A spectral analytic study of some postulates concerning engineering design and related processes

Abstract

An investigation is made of some of the important methodological problems arising in the study of engineering design processes. Independent support to the basis of previously developed generalized Poisson models of technological innovation and the reformulated technological progress function seems forthcoming. In light of the findings here, the coverage of the former would seem to be extended to both major as well as minor innovations. The ordinary version of the learning by doing hypothesis of technological change is decisively rejected. Spin-off, if any, in the invention process appears to be considerable and instantaneous. A formal theoretical view is emphasized to be a prerequisite to leading indicators approach to forecasting. An unambiguous indication provided by the results of this study is that short run changes in engineering design processes should be distinguished from those in the long run and transitory changes in design process from permanent changes.