This paper shows a method for decision analysis using the interval priority weight vector estimated from a crisp pairwise comparison matrix. The maximin or maximax rule has been adopted to order alternatives according to the estimated interval priority weight vector. However, maximin and maximax rules consider only the worst-case and best-case scenarios, respectively. In practical decision-making, people often have different attitudes. To address this, we incorporate the Hurwicz rule, which accounts for both the best and the worst scenarios. In addition to overcoming the nonuniqueness of the solution to the estimation problem of the interval priority weight vector, we propose a method for decision analysis based on a set of interval priority weight vectors. Specifically, we draw a map showing the areas of all possible orders of alternatives in the parameter space defined by the sum of centers t and the optimism parameter \(\lambda \) used in the Hurwicz rule. According to the map, the decision maker recognizes all possible orders of alternatives and suggestions based on the evaluation policy and attitude. The proposed analysis can give detailed suggestions for the DM even when the total utility scores of alternatives obtained by the classical AHP are close.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Decision Analysis with the Hurwicz Decision Map Under a Set of Interval Priority Weight Vectors

  • Yeyang Hong,
  • Masahiro Inuiguchi,
  • Shigeaki Innan

摘要

This paper shows a method for decision analysis using the interval priority weight vector estimated from a crisp pairwise comparison matrix. The maximin or maximax rule has been adopted to order alternatives according to the estimated interval priority weight vector. However, maximin and maximax rules consider only the worst-case and best-case scenarios, respectively. In practical decision-making, people often have different attitudes. To address this, we incorporate the Hurwicz rule, which accounts for both the best and the worst scenarios. In addition to overcoming the nonuniqueness of the solution to the estimation problem of the interval priority weight vector, we propose a method for decision analysis based on a set of interval priority weight vectors. Specifically, we draw a map showing the areas of all possible orders of alternatives in the parameter space defined by the sum of centers t and the optimism parameter \(\lambda \) used in the Hurwicz rule. According to the map, the decision maker recognizes all possible orders of alternatives and suggestions based on the evaluation policy and attitude. The proposed analysis can give detailed suggestions for the DM even when the total utility scores of alternatives obtained by the classical AHP are close.