城市接驳交通场景下的多智能体模型研究  

Multi-agent Model in Urban Intermodal Transportation Scenarios

作　　者：潘理虎[1] 杨楠 张林梁 张睿 谢斌红[1] 闫慧敏[3] PAN Li-hu;YANG Nan;ZHANG Lin-liang;ZHANG Rui;XIE Bin-hong;YAN Hui-min(School of Computer Science and Technology,Taiyuan University of Science and Technology,Taiyuan 030024,China;Shanxi Intelligent Transportation Research Institute Company Limited,Taiyuan 030036,China;Institute of Geographic Sciences and Natural Resources Research,CAS,Beijing 100101,China)

机构地区：[1]太原科技大学计算机科学与技术学院,山西太原030024 [2]山西省智慧交通研究院有限公司,山西太原030036 [3]中国科学院地理科学与资源研究所,北京100101

出　　处：《计算机技术与发展》2025年第4期149-155,共7页Computer Technology and Development

基　　金：山西省自然科学研究面上项目(202203021221145)。

摘　　要：随着城市交通系统日益复杂,接驳交通已成为优化交通出行的重要策略。接驳交通系统主要受到公共交通供给和停车设施不足的限制。本研究基于多智能体复杂系统,构建了一个应用于接驳交通场景下的多智能体模型,旨在研究分析不同管理策略下的交通指标,为城市交通管理者提供决策支持。该模型采用分布式并行框架开发,以太原市小店区的高密度出行区域为例,缩短公交发车间隔可将接驳交通吸引力提升至25.79%,明显高于调整公交票价方案下的19.19%;适度增加预约车位比例可使公共交通承担率达到49.85%,进一步提高公共交通在日常出行中的参与度;组合策略的协同作用则可将公共交通承担率提升至50.62%,同时将接驳交通吸引力提高至33.6%,充分体现出多种策略联合实施的综合优势。As urban transportation systems become increasingly complex,intermodal transportation has emerged as an important strategy for optimizing travel.However,intermodal transportation systems are primarily constrained by insufficient public transportation supply and limited parking facilities.We developed a multi-agent model for intermodal transportation scenarios based on a complex multi-agent system,aiming to analyze various traffic indicators under different management strategies and provide decision-making support for urban transportation managers.The model was implemented by a distributed parallel framework.Taking the high-density travel area of Xiaodian District,Taiyuan City,as an example,shortening bus intervals could increase the attractiveness of intermodal transportation to 25.79%,significantly higher than 19.19%under the adjusted bus fare strategy.Moreover,moderately increasing the proportion of reservable parking spaces could raise the public transportation utilization rate to 49.85%,further enhancing its role in daily travel.The synergistic effects of combined strategies could further increase the public transportation utilization rate to 50.62%and improve the attractiveness of intermodal transportation to 33.6%,demonstrating the comprehensive advantages of implementing multiple strategies jointly.

关 键 词：多智能体系统 接驳交通 并行仿真 交通管理 策略评估 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]