Engineering design is a process that requires information to be processed and decisions to be made in the presence of significant levels of imprecision and uncertainty. Traditional approaches to engineering have focused on analyzing precise information. Our research has introduced a formal method for representing and manipulating imprecise information in engineering design to enable designers to compare the performance of design alternatives, even at the highly imprecise preliminary stages.
Engineering design is a process comprised of many recognizable stages: analysis of requirements, concept generation, concept evaluation and refinement, embodiment design, detail design, etc. Once several concepts have been generated, this process is one of reducing the uncertainty with which each design alternative is described. This uncertainty is comprised of uncertainty in choosing among alternatives (imprecision) and also stochastic uncertainty (usually pertaining to random, uncontrollable events). The level of imprecision in the description of design elements is typically high in the preliminary phase of engineering design. As the design process proceeds, the imprecision with which each design variable is known is reduced, although the stochastic uncertainty usually remains).
An imprecise variable in preliminary design is a variable which may potentially assume any value within a possible range because the designer does not know, a priori, the final value that will emerge from the design process. The nominal value of a length dimension is an example of an imprecise variable. Even though the designer is uncertain about the final value of a variable, he or she usually has a preference or desire for choosing certain values over others. This preference is used to quantify the imprecision with which design variables are known in the preliminary phase of engineering design. Imprecision is represented and manipulated in this way as a basis for decision-making during the progression of the design process from imprecise design configurations to a precise physical device or process.
Stochastic uncertainty arises from a lack of exact knowledge of a variable due to some process the designer has no direct control or choice over. A manufacturing tolerance is an example of such an uncertainty.
We have also developed methods that permit the designer to implement an explicit design strategy, depending on the particular problem. The strategy can range from non-compensating (which reflects a conservative design process) to a compensating, aggressive approach. These strategies permit the designer to choose values for design variables for each alternative configuration, including the effects of imprecision and uncertainty, by trading-off one or more performance variables against others, and/or one or more design variables against others.
Our semi-automated design method was developed to aid the designer in making decisions in the the preliminary phase of design, and has the following objectives:
This material is based upon work supported, in part, by the National Science Foundation under NSF Grant Numbers DMI-9813121 and DMI-9523232. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the sponsor(s).