基于高度冲突专家意见融合的水利工程应急决策

为解决D-S(Dempster-Shafer)证据理论无法有效集结高冲突专家意见对水利工程应急决策结果的影响,提出了基于Lance距离和信息熵改进D-S证据理论的水利工程应急决策中的专家意见融合方法。首先,利用改进的直觉模糊熵计算属性权重,充分考虑了决策信息的模糊性,得到各专家对方案的Mass函数；然后,使用Lance距离衡量任意两个专家之间的冲突和差异性得到评价证据的可信度,使用信息熵描述专家决策时的随机性问题得到评价证据的不确定度,并将二者折扣处理后得到专家综合权重,保证修正权重的准确合理；最后使用D-S证据组合规则计算修正后各专家对方案的Mass函数,得到最终决策结果。通过实例分析及方法对比验证了模型的有效性,结果表明,提出的方法可以有效集结高度冲突专家意见,与其他方法相比融合结果精度更高,考虑问题更为科学有效,可为水利工程应急决策提供依据。

Emergency Decision Making in Hydraulic Engineering Based on Highly Conflicting Expert Opinions Fusion

An expert opinion fusion method is proposed to address the problem of the inability of D-S (Dempster-Shafer) evidence theory to effectively assemble high-conflict expert opinions on the emergency decision-making results of hydraulic engineering. This method introduces improvements to D-S evidence theory by incorporating Lance distance and information entropy. Initially, the improved intuitionistic fuzzy entropy was utilized to calculate attribute weights, considering the fuzziness of decision-making information. This process resulted in Mass functions for each expert regarding the proposed solutions. Subsequently, the credibility of the evaluation evidence was determined using Lance distance to measure the conflict and discrepancy between any two experts. Information entropy was employed to describe the randomness in expert decision-making, thus obtaining the uncertainty of the evaluation evidence. These aspects were discounted to obtain the expert''s comprehensive weight, ensuring accuracy and reasonableness in the revised weight. Finally, the D-S evidence combination rule was applied to compute the Mass functions of experts'' revised opinions on the solutions, ultimately providing the final decision outcome.The effectiveness of the model was verified through case analysis and method comparison. The results demonstrate that the proposed method can effectively assemble highly conflicting expert opinions. In comparison with other methods, the fusion results are more accurate, and the considerations regarding the problem are more scientific and effective. This method can serve as a basis for emergency decision-making in hydraulic engineering.