Decision Making in Engineering Design Tasks: Do Designers Benefit from Representations of Uncertainty?

Abstract

Engineering design tasks require designers to compare, weigh and choose between many complex alternatives throughout the design processes. The uncertainty in design decisions and the high cost of poor choices has long made decision methods incorporating representations of uncertainty appealing from a theoretical standpoint. Yet, such methods have not been widely adopted in practical settings. This paper describes a study exploring reasons why this is so. The study found that in task of rank ordering a set of design alternatives from best to worst, a decision making method which incorporated a representation of uncertainty produced no better performance than a deterministic one. However, when compared to the informal methods typically followed by designers for rank ordering alternatives, both methods resulted in more consistent rankings for expert designers, but required more time. We will discuss the practical implications these findings may have for use of these methods in practice.