While discussing design methodologies, it is important to understand that what is being described is a set of conceptual approaches that are used to develop the product realization process. In light of the coupled decision areas in composites and the need for early design decisions, the design process itself should be thought of as one that stretches from conceptual design to actual process design and fabrication. The methodologies outlined below, with the exception of the total quality design (TQD) process, were however developed for generic use, but have been found to be universally applicable. There are two major schools of thought regarding product development methodologies: (1) the scientific, as represented by Dixon (1988), and (2) the engineering method, as represented by Koen (1985). In the former, prescriptive methodologies are only developed following the development of accurate descriptive methodologies that lead to testable theories of product development. The latter prescribes that the prescriptive methodology be put forth based on the best available information, and then modified as necessary. Again there is perhaps more evidence of their use in the area of composites in Japan, than there is in the U.S.
This is also termed the "general procedural model of design engineering," and is a highly structured and detailed model whose main characteristics can be captured in Table 7.1. Although fairly detailed, it is perhaps more useful as a teaching aid than a methodology for composites product development. More than anything else its lack of applicability to composites stems from the fact that it does not focus on the early part of the design process, but concentrates on the later stages, which makes it unable to address the concerns in Figure 7.2. However, it is perhaps this method that is still used most often in composites development.
This methodology is very adaptive and actually prescribes little except a set of three rules: (1) adopt a framework; (2) select design methods to flesh out the framework; and (3) continually review and update the framework during the development effort.
Cross suggests that the framework could be made up of six basic steps: (1) clarifying objectives; (2) establishing objectives; (3) setting requirements; (4) generating alternatives; (5) evaluating alternatives; and (6) improving details. The process is thus almost linear and in fact completely leaves out the important aspects of team-customer interactions.
Clausing's methodology is more structured and attempts to evaluate each element on the basis of its impact on competitiveness, but it does focus on teamwork and collaboration. The methodology is organized around the following 10 cash drains of product development: (1) technology push, but where's the pull; (2) disregard for the voice of the customer; (3) eureka concept; (4) pretend designs; (5) pampered products; (6) hardware swamps; (7) here's the product -- where's the factory; (8) we've always made it that way; (9) inspection; and (10) give me my targets -- let me do my thing.
Clausing further prescribes the use of Pugh's concept selection method (Pugh 1981) and the House of Quality (Hauser 1988) as tools to be used. Although a very useful and illuminating methodology, it lacks the tools for actual application.
Summary of the General Procedural Model
Developed initially by Akao in Japan in 1966, quality function deployment (QFD) provides a formal structure for ensuring that customer wants and needs are carefully heard, and then directly translated into a company's requirements for a product as shown in Figure 7.3.
The QFD process maps customer driven requirements to technical requirements, allowing competitive evaluation and weighing of factors to achieve a technical evaluation. The process was developed to this level by Bob King (1988) and further used by Clausing.
Developed at the Center for Composite Materials at the University of Delaware, this methodology attempts to synthesize the best of the other methodologies with an emphasis on hearing the voice of the customer and developing design interactions early in the design process. The methodology consists of five main elements described in Figure 7.4.
Figure 7.3. The Four-Phase Approach to QFD
Figure 7.4. The Elements of the TQD Process
The use of these elements allows for organizational flexibility, a clear management vision, and an integrated implementation plan. Most importantly, it provides a framework for the integration of the functional disciplines of economics, design, and manufacturing towards a common goal of composites product development. In order to facilitate the easy use of this methodology, a set of Macintosh-based spreadsheets have been developed to organize the process of conceptual development of products. These are analogous to the House of Quality matrices in which the "Hows" of one stage in development are broken up into clarifying "whats" of the next stage (Hauser and Clausing 1988). The schematic in Figure 7.5 shows the link between each of the spreadsheet based TQD programs, and the links to other activities that must be conducted simultaneously, such as concept generation. The templates serve as facilitators for the integration of the various disciplines, as well as an electronic record of decisions made and the reasoning behind each decision.
Figure 7.5. Schematic of Use of TQD Templates
Winner et al. define concurrent engineering as:
a systematic approach to the integrated, concurrent design of products and their related processes, including manufacture and support. 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 quality, cost, schedule, and user requirements. (Winner 1988)
In the context of this paper, the term concurrent engineering is then taken to represent the approach of collaborative product development with input from each functional group involved in the product development process. The approach taken herein is to include as many tools as necessary, which may include the methodologies mentioned above. The need for the use of a number of methodologies is due to the fact that no single methodology addresses all the concerns for successful product development. A snapshot of this diversity is provided through a comparison made by Henshaw (1989) of the methodologies based on a number of key factors as needed in composites product development (Table 7.2).
A test bed for the use of the concurrent engineering approach through software integration has been established at University of West Virginia at the Concurrent Engineering Research Center (CERC). A major component is a demonstration bed of the use of concurrent engineering for the development of metal-matrix composite (MMC) components for aircraft. An extensive list of 26 tools usable for concurrent engineering is given by Pennell and Akin (1990) and hence will not be repeated herein. Further discussion of case studies and frequency of the use of formal methods (including design of experiments, pareto charts and QFD) is given in the IDA report by Winner et al. (1988) and is not included herein since it contains no breakup to show the actual extent of use by the composites community. It is however known that the DoD routinely attempts to use some form of concurrent engineering and TQM for its own composites projects. It is also being used in the Composite Armored Vehicle (CAV) project.