A novel integration of MCDM methods and Bayesian networks: the case of incomplete expert knowledge

Abstract In this study, we propose an effective integration of multi criteria decision making methods and Bayesian networks (BN) that incorporates expert knowledge. The novelty approach is it provides support in case the experts have partial We use decision-making trial evaluation laboratory (DEMATEL) to elicit causal graph BN based on knowledge experts. alternatives which make up initial matrix technique for order preference by similarity ideal solution (TOPSIS). then parameterize using Ranked Nodes allows submit their with linguistic expressions. analytical hierarchy process determine weights TOPSIS rank alternatives. A supplier selection study conducted illustrate effectiveness proposed approach. Two measures, namely, number mismatches distance due mismatch are developed assess performance scenario analysis 5% 20% missing values increment demonstrate our remains robust as level increases.