Fuzzy Concept Evaluation Based on Prospect Theory and Heterogeneous Evaluation Information

Abstract

Concept evaluation is the core stage of new product development and has a significant impact on the downstream process of product development. Because of the uncertainty and ambiguity of early design information, the concept evaluation process not only relies on the semantic terms of decision makers (DMs) but also includes uncertain criteria values (such as crisp numbers and interval numbers). In addition, DMs will have psychological preference errors when evaluating concepts owing to the risks taken in the evaluation. To address these drawbacks, a fuzzy concept evaluation approach based on prospect theory and heterogeneous evaluation information is proposed. Initially, based on the definition of intuitionistic fuzzy numbers (IFNs), a numerical model is developed to unify the representation of crisp numbers, interval numbers, and fuzzy numbers, and a normalized decision matrix is constructed. Second, a weight distribution method of DMs is proposed, which introduces hesitation and similarity, and the weighted intuitionistic fuzzy evaluation information is transformed into interval IFNs. Third, the weight of the evaluation criteria is determined using the decision-making trial and evaluation laboratory model (DEMATEL), and an evaluation information correction model based on the prospect theory is established to select the optimal concepts. Finally, the feasibility of the proposed approach is demonstrated using a bar-peeling machine, and a comparison and sensitivity analysis is conducted to verify the robustness of the decision results.