中文摘要：

碾压混凝土坝施工方案优选需要综合考虑施工工期、机械利用率、投资规模等多个因素的影响,是一个典型的多属性决策过程。当前研究大多未考虑专家知识水平的差异性和专家判断的模糊性对施工方案优选的影响;同时在施工方案优选研究中,指标权重也多为主观设定,没有明确的数理依据。针对以上问题,利用D数理论具有不考虑决策信息完备性的优势,本研究提出了基于熵权-D数模型的碾压混凝土坝施工方案优选方法。首先,由仿真计算结果得到不同施工方案评价指标,利用Pearson相关性检验方法分析处理评价指标之间的相关性,然后利用熵权法确定各个指标的权重,应用D数理论得到不同施工方案在不同辨识框架下的指标评价D数,最后根据不同专家的信任度排序和不同指标权重大小关系,确定施工方案综合评价D数的融合规则,实现方案优选指标的递归融合,将融合结果进行集成得到施工方案综合评价指标,并将其排序从而优选出最佳施工方案。工程实例表明,此方法可以在决策信息不完备的情况下实现碾压混凝土坝施工方案优选。

英文摘要：

Optimization of RCC dam construction schemes is a typical multi-attribute decision-making process that requires a comprehensive consideration of several factors such as construction period, utilization rate of machines, and investment scale. Most previous studies neglected the influence of differences in the experts＇ knowledge and the fuzziness in the experts＇ judgment on construction scheme decision and assumed the index weights subjectively in optimization of the schemes without a clear mathematical basis. This paper describes an optimization method of RCC dam construction schemes using an entropy-D model that, based on the D-number theory, does not need the completeness of decision-making information. First, the Pearson correlation test method is used to analyze the correlation between evaluation indexes obtained from simulations of four schemes. Then, index weights are determined using an entropyweighted method and the D-number theory is applied to calculate the D-numbers of each scheme in different identification frames. Finally, combination rules of the D-numbers are determined and recursive combination is carried out according to the trust degrees of different experts and the weights of different indexes. And integration of these fusion results gives a comprehensive evaluation index of the schemes and hence an optimal scheme can be selected. Engineering applications show that this method can achieve an effective scheme optimization for RCC dam construction with incomplete decision information.