STATE OF THE ART IN JAPANESE COMPUTER-AIDED DESIGN METHODOLOGIES FOR MECHANICAL PRODUCTS -- INDUSTRIAL PRACTICE AND UNIVERSITY RESEARCH The author spent 3 months in Japan as a temporary liaison scientist with the INTRODUCTION I spent from 3 June to 13 September as a temporary liaison scientist at ONRASIA, the Far East office of the Office of Naval Research, while on leave from the Charles Stark Draper Laboratory, Inc., Cambridge, MA. ONR asked me to survey Japanese use of computers in design of mechanical products, to report on the state of practice in Japanese companies, and to determine research needs and trends in both industry and academia. The resulting report contains a thorough mix of engineering, management, science, and technology issues. This is appropriate for manufacturing, a complex and challenging area where practical solutions are often needed before theoretical understanding is available, and where major innovations have come from practitioners. Research often plays the role of replacing the idiosyncratic nature of these solutions by Daniel E. Whitney with an intellectual foundation and Japanese companies have BACKGROUND This is my seventh visit to Japan since 1974, but the first to last more than 2 weeks. My research interests are in robotics, computer-aided design (CAD), product design methodologies, design for assembly, and the relation between technical and business aspects of product design. On prior visits I have regularly visited the same laboratories, companies, and people over and over. The reason is to follow the maturing of the people and the companies and the evolution of their approaches to generic problems. The topic of discussion on prior visits usually was robotics and its application to industrial assembly. On prior trips I visited mostly industrial sites because I found that industrial people were closer to real problems and solved them with remarkable ingenuity combined with significant intellectual depth. This contrasted sharply with university research in robotics, which in my view (in the United States as well as Japan) tended to focus too much on far advanced technology and attempted to solve all robotics problems by enhancing the technology of the robot. Industry took a more balanced view and attempted to improve robot technology in feasible as well as economically high-leverage ways (mostly by increasing motion speed) and improve robot applicability to tasks by redefining and redesigning the tasks (mostly by pervasive and thoughtful product redesign). This combination has proven to be very powerful. I have written about this contrast in several venues (Ref 1). On this visit I have focused on design and found the universities to be more interesting than before, largely because they have tightened their contacts with industry and have identified several very relevant research tracks that should flow directly to industry very soon. Still, companies are on the front line of design and have, as in robotics, applied considerable intellectual force to defining their problems and evolving impressive responses. Companies visited represent heavy, medium, and precision industry. I identified many of them from prior visits, while others were identified and contacted by Prof. Fumihiko Kimura of the University of Tokyo, whose collaboration is greatly appreciated. CAVEATS In a 3-month visit, it is not possible to obtain a statistically significant sample. I visited companies and laboratories that are leaders in one way or another. A typical visit took most of one day and involved from 3 to 10 company personnel. While they hosted me attentively (in some cases lavishly), they did not have time to tell me everything and in no case did they place anything on the agenda that had not been published or discussed with outsiders before. I am certain they were frank and responsive to my many questions, however. Where capabilities or activities of different companies are mentioned below, they should therefore not be taken as having comparative value. That is, if company A mentioned capability Management Style Versus The business schools and the engineering schools have differing views of product design. Several business schools, notably Harvard and the Massachusetts Institute of Technology (MIT), have extensively studied the world automobile business to determine why the Japanese auto industry can deliver new models faster than their European or U.S. competitors. The marketing value of doing so has long been recognized, and Japanese companies in other industries such as laptop computers and consumer electronics are similarly adept. Prof. Kim Clark of Harvard and his former student, Prof. Takahiro Fujimoto, now at the University of Tokyo, say that the main reason is specific management methods such as overlapping design tasks that are normally done in sequence (Ref 2). In offering such explanations, the business researchers tend to ignore or downplay the role of computer design aids, such as computerized drafting, solid modelers, data management, rapid electronic communication, and so on. Several companies agree. Engineering schools and researchers tend to ignore the management factors and look exclusively at the computer tools. To put the contrast bluntly, the business researchers think computers are commodities that anyone can buy and use, while management techniques are the product of decades of "corporate learning" that others cannot buy or copy. The engineering researchers feel that management practices can be copied and learned, too, or are merely "social factors," whereas design and manufacturing engineering software, knowledge, and data about past designs, properly captured and deployed in computers, can convey considerable competitive advantage in terms of design speed, accuracy, and quality. Toyota |