当前位置:主页 > 科技论文 > 路桥论文 >

公共自行车系统自然租赁需求预测与多目标调度方法

发布时间:2018-08-26 16:38
【摘要】:公共自行车系统作为破解公共交通“最后一公里”难题的交通方式,能有效提升城市公共交通整体服务水平,但公共自行车系统在迅速发展的同时,也存在着经常出现的“无车可借”与“无法还车”问题。因此,为解决出行者租/还车难问题,需要获取公共自行车系统服务点的实时状态信息及调度需求信息,设计合理的调度计划,均衡分配各服务点的公共自行车数量。本论文以浙江省科技计划项目——公共自行车运行数据的空间分析及其调度系统的研究与开发(2013C33047)为依托,研究了公共自行车系统的自然租赁需求与调度模型及求解算法。首先提出了公共自行车系统自然租赁需求的概念,并且给出了这种租赁需求的预测方法。通过Multi-logit改进模型估算公共自行车在客运总量中的分担率,利用模型估算所得分担率与调查统计所得到分担率的相关性,结合服务点历史租赁数据,实现公共自行车系统自助服务点自然租赁需求的预测。服务点自然租赁需求是公共自行车系统调度的前提,决定了服务点提供租还服务的能力和服务点的调度需求类型及调度需求量,对提高公共自行车系统的调度计划质量起着非常重要的作用。其次,从不同角度对公共自行车系统调度问题进行了分析,将公共自行车调度问题界定为公共自行车自流动与机动车调度相叠加的复杂动态车辆调度问题。通过对公共自行车系统服务点调度优先级、调度需求量、服务时间窗约束等特性进行的研究,建立了统筹用户满意度与企业调度成本的公共自行车系统多目标动态调度模型。同时,设计了禁忌搜索算法与遗传算法相结合的混合算法对所建立的调度模型进行求解,得到公共自行车最优调度方案。最后,通过C#语言、Sql Sever 2008数据库,开发了基于客户端/服务器(C/S)架构的公共自行车系统调度软件,并验证了本文所提出的公共自行车系统多目标调度方法。结果表明,调度方法在平衡各服务点公共自行车数量方面具有预期效果,能够达到缓解租/还车难问题的目的。
[Abstract]:As a way to solve the "last kilometer" problem of public transport, the public bicycle system can effectively improve the overall service level of urban public transport, but the public bicycle system is developing rapidly at the same time. There are also frequent problems of "no car to borrow" and "unable to return the car." Therefore, in order to solve the problem, we need to obtain the real-time status information and scheduling requirement information of public bicycle service points, design a reasonable scheduling plan, and distribute the number of public bicycles in each service point evenly. Based on the spatial analysis of public bicycle operation data and the research and development of scheduling system (2013C33047), a scientific and technological project in Zhejiang Province, this paper studies the natural rental demand, scheduling model and solution algorithm of public bicycle system. Firstly, the concept of natural lease demand of public bicycle system is put forward, and the forecasting method of this kind of rental demand is given. The Multi-logit improved model is used to estimate the share rate of public bicycle in the total passenger transportation, and the correlation between the share rate estimated by the model and the share rate obtained by the investigation and statistics, and the historical rental data of the service point are combined. Realize the forecast of the demand of the natural lease of the self-service point of the public bicycle system. The demand for natural lease of service points is the premise of public bicycle system scheduling, which determines the ability of service points to provide rental and return services, and the type of scheduling requirements and scheduling requirements of service points. It plays a very important role in improving the quality of the scheduling plan of the public bicycle system. Secondly, the public bicycle scheduling problem is defined as the complex dynamic vehicle scheduling problem which is superposed by the self-flow of the public bicycle and the motor vehicle scheduling. By studying the characteristics of service point scheduling priority, scheduling demand and service time window constraints of public bicycle system, a multi-objective dynamic scheduling model for public bicycle system is established, which can balance user satisfaction with enterprise scheduling cost. At the same time, a hybrid algorithm combining Tabu search algorithm and genetic algorithm is designed to solve the scheduling model, and the optimal scheduling scheme of public bicycle is obtained. Finally, through C # language SQL Sever 2008 database, the public bicycle system scheduling software based on client / server (C / S) architecture is developed, and the multi-objective scheduling method proposed in this paper is verified. The results show that the scheduling method has the expected effect in balancing the number of public bicycles in various service points and can alleviate the problem of renting / returning vehicles.
【学位授予单位】:浙江工业大学
【学位级别】:硕士
【学位授予年份】:2015
【分类号】:U491.225

【参考文献】

相关期刊论文 前1条

1 柳祖鹏;李克平;朱晓宏;;基于蚁群算法的公共自行车站间调度优化[J];交通信息与安全;2012年04期



本文编号:2205512

资料下载
论文发表

本文链接:https://www.wllwen.com/kejilunwen/daoluqiaoliang/2205512.html


Copyright(c)文论论文网All Rights Reserved | 网站地图 |

版权申明:资料由用户d4d29***提供,本站仅收录摘要或目录,作者需要删除请E-mail邮箱bigeng88@qq.com