基于组合赋权-可变模糊集的装配式建筑施工安全评价*

为解决装配式建筑施工安全影响因素的复杂性和模糊性，引入组合赋权和可变模糊集理论相结合的方法对装配式建筑施工安全进行评价。通过分析装配式建筑施工特点，以WSR方法论为基础，选取“人、机、法、环、管”5方面因素构建装配式建筑施工安全评价指标体系并划分指标等级标准；采用博弈论思想对G1法确定的主观权重和熵权法计算的客观权重进行最优组合，进而确定综合权重；根据可变模糊集理论量化后的级别综合特征值确定最终评价结果；将该方法用于甘肃省某装配式试点项目，评价结果与其他方法进行对比，并采用单因素轮换OAT法分析5个子系统内的指标敏感性，以便决策者识别出最敏感指标。结果表明:该实证项目安全等级为Ⅱ级，偏向Ⅲ级，与实际情况吻合；组合赋权-可变模糊集方法所得评价结果与灰色聚类分析、未确知测度方法基本一致，使评价结果准确性和稳定性得到有效提高。

Safety evaluation of prefabricated building construction based on combination weighting and variable fuzzy sets

In order to solve the complexity and ambiguity of the factors affecting the safety of prefabricated building construction,a combined method of combination weighting and variable fuzzy set theory was introduced to evaluate the safety of prefabricated building construction.Firstly,through analyzing the characteristics of prefabricated building construction,and based on the WSR methodology,five factors of “man,machine,method,environment and management” were selected to construct an evaluation index system for the safety of prefabricated building construction,and the standards of indexes level were divided.Secondly,the thought of game theory was used to optimally combine the subjective weights determined by the G1 method and the objective weights calculated by the entropy weight method,thus the comprehensive weight was determined.Finally,the final evaluation results were determined according to the comprehensive characteristic value of level after quantization by the variable fuzzy set theory.This method was used in a prefabricated pilot project in Gansu Province,and the evaluation results were compared with other methods.Then,the single factor rotation OAT method was used to analyze the sensitivity of indexes in the five subsystems so that the decision makers could identify the most sensitive indexes.The results showed that the safety level of the empirical project was level II,which was biased towards level III and in line with the actual situation.Under the premise that the evaluation results obtained by the combined weighting variable fuzzy sets method were basically consistent with the grey cluster analysis and the unascertained measurement method,the accuracy and stability of the evaluation results were effectively improved.