基于聯盟博弈?云化TODIM的方案評價方法研究

張志霞 張二雙

摘? 要： 針對方案評價中評級主體多考慮評價信息的模糊性而忽略信息隨機性，在主觀判斷時常伴有參照依賴和損失規避等特性，提出一種基于云模型和聯盟博弈法改進的交互式多屬性決策（TODIM）方案評價方法。在方案評價中，選用云模型量化語義評價信息，計算專家評價云與算術平均云兼顧形狀?距離的云模型綜合相似度以獲取專家的動態權重，引入聯盟博弈法求取各指標組合權重，采用云模型與聯盟博弈法改進的TODIM方法計算每個方案相對于其他方案的綜合排序值，實現對方案的優選，最后通過智能識別系統候選方案的擇優證明該方法的有效性。

關鍵詞： 方案評價; 聯盟博弈; TODIM法; 云模型; 組合權重求取; 方案優選

中圖分類號： TN911?34; TH122? ? ? ? ? ? ? ? ?文獻標識碼： A? ? ? ? ? ? ? ? ? ? ? ?文章編號： 1004?373X（2020）08?0022?07

Research on TODIM scheme evaluation method based on alliance

game and cloudification

ZHANG Zhixia， ZHANG Ershuang

（School of Management， Xian University of Architecture and Technology， Xian 710000， China）

Abstract： In allusion to the problem that the rating bodies lay stress on the fuzziness of evaluation information but ignore the information′s randomness in the scheme evaluation， which is often accompanied by the characteristics of reference dependence and loss avoidance in the subjective judgment， an improved interactive multi?attribute decision?making （[TO] mada de Decisao Interativa Multicritévio， TODIM） scheme evaluation method based on cloud model and alliance game method is proposed. In the scheme evaluation， the cloud model is adopted to quantize semantic evaluation information， and the comprehensive similarity of the cloud model that the specialist evaluation cloud and the arithmetic average cloud give consideration to both shape and distance is calculated to obtain the specialists′ dynamic weights， and get the combination weight of each evaluation index by introducing the alliance game method. The improved TODIM method based on cloud model and alliance game method is used to compute the comprehensive sorting value of each scheme relative to the others， so as to realize the optimization of schemes. The effectiveness of this method is proved by the merit of candidate schemes in the intelligent identification system.

Keywords： scheme evaluation; alliance game; TODIM method; cloud model; combination weight derivation; scheme optimal selection