Center for Innovation in Product Development
The Center for Innovation in Product Development (CIPD) links representatives from academia, industry, and government who share our dynamic vision of the future of product development: new products will be developed by just-in-time collaborations of globally-distributed teams linked seamlessly by web-based tools and processes. These collaborations will be formed by means of a services marketplace, where lead firms will find the world's best suppliers of information, components, and support services.
Our mission is to lay the conceptual groundwork for, and contribute core components to, a product development infrastructure that will help companies to thrive in the new services marketplace. Using industrial sites as our laboratories, we work with engineers and managers in product development environments to extend our fundamental understanding of the product development process, and provide innovative improvements to current practice.
Established in 1996 as one of the National Science Foundation's (NSF) Engineering Research Centers, CIPD is interdepartmental research program joining MIT's School of Engineering with the Sloan School of Management. Our 16 faculty and 18 graduate students pursue research that has been funded by the NSF and 10 industrial partners, with a budget in FY 2001–2002 of $2.2 million.
Since its inception, CIPD has executed over 173 research projects, and produced over 110 refereed journal articles, 81 conference papers, and 21 awards for outstanding research. Under the guidance of Co-Directors Professor Maurice Holmes and Professor Steven Eppinger, and Assistant Director Nils Nordal, CIPD continues to advance the theory and practice of product development with integrated programs of research, education, and outreach.
CIPD's seven interrelated research initiatives are:
- Virtual Customer (VC): Professor John Hauser, lead
- Distributed Object Modeling Environment (DOME): Professor David Wallace, lead
- Incentives and Boundaries (IB): Professor Rebecca Henderson, lead
- Implementation Dynamics (ID): Professor Nelson Repenning, lead
- Information Flow Modeling (IFM): Dr. Daniel Whitney, lead
- Platform Architectures (PA): Dr. Kevin Otto, lead
- Product Development Integration Lab (PDIL): Professor Maurice Holmes, lead
Successful product development (PD) demands good customer input. However, PD teams often are unable to get good customer input because of cumbersome, slow, expensive, or outdated collection methods. The Virtual Customer initiative seeks to improve customer input through a portfolio of web-based methods that allow PD teams to seamlessly incorporate customer preferences into their PD process. Because PD teams are increasingly distributed around the world, the virtual customer "engine" emphasizes web-based methods that facilitate globally-dispersed teams.
This past year, we made significant progress on our six virtual customer methods—web-based conjoint analysis, fast polyhedral adaptive conjoint analysis, user design, virtual concept testing, securities trading of concepts, and the information pump. These methods have been applied by our researchers at Xerox, Ford, and Polaroid (at Polaroid, they contributed to the successful redesign of the I-Zone camera).
Demonstrations and working papers are available at the Virtual Customer web site (mitsloan.mit.edu/vc). This site is a major practical output of the project, and includes demos of various virtual customer methods. It will also begin to post, over the next year, open-source code that will allow researchers and practitioners to download software applicable to market research.
Lastly, we have developed tools for User Design and for web-based conjoint analysis. These tools have been used in course 15.828 Design and Marketing of New Products, by student teams to quickly create virtual customer web sites. These tools will be refined and posted in open-source format on the VC web site.
This CIPD centerpiece is our largest initiative, and commercial interest in it has grown steadily over the past year. DOME offers a fundamentally new infrastructure for integrated modeling and simulation. Overcoming the extreme difficulties of large product simulations, DOME provides a "World-Wide Modeling Web" that allows all developers
to participate equally in simulation—even when they are separated by great distances, and are using different modeling tools and computing environments. DOME also secures proprietary knowledge, ensuring that each member of the design process has only the appropriate and necessary information.
Accomplishments for the DOME project this past year include: a patent (in progress) for a novel assembly process using DOME; journal publications in the American Society of Mechanical Engineers (ASME) Journal of Mechanical Design and the Journal of Industrial Ecology; and major new pilot studies initiated at Ford and Boeing. In addition, two commercial startup ventures (Oculus and Zionex) developed an independent interpretation of the published DOME concept, and are producing a commercially available product. Finally, the industry-driven DOME working group that began in the fall of 1999 has continued to meet regularly and successfully.
Our primary plan for the next year is to design and research the implementation of a next-generation DOME concept that will provide the foundation for a world-wide simulation web.
The explosion of information and information technologies has led many firms to evolve a dispersed product development process with people and organizations spread around the world. Incentives within an organization must therefore communicate the company's fundamental intentions not only to people within the organization, but to those beyond it. In the services marketplace that we envision, incentives must cross the boundaries of the organization and reach the virtual enterprises.
To coordinate such dispersed processes, managers must foster a culture that implicitly rewards and motivates product development teams to perform against a set of common strategic metrics. Those metrics may include customer satisfaction, time to market, defect reduction, and platform reuse. We focus on a practical method to fine-tune a firm's relative emphasis on these metrics. In particular, we seek to advise a firm how to modify their emphasis on each metric in order to improve profits and/or efficiency.
Journal articles detailing the results of the IB initiative are forthcoming in Organization Science, and have been submitted to Management Science. Over the next year, we will add greater emphasis on evaluation and industrial testing of our methods. We will also build a model for the US Navy that provides insight into how to best modify incentives for reduced lifetime costs of shipboard systems. Finally, we will integrate this initiative's results with other CIPD research. For example, the DOME project is now using the IB initiative's "3-T" model to frame boundary issues that exist in implementing DOME.
We are developing a methodology for estimating a given product development system's capacity. Complementing many of the advances in the theory of process design, this methodology will advance the theory of process improvement and implementation. Such a theory should explain high rates of failure, and should provide the basis for developing an implementation methodology that will prevent future failures.
During this past year, five of our research papers on PD implementation have been published (or accepted for publication) in distinguished academic journals. We have also presented our work widely at both academic and practitioner conferences.
Having recently completed a four-month study of a DOME implementation effort on-site at Ford Motor Company, we will write up our results in a paper that will be complete in the next year. Lastly, we are completing a three-year study of how development processes are used at Xerox, and will finish writing up our conclusions by the end of 2001.
We are developing advanced methods for the management of knowledge used in the engineering design process. These methods are based on the techniques of information flow modeling through the application of design structure matrices (DSMs). The resulting visual representations of development activities serve as maps for understanding and improving the PD process.
IFM research serves to help engineers and managers to achieve greater value in the development cycle of complex products. Over the coming year we will complete the development of decision support tools that will allow product developers to reduce risks in the PD process and optimize product performance (with specific application to jet engine design). This work has special significance to the DOME initiative, which began collaboration in the fall of 2000. Thesis work in this area will begin in the spring of 2002.
IFM research on spiral development aims to determine which conditions warrant the use of spiral product development processes rather than stage-gate processes. We laid the groundwork for our investigation in the past year by rigorously defining the spiral and stage-gate processes. In June 2001, we began applying mathematical models to spiral processes at Verizon, and will complete that work in September. Also in June we completed a case study on spiral applications in the computer hardware industry (ITT Industries). In July we will have completed case studies of Ford and Pratt & Whitney, and in August at IDE and Xerox. By the project's end in April 2002, we will have determined the selection criteria necessary to make an effective choice between spiral and stage-gate product development processes.
IFM research on system level knowledge strives to increase understanding about the capture and management of design information in complex products. Using DSM, we map out the exchange of knowledge during a project's early development. Our aim is to better convert top-level specifications into component specifications, and to insure that components in complex systems work harmoniously.
Collaborations between Ford, Pratt & Whitney, Otis Elevator, and Veeco-CVC have proven the value of our novel uses of DSM. In the summer of 2001 we will conduct a case study at Johnson and Johnson, and will conclude analysis in the fall. Deliverables will include a paper presented the ASME 13th International Conference on Design Theory and Methodology in September.
This initiative has focused on defining a theory of platform architecture, with an emphasis on the definition of successful product families. This theory would :
- Match customer needs with function;
- Match function with technology;
- Transform technology into a set of integrated design modules; and
- Screen and optimize these modules against a set of portfolio architecture principles.
We completed this research on May 31, 2001, and results have been disseminated throughout the Xerox Corporation, especially in the System Design and Integration (SDI) group. Our methodologies have been implemented as case studies that help engineers to more effectively define products and product architectures.
CIPD established the Product Development Integration Laboratory (PDIL)for several reasons:
- Foster intellectual ties with engineers and scientists from industry;
- Facilitate a two-way flow of ideas leading to participation in the Center's research and education programs; and
- Assist in the transfer of knowledge and technological advances to industry.
Equipped with $150,000 of state-of-the-art displays and video conferencing equipment, the PDIL can demonstrate the feasibility of web-based product development systems that integrate research from CIPD with commercial enterprises and other universities. In addition, the lab makes new technologies available for undergraduate, graduate, and professional education.
This past year, the PDIL supported efforts toward the development and application of a systems engineering submarine design methodology. In partnership with MIT's Ocean Engineering Department, this project investigated the worldwide web as a vehicle for including many collaborators in the submarine design process. We provided computer, network, and some CIPD student desk space; collaboration also included employees from Xerox.
Our current on-site work includes the deployment of DSM tools and methodologies at General Motors; a thesis drawn from this effort will be finished in the spring of 2002. We are also exploring the feasibility of deploying Virtual Customer for the Gould Pump Division of ITT Industries.
Finally, the Lab supported various projects in other departments at MIT: the Space Systems, Policy, and Analysis Research Consortium (SSPARC) program in Aeronautics and Astronautics, the Sharing FIRST (For Inspiration and Recognition of Science and Technology) program, and the Ocean Engineering Department all used our conference, computer, and video conference facilities.
Professor John Hauser, Kirin Professor of Marketing at the Sloan School of Management, and CIPD Virtual Customer research initiative leader, has been chosen as the next Charles Coolidge Parlin Award winner for distinguished achievement in marketing research. The yearly award was established in 1945 by the Philadelphia chapter of the American Marketing Association and the Wharton School in association with Curtis Publishing Company to honor Charles Coolidge Parlin, "the father of marketing research." The Parlin Award is the oldest and most distinguished award in the marketing research field, recognizing outstanding leadership and sustained influence on the evolving profession of marketing research.
Research at Xerox Palo Alto Research Center PARC
Research co-sponsored by CIPD at the Xerox Palo Alto Research Center (PARC) this year investigated knowledge sharing among product development teams. Two papers resulted, and were accepted for publication in the Journal of Engineering Education and the Proceedings of International Conference on Engineering Design (August 2001). Related research on the sharing of PD process knowledge among non-competitive firms resulted in various conference papers, and in one paper accepted for publication in the Journal of Knowledge and Process Management.
Plans for the coming year include an empirical analysis of process implementation based on case studies of three very different product teams: one using a radically new technology; one targeting a relatively new market; and one using the firm's traditional technology and targeting the firm's traditional markets.
CIPD is an interdisciplinary program between the School of Engineering and the School of Management. We believe that students' course experiences should address the interplay between the technical, social, and system elements of product development, and prepare them for work in the globally distributed services marketplace.
As an NSF Engineering Research Center, our educational mission has been to promote product development as part of core engineering curriculums both at MIT and throughout the United States. As noted in last year's Report to the President, the National Science Foundation elected not to extend funding to CIPD beyond January 2002, and we thus began the process of transitioning into a more independent research center. After carefully evaluating our strategy and programs, we decided to maximize our efforts in research to create more value for our industry sponsors. Our research has made us a vital ally to our current industry membership, and all nine of our current partners (Ford, General Motors, IBM, IDE, ITT Industries, Polaroid, Product Genesis, US Navy, Xerox Corporation) remain fully committed to CIPD. To better serve them, we have concluded ancillary education programs that, valuable in themselves, draw resources away from our primary research focus. In particular, we have concluded our obligations to the outreach programs FAIHM and MITE²S (Minority Introduction to Engineering, Entrepreneurship, and Science). Our substantially reduced commitment to those programs frees resources for the research that will serve MIT and our industry sponsors in the years to come.
Our current PD educational programs target three communities: working professionals, graduate students, and undergraduates.
For Working Professionals
The System Design and Management (SDM) Product Development Track was created in collaboration with MIT's SDM program. This two-year degree program targets mid-career engineering professionals who are potential leaders in product development. Students in the program continue to work for their employers at least half-time while pursuing their degree through an innovative educational structure: courses are broadcast to the students' company sites. Several times per year, students visit MIT to participate in special PD course modules, and for one semester in their program, students study on campus. At course completion, students receive an MIT S.M. degree in Engineering and Management.
To date, 13 faculty have participated in the SDM track, and 195 students from 54 organizations have enrolled. In January 2001, those organizations included Arvin Meritor, BAE Systems, the Chilean Army, Ford, Kodak, Lucent, National Aeronautics and Space Administration (NASA), Nortel, Northeast Utilities, US Air Force, and UTC.
SDM plans to keep its cohort at approximately 50 students per year using the current format. However, we anticipate growth through collaboration with industry partners who are interested in extending SDM education through other formats. For example, United Technologies Corporation has approached SDM with an interest in a systems integration curriculum that could impact 3,000 employees over the next three years. If successful, work on this project could be leveraged to include other Leaders for Manufacturing-System Design Management (LFM-SDM) industry partners.
Product Development for the 21st Century
CIPD began disseminating the SDM product development curriculum in 1998. Collaborating with the University of Detroit Mercy (UDM) and the Rochester Institute of Technology (RIT), we helped to create and implement the program called Product Development for the 21st Century (PD21), the Education Consortium for Product Development Leadership in the 21st Century. Guided by our industrial sponsors Xerox and Ford, the schools copy MIT's core PD curriculum, but emphasize project and case study work relevant in their geographic areas (optics in Rochester and automobiles in Detroit). The first PD21 students graduated from their respective schools in December 2000.
The Naval Postgraduate School (NPS) at Monterey joined the PD21 consortium in May 2000, and the first students entered its program in September. CIPD Co-Director Maurice Holmes participated in the program launch by presenting a concentrated course on-site at NPS. Introducing case studies on PD leadership, Professor Holmes taught 21 mid-career fast track Naval officers for one week in September. In addition, a cohort of 15 NPS students and 2 faculty visited MIT in June 2001, linking NPS students with their SDM and CIPD counterparts.
PD21 workshops brought together numerous students, faculty, and industry leaders this past year. In July 2000, CIPD hosted the two-day workshop, "Capturing the Value of the PD21 Program," which focused on gaining the most worth from the PD21 program, both from the point of view of students and their industry sponsors. In June 2001, NPS at Monterey hosted the three-day workshop, "System Engineering," which emphasized the integration of project management and system engineering.
PD21 consortium members met in June 2000 to review progress thus far and consider plans for the future. Members agreed unanimously that that the first phase of PD21-that of establishing the consortium as a multi-institutional platform for PD education-was a clear success. Over 400 students to date have enrolled in the program at the four schools, and 145 have graduated. The consortium established a committee to draft a statement of vision and objectives for PD21's second phase-disseminating the program beyond the core membership. This plan will be complete by the end of 2001.
With the PD21 program now firmly established, CIPD is free to reevaluate our commitment to PD21, given our transition away from a role as an NSF Engineering Research Center, and our increased allocation of resources on research that will more directly benefit MIT and our industry sponsors.
Executive Education Courses
CIPD faculty created and are teaching a full suite of short courses as non-degree offerings. These industrial mini-courses (15 this year) are offered throughout the year through MIT's office of summer professional programs and through Sloan's office of special executive programs. In addition, SDM faculty offer 12 courses by distance learning through the SDM Product Development Track.
For Graduate Students
Center Co-Director Steve Eppinger teaches a semester-long class, 15.783J (or 2.739J, or ESD.32J), entitled Product Design and Development. The class focuses on integration of the marketing, design, and manufacturing functions of the firm in creating a new product. Student teams develop new product concepts and create prototypes. This past semester, the class was composed of 90 graduate students from the School of Engineering, the Sloan School of Management, and the Rhode Island School of Design. Professor Eppinger was joined by other CIPD faculty who helped to teach this class: Professor Maurice Holmes, Dr. Dan Whitney, and Professor Thomas Roemer.
An associated web site (with its construction supported by CIPD) provides an extensive set of resources for students and support materials for faculty teaching comparable courses at other institutions. The site is organized around the book Product Design and Development as taught at MIT, and can be found at http://www/ulrich-eppinger.net/.
CIPD collaborated on the design of a semester-long MBA course based on an integrated theory of the firm. The course, 15.903 Integrative Course in Organizational Theory incorporates the latest (often CIPD-funded) research while analyzing insights from organizational theory and economics. The course was taught for the first time in the fall of 2000.
In June 2001, MIT awarded 10 Master's degrees and one Ph.D. degree to students funded by CIPD. An additional two Master's students and four Ph.D. students are expected to graduate later in 2001.
For Undergraduate Students
CIPD supported Professor Earll Murman in the development of a freshman seminar Introduction to Product Development. The seminar introduces freshman to product development with a project they undertake alongside PD professionals. (The course is co-taught by an MIT faculty member and a PD leader from industry.) CIPD sponsored Professor Murman is participating in the creation of a seminar instructor's guide. Available in May 2001, the guide is designed to enable institutions outside MIT to offer this course.
CIPD funded three UROP students for a project at the Center for Sports Innovation. They concentrated on applying the product development process to the sport of baseball. In particular, students investigated the design of a system for automatic detection of balls and strikes thrown during a game.
The class 2.007 Design and Manufacturing I teaches undergraduates the process of design by having them build machines that, in a grand finale at the end of the term, compete in a class-wide contest. This year, thanks to grants from Ford and GM, the students had no lab fee, and they could keep their machines when the class concluded. Led by Alex Slocum, teaching staff included CIPD faculty David Wallace and Woodie Flowers, and CIPD industry affiliates Sandy Campbell (Xerox) and Chris Magee (Ford).
Through a past grant from the National Science Foundation, CIPD provided equipment for lab facilities used by undergraduates. As part of their PD classes and student-defined activities in several engineering departments, undergraduates could develop and test prototypes to better understand how products are made. The final complement of new equipment was installed by the end of summer 2000, in new laboratory space that MIT elected to construct in the Aeronautics and Astronautics complex. The facility became available in the fall of 2000, and has been used by hundreds of undergraduates working on PD projects.
The center's outreach activities aim to provide life-long learning experiences for business professionals.
A collaboration of researchers and industry participants who are actively engaged in Distributed Object Modeling Environment Working Group (DOME) projects, the DOME Working Group is open to individuals and organizations that are actively involved in planning or executing a DOME project or pilot. The DOME Working Group meets regularly, with each member organization taking turns hosting a meeting. In February 2001, the Ford Motor Company served as host.
In September 2000, CIPD co-sponsored the Second International Design Structure Matrix Workshop on DSM along with the Lean Aero-space Initiative, Leaders for Manufacturing, and Ford Motor Company. Over 25 speakers presented research at this highly interactive gathering, which was attended by over 75 participants from industry, government, and academia (a 50 percent increase in attendance from the previous year). The Third International DSM Workshop will be held in October 2001.
While funded by the NSF, our industrial partnerships were limited to companies within the United States. CIPD is now actively seeking new partnerships, and in 2000-2001 we hosted visits by nearly 30 companies from around the world. They included such well-known firms as 3M, Boeing, Dessault, GE Corporate Research, Harley-Davidson, Nokia, Newell-Rubbermaid, Oracle, Saab, and Volvo.
Our goal for the coming years is to recruit partners in industry sectors that complement our existing membership. Our current partners operate mainly in the industrial, capital goods, technology hardware and equipment sectors, and we will target the telecommunications, aerospace, software, and information technology sectors.
Our partner companies have indicated a desire for a new sponsorship arrangement, one that better integrates our research and educational offerings. In July 2001, CIPD Co-Director Steven Eppinger assumed additional responsibility as Sloan Co-Director of LFM-SDM. With closer ties to these two programs, we are exploring an arrangement whereby LFM-SDM partner companies would sponsor CIPD doctoral students in conjunction with their LFM-SDM Master's degree internships.
The following is a list of recent events for the academic year 2000-2001:
- July 2000: Sloan Executive Education—Managing Complex Product Development Projects
- August 2000: Industry/NAVY R&D Partnership —This conference promoted dialogue between government, industry, and academia. The first of its kind, this conference focused on reducing barriers to integrating commercial products with R&D in Naval systems.
- September 2000: Second International Design Structure Matrix Workshop—Researchers presented the latest DSM theory and techniques, while practitioners had a chance to report on successful (or unsuccessful) experiences with DSM, including critical success factors and barriers to successful DSM methodology implementation. In addition, software developers demonstrated current DSM software implementation.
- October 2000: CIPD Executive Council Meeting—This meeting reviewed the summer's activities, and planned CIPD research for the following year; and CIPD Fall Research Review-This first of two annual reviews presented the latest theoretical and applied research conducted at CIPD. Intended for senior executives and managers involved in PD activities, it combined presentations about research projects with networking sessions for industry partners, CIPD faculty, and students. The review focused on three of CIPD's core initiatives: Information Flow Modeling (IFM), Implementation Dynamics (ID), and Distributed Object Modeling Environment (DOME).
- December 2000: CIPD Industry Workshop; Sloan Executive Education—Managing Complex Product Development Projects; and Sloan Executive Education-Developing and Managing a Successful Technology and Product Strategy.
- February 2001: CIPD Governing Board Meeting; DOME Working Group Meeting—Hosted by Ford Motor Company; and CIPD Spring Research Review-This review complemented the Fall Research Review by focusing on CIPD's remaining core initiatives: Virtual Customer (VC), Incentives and Boundaries (IB), Platform Architectures (PA), and the Product Development Integration Lab (PDIL).
- March 2001: Sloan Executive Education—Developing and Managing a Successful Technology and Product Strategy; and Sloan Executive Education-Managing Complex Product Development Projects.
- May 2001: Sloan Executive Program: Developing and Managing a Successful Technology and Product Strategy.
- June 2001: Sloan Executive Program: Product Design, Development, and Management.
The center's leadership structure has changed only nominally in the past year, providing evidence of our greater assurance in a role as a predominantly research-based center. Professors Steven D. Eppinger and Maurice Holmes remain Co-Directors of CIPD. Nils Nordal was promoted to Assistant Director in January 2001.
In addition, Michael Mack was hired as Communications Coordinator, and Sharon Bradshaw was hired as Manager of Finance (Ms. Bradshaw is also a member of the Sloan School's Office of Finance and Administration). Finally, Kathleen Sullivan joined CIPD as Office Manager, Cara Barber joined as Assistant to Professor Eppinger, and Sarah Davitt joined as Assistant to Professor Holmes.
More information about the Center for Innovation in Product Development can be found online at http://web.mit.edu/cipd/.