模糊理论在铁路客运安全指标权重优化算法中的应用
Application of fuzzy theory to weight optimization algorithm of railway passenger transport safety index

摘要 客运安全评价是保证铁路运输安全高效的重要手段.为了解决安全评价中指标权重确定的问题，将模糊理论与层次分析法相结合，给出各评价指标权重的变化区间，并在各区间内生成不同权重值.通过蒙特卡洛模拟的仿真实验来生成针对不同指标权重值所产生的评价结果，并通过定量化方法得出该评价结果的方差.以方差作为遗传算法的适应度函数进行后向反馈，实现指标权重的优化，并将优化后的权重用于我国各铁路企业客运安全评价结果的计算.实验结果表明：该方法可以在充分尊重专家对指标权重评定的基础上，对权重进行微调，使得评价结果对指标得分更加敏感.
Abstract：Passenger safety evaluation is an important means to ensure the safety and efficiency of railway transportation. In order to solve the problem of index weight determination in security evaluation, this paper firstly combines fuzzy theory with Analytic Hierarchy P rocess (AHP) to give the weight variation interval of each index. Then, the simulation results of Monte Carlo simulation are used to generate the evaluation results for different index weights, and the variance of the evaluation results is obtained by the quantitative method.Finally,the variance is used as the fitness function of the genetic algorithm to carry out the backward feedback to realize the optimization of the index weight.And the optimized weights are applied to the calculation of passenger safety evaluation results of railway enterprises in China. The experimental results show that the method can fine tune the weights on the basis of full respect of the experts’ evaluation of the in dex weights, and make the evaluation results more sensitive to the index score.