基于前景理論和證據(jù)推理的混合型多屬性決策方法

摘要： 針對(duì)屬性權(quán)重完全未知的混合型多屬性決策問(wèn)題,提出一種基于前景理論和證據(jù)推理的決策方法。通過(guò)直覺(jué)模糊數(shù)對(duì)精確數(shù)、區(qū)間數(shù)和語(yǔ)言變量3種混合型屬性的決策信息進(jìn)行統(tǒng)一,根據(jù)前景理論對(duì)決策信息進(jìn)行轉(zhuǎn)化;提出基于直覺(jué)模糊熵與相似度的屬性可靠性評(píng)估方法,結(jié)合屬性重要度確定屬性權(quán)重;采用證據(jù)推理算法集結(jié)屬性信息,得到方案的綜合前景值,并以此進(jìn)行方案排序。算例分析結(jié)果表明,所提方法具有較強(qiáng)的區(qū)分能力,能夠有效降低決策結(jié)果的不確定性,對(duì)混合型多屬性決策問(wèn)題具有較好的適用性。
For the hybrid multi-attribute decision making problems with completely unknown attribute weights, a decision-making method based on the prospect theory and the evidential reasoning was proposed. The decision-making information of three kinds of hybrid attributes including precision numbers, interval numbers and linguistic variables was unified by intuitionistic fuzzy numbers and transformed according to the prospect theory. An attribute reliability evaluation method based on intuitionistic fuzzy entropy and similarity was proposed. And the attribute weights were determined by combining the importance of attribute. The evidential reasoning algorithm was used to assemble the attribute information to obtain the comprehensive prospect values of alternatives, and then the alternatives can be sorted. The results of numerical example analysis show that the proposed method has a strong discriminating ability and can effectively reduce the uncertainty of decision-making results. The proposed method has good applicability to hybrid multi-attribute decision making problems.