On the Fundamentals of the Engineering Design Process

Abstract: In an industrial context when the product development cycles are shortened and constraints on technical innovations are high, it is very important to provide a very efficient and effective engineering design activity. In the current engineering design literature, most methodologies are by and large based on the so-called systematic engineering design approach. This approach is characterized by an engineering design process model (well-defined design phases and steps) and a set of core methods that are both assumed to be generic. The initial focus of the thesis was set on a further improvement of the systematic engineering design methodology, specifically with the support of the embodiment design and detail design phases of the engineering design process, that is, support of the activities of establishing the product architecture of the technical system, selecting components, and creating and fully dimensioning each and every one of the subsystems and details. The first step of this overall research project was to observe and analyze the engineering designer at work to understand his or her thinking patterns and to map his or her skills and know-how, in oder to form the theoretical foundation upon which the support methodology could be developed. A subsequent procedural model was developed and tested with students. Different problems appeared during this investigation: the students using a systematic approach did not attain much better results than those who relied on their expertise and intuition; it seemed that other factors were at least as important as the systematic approach in terms of efficiency and effectiveness. These severe limitations were corroborated by similar observations in the literature. It was therefore decided to modify the objectives set out for the thesis towards a critical review of the systematic engineering design approach. Three aspects were investigated: the effectiveness of the systematic engineering design approach, that is, the degree to which it permits the design of technical systems to fulfill the requirements; the coherence and relevance of the fundamentals of the approach; and the interaction of the systematic engineering design process with the product development process models. Many empirical studies show that application of the methodology increases effectiveness, but it could also be observed that the central features of systematic engineering design process models — a step-by-step concretization of the product with a systematic variation of subsolutions and recombination — are not absolutely necessary, and that the proposed engineering design process model is far from being as generic as is supposed. The interaction between the engineering design process and product development models is not unproblematic either. This discrepancy between the reported shortcomings and the good empirical results of the systematic approach is explained by a substantial adaptation of the methodologies and by the positive indirect effects produced (favoring structure and documentation, easing consensus building among the team members, etc.). As many fundamentals of the systematic approach were questioned, the subsequent step was to search for new fundamentals that would constitute a core set of elements necessary for the development of new and/or revision of existing methodologies. To that end, the engineering designer's cognitive processes, the nature of problem solving and the context of product development were redefined. From these elements and the critical review of the systemic approach, a certain number of fundamentals are proposed: redefine the conceptual design activity as the development of the product until all critical risks are eliminated; search for a satisfying design, not necessarily an optimal one; use a pragmatic approach instead of breadth-first top down strategy; find a design process unique for each project, enabled by knowledge about the different product development approaches presented in the thesis. Hence the function modeling and morphological matrix methods are no longer central to the design activity. This set of fundamentals should also contribute to a deeper theoretical insight into today's engineering design methodologies and thus help establish those of tomorrow.