Pseudo-criteria versus linear utility function in stochastic multi-criteria acceptability analysis

摘要: Abstract Stochastic multi-criteria acceptability analysis (SMAA) is a decision support method for multiple decision-makers (DMs) in discrete problems. SMAA does not require explicit or implicit preference information from the DMs. Instead, based on exploring weight space order to describe valuations that would make each alternative preferred one. Partial can be represented through distributions. In this paper we compare two variants of using randomly generated test problems with 2–12 criteria and 4–12 alternatives. original SMAA, utility value function models DMs' structure, inaccuracy uncertainty by probability SMAA-3, ELECTRE III-type pseudo-criteria are used instead. Both methods compute an index measuring variety different supports alternative, central vector representing typical resulting decision. We seek answers three questions: (1) how similar results provided models, (2) what kind systematic differences exists between (3) could one select indifference thresholds model match given distributions?