城市公共交通监管博弈建模与仿真  

Game modeling and simulation of urban public transport supervision

作　　者：张亚平[1] 宋成举[1,2] 程绍武[1] 郑柯[3] 

机构地区：[1]哈尔滨工业大学交通科学与工程学院,哈尔滨150090 [2]黑龙江工程学院汽车与交通工程学院,哈尔滨150050 [3]福州大学土木学院,福州350116

出　　处：《哈尔滨工业大学学报》2016年第9期24-29,共6页Journal of Harbin Institute of Technology

基　　金：黑龙江省自然科学基金(QC2014C060);黑龙江省交通运输厅科技项目(HJK2012A032);黑龙江工程学院博士基金(2015BJ03)

摘　　要：针对城市公共交通存在违规运营的问题,对管理部门与城市公共交通企业间的博弈关系展开研究。根据公交企业和管理部门相应的策略集建立双方博弈关系的收益矩阵,给出了不同策略下双方的期望收益和整体平均收益,利用复制动态方程描述双方策略的变化速度,给出了纳什均衡,利用系统动力学理论构建城市公共交通监管问题的博弈模型,并对模型进行了仿真分析。分析结果表明:采用静态惩罚策略时,博弈双方除纳什均衡外不存在演化稳定策略,波动难以控制;采用动态惩罚策略时,博弈双方存在演化稳定策略,演化时间和幅值与最大惩罚收益正相关,且博弈的稳定态与初始状态无关。Against the problem of urban public transport operation irregularities, the research was carried out to study the game relationship between management departments and public transport enterprises. The profit matrix of the game relations was established according to the corresponding strategy sets of public transport enterprises and management departments. Both expected profit and overall average profit under the different strategies are obtained. Nash equilibrium was obtained using replicator dynamics function to describe the changing rate of both strategies. The paper also establishes the game model of public transport supervision problem by using system dynamics theory, and simu[ation analysis to the model was carried out. The analysis results show that, when using static penalty strategy, there is no evolutionary stable strategy except Nash equilibrium on both sides, and fluctuations are difficult to control; while using dynamic penalty strategy, there exists evolutionary stable strategy on both sides, the evolution time and amplitude are positively correlated with maximum punishment revenue, and game steady-state is independent of the initial state.

关 键 词：城市公共交通 监管 系统动力学 博弈 纳什均衡 动态惩罚策略 

分 类 号：U121[交通运输工程]