Integration of behavior models' accuracy in design decisions using AHP, FMEA and physical prototyping

Abstract

The embodiment design phase consists of rough selections/arrangements of materials, technologies used, dimensions, and main components. During this phase, many behavior models are used to verify the achievement of design objectives. The lack of confidence in these models due to the assumptions adopted causes designers to realize many prototypes during product development, causing time/cost-consuming loops of the "trial-and-error" procedure. We propose a decision model that integrates the accuracy of behavior models into decision-making. The objective is to limit the use of physical prototypes and improve the quality of decision-making. Each design alternative is evaluated using two indicators. The first is a desirability indicator that measures the level of completion of design objectives. The second indicator assesses the risk associated with the accuracy of behavior models using AHP, FMEA, and experimental tests on a prototype. The proposed approach was applied to the development of a solar collector.