An evidential-reasoning-interval-based method for new product design assessment

A key issue in successful new product development is how to determine the best product design among a lot of feasible alternatives. In this paper, the authors present a novel rigorous assessment methodology to improve the decision-making analysis in the complex multiple-attribute environment of new product design (NPD) assessment in early product design stage, where several performance measures, like product functions and features, manufacturability and cost, quality and reliability, maintainability and serviceability, etc., must be accounted for, but no concrete and reliable data are available, in which conventional approaches cannot be applied with confidence. The developed evidential reasoning (ER) interval methodology is able to deal with uncertain and incomplete data and information in forms of both qualitative and quantitative nature, data expressed in interval and range, judgment with probability functions, judgment in a comparative basis, unknown embedded, etc. An NPD assessment model, incorporated with the ER-based methodology, is then developed and a software system is built accordingly for validation. An industrial case study of electrical appliances is used to illustrate the application of the developed ER methodology and the product design assessment system. © 2008 IEEE.