摘要
在多式联运路径优化模型中,权重赋值是复杂的多目标决策问题。提出一种主客观相结合的综合权重确定方法,先进行主观赋值,再结合贝叶斯网络和极大熵准则进行自学习。通过贝叶斯网络将各目标属性及影响因素相关联,再结合极大熵准则对权重进行自学习来输出各目标权重。通过该方法得到的权重结合了主观判定并通过自学习来减少人为因素偏差,提高了权重的客观准确性,为多式联运中的多目标决策提供技术支持。
In the multimodal transport path optimization model,the weight assignment is a complex multi-objective decision problem. We proposed a comprehensive weight determination method which combined subjectivity and objectivity. In the method,subjective assignment was done,and then self-learning was conducted by the combination of Bayesian network and maximum entropy criterion. The target attributes and the influencing factors were correlated by Bayesian network,and the weight was self-learned to output the weight of each target by using the maximum entropy criterion. The weights obtained by this method combine subjective judgments and reduce the deviation of human factors through self-learning. It improves the objective accuracy of weights and provides technical support for multi-objective decision-making in multimodal transport.
引文
[1]郭亚军.综合评价理论、方法及应用区[M].北京:科学出版社,2007.
[2]Saaty T L.The Modern Science of Multicriteria Decision Making and Its Practical Applications:The AHP/ANP Approach[J].Operations Research,2013,61(5):1101-1118.
[3]Reazaei J.Best-worst multi-criteria decision making Method[J].Omega,2015,53:49-57.
[4]李刚,迟国泰,程砚秋.基于熵权TOPSIS法的人的全面发展评价模型及实证[J].系统工程学报,2011,26(3):400-407.
[5]李刚,程砚秋,董霖哲,等.基尼系数客观赋权方法研究[J].管理评价,2014,26(1):12-22.
[6]张文宇,王磊.基于集成赋权的微信营销效果模糊综合评价[J].计算机工程与应用,2016,52(11):68-76.
[7]顾欣欣,任倩,安睿.基于组合赋权和灰色相似关联度的区域公路网服务水平评价方法[J].交通运输研究,2015,1(3):1-6.
[8]Zografos K G,Davis C F.Multi-Objective Programming Approach for Routing Hazardous Materials[J].Journal of Transportation Engineering,1989,115(6):661-673.
[9]吴宗之,多英全,刘茂,等.危险品道路运输过程风险分析与评价方法研究[J].应用基础与工程科学学报,2004(12):36-44.
[10]黄解军,万幼川,潘和平.贝叶斯网络结构学习及其应用研究[J].武汉大学学报(信息学版),2004,29(4):315-318.
[11]徐梦佳,王渝红,竺懋渝,等.家居智能用电水平指标体系构建方法研究[J].现代电力,2017,34(2):74-79.
[12]邱菀华.管理决策与应用熵学[M].北京:机械工业出版社,2002.