兰州市公共自行车租赁点配置评估与优化

发布时间：2019-05-15 18:23

【摘要】：随着城市化进程的快速推进,城市的发展与交通拥塞的矛盾日益凸显,大力发展公共交通是缓解该矛盾的一个有效手段。在选择公共交通工具时,地铁和公交车无疑成为人们的首选,但是,无论是地铁还是公交车,均不能完全解决末端交通问题,即无法将乘客直接送达到目的地。在公共交通中,公共自行车作为一种解决末端交通的工具,为人们的出行提供了方便,然而,由于城市各区域的部分公共自行车租赁点设置不尽合理,经常会出现“有车无桩锁,桩空无车取”的现象,给市民选择使用公共自行车到达目的地带来困难。因此,提高公共自行车系统的服务效率,引导市民选用公共交通工具出行,成为亟待解决的问题。本文以兰州市公共自行车租赁点的运行数据为基础,经过对运行数据的分析,提出了公共自行车系统租赁点设置和锁桩配置评价模型,该模型通过改进C4.5分类算法对公共自行车系统进行评价,并根据系统动力学原理,构建基于AnyLogic的公共自行车系统仿真模型,以租赁点的设施利用率和服务度为评价指标,利用敏感性分析和迭代相结合的方法优化公共自行车租赁点的配车数。本文的研究内容如下:(1)公共自行车系统运行数据的统计分析。基于兰州市公共自行车系统的实际运行数据,利用R语言对租赁点的租车频次、还车频次、锁桩数量和用车时长等数据进行统计,计算数据间的耦合协调度,并结合兰州市的交通地形特征,确定单个租赁点设置的效果等级。(2)构建公共自行车租赁点设置效果的评价决策树。首先分析了常用决策树分类算法的特点,选择了C4.5分类算法,并对算法的计算过程进行改进,利用改进的C4.5分类算法,构建公共自行车租赁点设置和锁桩配置评价决策树,快速直观地获取租赁点设置的优劣。这部分研究为公共自行车租赁点的优化提供依据。(3)构建公共自行车系统仿真模型,提出优化租赁点配车数的方案。针对部分公共自行车租赁点自行车利用率低和服务质量不高的问题,首先,根据系统动力学原理,构建基于AnyLogic的公共自行车系统仿真模型。然后,以租赁点的设施利用率和服务度为评价指标,利用敏感性分析和迭代相结合的方法,优化公共自行车租赁点的配车数。最后,以公共自行车系统仿真模型为实验平台,对比分析优化前后租赁点的评价指标。通过对比实验,结果表明优化后的租赁点有利于提升系统效率与服务质量。
[Abstract]:With the rapid development of urbanization, the contradiction between urban development and traffic congestion is becoming more and more prominent. Vigorously developing public transport is an effective means to alleviate this contradiction. In the choice of public transport, subway and bus have undoubtedly become the first choice, but neither subway nor bus can completely solve the end of the traffic problem, that is, it is impossible to send passengers directly to their destination. In public transport, public bicycles, as a means to solve the problem of terminal transportation, provide convenience for people to travel. However, due to the unreasonable setting of some public bicycle rental points in various areas of the city, There is often the phenomenon of "there is no pile lock and no pile lock", which makes it difficult for citizens to choose to use public bicycles to reach their destination. Therefore, it is an urgent problem to improve the service efficiency of public bicycle system and guide citizens to choose public transport. Based on the operation data of Lanzhou public bicycle rental point, through the analysis of the operation data, this paper puts forward the evaluation model of the rental point setting and lock pile configuration of the public bicycle system. The model evaluates the public bicycle system by improving the C4.5 classification algorithm, and according to the principle of system dynamics, constructs the simulation model of the public bicycle system based on AnyLogic, and takes the facility utilization rate and service degree of the rental point as the evaluation index. Sensitivity analysis and iterative method are used to optimize the number of public bicycle rental points. The research contents of this paper are as follows: (1) Statistical analysis of public bicycle system operation data. Based on the actual operation data of Lanzhou public bicycle system, the frequency of rental car, the frequency of return car, the number of lock piles and the length of vehicle use are counted by using R language, and the coupling coordination degree between the data is calculated. Combined with the traffic terrain characteristics of Lanzhou City, the effect grade of single lease point setting is determined. (2) the evaluation decision tree of the effect of public bicycle rental point setting is constructed. Firstly, the characteristics of common decision tree classification algorithms are analyzed, and the C4.5 classification algorithm is selected, and the calculation process of the algorithm is improved, and the improved C4.5 classification algorithm is used. The decision tree of public bicycle rental point setting and lock pile configuration evaluation is constructed, and the advantages and disadvantages of lease point setting are obtained quickly and intuitively. This part of the study provides the basis for the optimization of public bicycle rental points. (3) the simulation model of public bicycle system is constructed, and the scheme of optimizing the number of rental points is put forward. In order to solve the problems of low utilization rate and low service quality of bicycles in some public bicycle rental points, firstly, according to the principle of system dynamics, the simulation model of public bicycle system based on AnyLogic is constructed. Then, taking the facility utilization rate and service degree of the rental point as the evaluation index, the number of car allocation at the public bicycle rental point is optimized by using the method of sensitivity analysis and iteration. Finally, taking the simulation model of public bicycle system as the experimental platform, the evaluation index of rental point before and after optimization is compared and analyzed. Through the comparative experiment, the results show that the optimized lease point is beneficial to improve the system efficiency and service quality.
【学位授予单位】：兰州理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U491.225;N941.3

【相似文献】