不确定信息环境下的公交调度鲁棒优化及其控制策略

项目名称： 不确定信息环境下的公交调度鲁棒优化及其控制策略

项目编号： No.61473122

项目类型： 面上项目

立项/批准年度： 2015

项目学科： 自动化技术、计算机技术

项目作者： 靳文舟

作者单位： 华南理工大学

项目金额： 83万元

中文摘要： 城市公交调度是一个复杂的离散事件大系统，面临着客流需求、行程时间、车次延误时间等随时空波动的不确定因素。为解决目前对不确定环境下公交调度相关研究方法的缺陷，本课题从计算机模拟的角度，应用鲁棒优化理论，从单线到多线的公交班次延误扩散机理解析切入，构建公交调度鲁棒性评价指标，以提高公交调度计划的抗干扰能力为目标，建立基于数学规划的公交调度鲁棒优化模型；搭建公交调度离散事件仿真平台，通过两阶段仿真研究置信区间下不同随机干扰阈值区间鲁棒性较好的公交调度方案及其特征参数值，探讨不同控制策略下各仿真性能评价指标与延误特征系数的关系，并给出最优控制策略组合；结合我国国情分析成本预算限制、运力约束等资源供给对公交调度鲁棒性的影响，给出运营管理资源优化配置的建议；以广州市第二公共汽车公司和澳门新福利公交公司为试点，促进研究成果的进一步完整和推广应用。

中文关键词： 不确定环境；公交调度；鲁棒优化；控制策略；离散事件仿真

英文摘要： City bus scheduling is a complex large scale discrete event system, facing the uncertainties such as passenger demand, journey time, and bus delay time that fluctuates with space and time. In order to overcome the defects in the current research methodologies for the bus scheduling under uncertain information environment, from the perspective of computer simulation，this project starts from exploring the spreading mechanism of the single-line and further multi-line bus-delay， establishing the index of bus scheduling robustness evaluation with robust optimization theory. Aiming to improve the anti-interference ability of bus scheduling, bus schedule robust optimization models based on mathematical programming are established. Furthermore, a discrete event simulation system platform is developed，and through two-stage emulating, the bus scheduling schemes with better robustness are obtained, as well as the characteristic parameters in different random disturbance threshold values within the confidence intervals. And then the relationship between simulation performance evaluation index and delay characteristic coefficient in different control strategies is investigated，which provides the optimal control strategy combination. Considering our national conditions，the impact of resource supply such as budget and fleet size constraints will be analyzed，and the advice on optimal configuration of operation resources will be given accordingly. These research outcomes will be illustrated for Guangzhou second and Macao Xinfuli bus company for further improvement and popularization.

英文关键词： uncertainty;bus schedule;robust optimization;control strategy;discrete event-based simulation