COGNITIVE-PROCESSES IN ENGINEERING DESIGN - A LONGITUDINAL-STUDY

Citation
Lj. Ball et al., COGNITIVE-PROCESSES IN ENGINEERING DESIGN - A LONGITUDINAL-STUDY, Ergonomics, 37(11), 1994, pp. 1753-1786
Citations number
38
Categorie Soggetti
Ergonomics,Ergonomics
Journal title
ISSN journal
00140139
Volume
37
Issue
11
Year of publication
1994
Pages
1753 - 1786
Database
ISI
SICI code
0014-0139(1994)37:11<1753:CIED-A>2.0.ZU;2-8
Abstract
In this paper we report an empirical study of the cognitive processes of semi-expert electronics engineers pursuing real-world design projec ts. Extensive diary and interview data were analysed so as to reveal t he organization of on-going design activity and the goals that were mo tivating behaviour. Our analyses indicated that subjects were implemen ting a highly systematic design strategy which deviated only a small a mount (12%) from a top-down, depth-first procedure. Some of this devia tion could be accounted for by social demands impinging on the individ ual designer, and whilst there was also evidence of 'opportunistic' de viation from a structured design approach, this did not appear to be a significant feature of our subjects' behaviour. We present a model of the design process as a set of production rules which describe an abs tract 'design schema' for electronic engineering. This design schema e mbodies processes which control and co-ordinate problem-understanding and problem-structuring activity as well as the essentially top-down, depth-first pursuit of design solutions. The schema also allows for fl exibility in the design process, permitting subjects to cope with cont ingencies arising through social influences and performance breakdowns . A further aspect of the design schema is that it encapsulates a 'sat isficing' procedure which reflects a principle that dominated our engi neers' solution-search and evaluation activity - they characteristical ly focused exclusively on initial, satisfactory (but usually sub-optim al) solution concepts rather than comparing alternatives with the aim of optimizing choices. We conclude our discussion by recommending faci lities that we feel computer-based design aids should offer users if t hey are to be genuinely useful adjuncts to design activity.