基于Dijkstra-PD-ACO算法的大城市公交线路优化与评价方法研究  

Metropolitan Bus Linesoptimization and Evaluation Method Based on Dijkstra-PD-ACO Algorithm

作　　者：潘若愚[1,2] 褚伟[1,2] 杨善林[1] 

机构地区：[1]合肥工业大学管理学院过程优化与智能决策教育部重点实验室,安徽合肥230009 [2]安徽省公路交通物联网关键技术及装备工程技术研究中心,安徽合肥230009

出　　处：《中国管理科学》2015年第9期106-115,共10页Chinese Journal of Management Science

基　　金：国家自然科学基金资助项目(71331002;71501054;71271072;71201045)

摘　　要：城市公交线路的布设受到诸多影响因素的作用,单纯的追求线路单一指标的"最优"在实际公交线路优化时往往难以取得满意的效果。首先针对传统公交线路优化算法在求解线路优化问题中存在的不足,基于蚁群优化算法的寻优特性,结合Dijkstra算法在局部路径寻优中的优点,提出了Dijkstra蚁群混合优化算法。其次对于线路优化所得到的可行备选方案集,基于分层聚类主成分分析评价法进行优化效能评价。最后以合肥市一环内公交线路优化为实例进行验证。结果表明,本文提出的方法在兼顾客流密度最大、出行路径最短的同时给出了有效公交线路备选方案,优化结果也符合合肥市的实际情况。同时,该方法对我国大中城市公交线网优化问题的研究具有一定的借鉴意义。The city bus line layout is influenced by many factors.The simple pursuit of the best single indicator line in the actual bus route optimization is often difficult to obtain satisfactory results.Due to shortcoming of traditional bus route optimization algorithm for solving optimization problems that exist in the line,the paper analyzes the ant colony optimization algorithm optimization features.Combined with the advantages of the local path Dijkstra algorithm optimization,Dijkstra hybrid ant colony optimization algorithm is proposed.Secondly,for the optimization of the resulting line,a hierarchical clustering method of principal component analysis and evaluation to optimize the performance evaluation is proposed.Finally,the case study of bus lines optimization in Hefei is proposed as an example to verify the proposed algorithm.The results show that the proposed algorithm could take into account traffic density and maximum travel the shortest path,and gives an effective alternative.The optimization results are consistent with the actual situation in Hefei.The achievement of this paper has practical and realistic significance to the large urban public transportation network optimization in China.

关 键 词：大城市公交线路 最优路径 客流密度最大 优化效能评价 

分 类 号：U121[交通运输工程]