Analysis of the Engineering Design Process in an Industrial Context
Author: Hales, Crispin
Supervisor: Wallace, Ken
Institution: University of Cambridge
This dissertation is concerned with the process of engineering design, and the development of ways to improved design quality in industry through more effective management of design activities.
A review of the literature suggests that the way in which needs and ideas are converted into information for manufacturing products and technical systems is not yet fully understood, despite a long history of engineering design in industry. Better understanding of what happens in practice is needed. Participant observation of actual design processes, where the researcher takes an active part, records what happens and analyses the field data collected, is a recommended research approach. the participant observation of an engineerign design project involving gthe design of a high-pressure, high-temperature system for testing materials in a simulated coal gassification system is described. A systematic approach was used to structure the design work, and all activities were recorded during the 2.8-year project. In total 1180 pages of field notes, 76 hours of tape-recordings, 116 weekly report and 6 design reports were accumulated. These covered 1373 separate events or 'interchanges', and detailed the 2368 hours of project effort. 'Interchange data sheets' were compiled for each of the 37 participants, and the 2488 coded records were entered into a computer for sorting and categorizing. The reduced data resulting from this is analysed both quantitatively and qualitatively in terms of the engineering design process. To clarify the context within which the project took place and to help structure the analysis, a Context Model is described. It represents the phases of the engineering design process in terms of its 'activities' and 'outputs', set in context within the Project, within the Company, within the Market and within the Environment. The quantitiative analysis shows that the engineerign design process may be characterized by a set of overlapping phases, each consisting of a particular mix of procedural steps and other general activities. A comparison between the 'phase diagram' of design effort and an 'ideal' diagram indicates ways of assessing progress and identifying problems during an engineering design project. The design 'activities' during the project are compared with the procedural design steps referred to in the literature, and six general activities are added. The design 'techniques' used during the project are compared with those suggested in the literature, and thirteen working, communicating and motivating techniques are added. Theoretical and observed design outputs are compared. Work 'type', team location and team 'mood' are discussed.
A tentative list of 103 factors likely to influence the engineering design process is generated from the literature, divided into 20 categories of influence at five levels of resolution. The impact of each factor on the project is assessed. An attempt is also made to assess the effectiveness and efficiency of the design process and the success of the project. Several recommendations for further research are made, including: the use of phase diagrams; comparative studies of the observed 'activities' and 'techniques' for different projects; assessment of design process outputs; and development of a design terminology acceptable to related disciplines.
A dissertation submitted to the University of Cambridge for the Degree of Doctor of Philosophy, Department of Engineering, 1987