Less Certain products 50

6. Discussion Question • 본 연구에서 제시되는 Independent variables들의 concept 이 올바르게 측정되었다고 생각하는가? construct Planning Reward for schedule attainment total reported time spent/total elapsed time Planning: ü Extensive planning simply waste time, especially high-velocity industry. Asking whether the product development personnel were rewarded for meeting the schedule deadlines during the course of the project Reward: ü reverse causality: slow firms may use reward for schedule as a solution to their pace problem. measure ü Neglect quality problem. 51

Compression Model-IV Construct Measure Supplier Involvement Numbers of stages that suppliers participated CAD(Computeraided design system) 1) Percentage of design engineers on the product development team who used CAD 2) Rating by respondents of the importance of increasing CAD usage to decreasing product development time Project overlap Sum or overlaps across the six processes of the product development/total development time Multifunctional team Adding the total number of functions in every phases (Whether personnel from specific functions were involved on the product development team during the various stages of product development) 52

Experiential Model-IV Construct Measure Number of design iterations - Asking the number of design iterations for a project. - Adjusted by the average number of iterations for each industry segmentation. Test Reported time spent on testing/total elapsed time Time between - Average time between milestones during the project milestones - Adjusted by the average time between milestones for each industry segmentation Power of the 1)Asking respondent group where in the SBU hierarch the project leader project manager reported(1, 0) 2)Assessed power by whether or not the project manager was the final decision maker on the key issues of budget, team composition, and project timetable. 53

Dependent and Control Variables Construct Measure Development -Duration: start and end time dates of the project -Development time: adjusted by average project duration of its corresponding industry segment Project size Dummy variable 0: medium-scale project 1: major project Development time: -start of a project: 1)the time at which the first meeting was held to consider the development of this specific product. 2)First assignment of firm personnel to assess the feasibility of the product. -ending date: the date at which product stabilization was reached. *stabilization: the time at which no more changes are made to the product Project size: Major projects would take longer to complete than medium-scale ones. 54

6. Discussion Question • 본 연구의 실증분석 방법에 대해 어떻게 생각하는 가? ü Compression/Experiential strategy’s assumption : industry, product characteristics, degree of maturity 55

6. Discussion Question • 본 연구에서 밝혀낸 IV(multifunctional team, multiple iterations, extensive testing, frequent milestones, powerful leader)들이 제품개발시간 을 줄이는 것을 방해하는 요소에는 어떤 것들이 있을까? 56

Woodman, Sawyer, and Griffin(1993) 57