On synthesis in the later phases of the mechanical engineering design process

Abstract: Only 15% to 30% of product development projects require the development of new product concepts. This means that the majority of design projects are carried out within the embodiment design and detail design phases. However, the majority of the research studies on the mechanical engineering design process, or design process for short, have been focusing on the conceptual design phase. The embodiment design and detail design phases are considered to be more routine-like and less complex, but the existing embodiment design and detail design methodologies (a methodology consists in a design process model, with related methods and supporting tools) suffer from at least two shortcomings: they are not as thoroughly developed as conceptual design methodologies, and there is a lack of dedicated methods and supporting tools for the generative design activities (hereafter called synthesis) of these phases. This thesis constitutes the first part of an overall research project whose goal is to contribute to the development of a support methodology for the synthesis activities in the later phases of the design process. As a first step, it has been decided to put the engineering designer, or designer for short, at the center of the study. As the designer is the primary user of a design methodology, it is necessary to understand the designer's thinking patterns, to clarify his/her skills and know-how, and to identify some common or specific actions to form the theoretical foundation upon which the support methodology can be developed. Thus the first step of this overall research project is to observe and analyze how the designer performs the synthesis activities in the later phases of the mechanical engineering design process. There is a lack of empirical pre-knowledge of the embodiment design and detail design phases; this first step is therefore explorative in nature and aims at covering this research domain extensively. The design activity of six designers (three students and three experts) has been observed in depth under an experimental setup.Each of the participants had to solve the same problem: to design a support device for a hydraulic piston. The verbal protocol analysis method has been applied to extract a model of their design activities, both on a strategic and tactical level as well as in a problem-solving perspective. The findings concerning the problem-solving process show that the designer, though following a fairly structured approach, developed no more than two alternatives, rapidly selected one of them and then lengthily studied and developed this alternative. There was no systematic evaluation moment. The activity of solution finding, characterized by synthesis, is balanced by an activity of mechanical modeling of the problem, which serves to both understand the generated solution and to monitor its correctness. At the strategic and tactical levels, the experts' design process has the following pattern: rapid understanding of the problem; consideration, very early in the process, of the shape of the parts and their interactions; concrete selection of materials; optimized selection of standard components; dimensioning of the joints. The early selection of standard components led to an early focus on the spatial restrictions and interface compatibility problems. On the other hand, the students reasoned about abstract mechanical structures, without defining the components until late in the process, and thus faced complications later on. A list of "weaknesses" observed on the part of the designers' process has also been established. The results of this explorative study now need to be further investigated; some complementary studies have to be carried out to statistically verify them. In addition to that, an investigation of the design activity in an industrial environment is needed in order to establish whether or not there are additional factors influencing the embodiment design and detail design phases.