基于多智能体的跨区域综合交通应急调度  

Trans-Regional Comprehensive Traffic Emergency Dispatch Based on Multi-Agent System

作　　者：王晓阳 杨廷宇 李林谦 徐长安[1] 倪少权[1,3,4] WANG Xiaoyang;YANG Tingyu;LI Linqian;XU Changan;NI shaoquan(School of Traffic&Transportation,Southwest Jiaotong University,Chengdu 610031,China;China Railway Second Engineering Group Co.,Ltd.,Transportation and Urban Planning and Design Institute,Chengdu 610031,China;National and Local Joint Engineering Lab for Integrated Transportation Intelligence,Chengdu 610031,China;National Engineering Lab of Comprehensive Transportation Big Data Application Technology,Chengdu 610031,China)

机构地区：[1]西南交通大学交通运输与物流学院,四川成都610031 [2]中铁二院工程集团有限公司交通与城市规划设计研究院,四川成都610031 [3]综合交通运输智能化国家地方联合工程实验室,四川成都610031 [4]综合交通大数据应用技术国家工程实验室,四川成都610031

出　　处：《综合运输》2021年第9期27-33,共7页China Transportation Review

基　　金：国家自然基金项目(52072314;61703351);四川省科技计划项目(2020YFH0035,2020YJ0268,2020YJ0256,2020JDRC0032);成都市科技项目(2019-YF05-01493-SN,2020-RK00-00036-ZF,2020-RK00-00035-ZF);中国铁路总公司科技研究计划项目(P2018T001,2019KY10)。

摘　　要：针对突发事件下不同区域间无法有效配合的问题,引入多智能体技术对跨区域综合交通应急调度进行研究。首先,把受灾点和广泛分布于全国范围内的应急物资储备中心分别模拟为需求点智能体和出救点智能体,进而构建考虑运载工具能力限制的出救点智能体应急联运路径规划模型,设计并行蚁群算法进行求解,使用蚁群算法对比设计算法的有效性,最后进行算例分析。研究表明,基于本文模型和算法,决策者可以在满足救援时效性的要求下筛选出时间和费用成本最少的应急物资运输路径,为跨区域应急救援提供决策支持。Aiming at the problem of ineffective coordination between different regions under emergencies,multi-agent technology is introduced to study the cross-regional comprehensive transportation emergency dispatch.First,the disaster-affected points and the emergency material reserve centers widely distributed across the country are simulated as demands.Point agent and rescue point agent,and based on this,the rescue point agent’s emergency intermodal transportation route planning model considering the limitation of the vehicle capacity is constructed,and then a parallel ant colony algorithm is designed to solve it,also the ant colony algorithm is used to compare the effectiveness of the designed algorithm,finally an example is conducted.Results shows that based on the model and algorithm in this paper,decision makers can screen out the emergency material transportation path with the least time and cost under the requirements of rescue timeliness,and provide decision support for cross-regional emergency rescue.

关 键 词：跨区域 多智能体 应急联运 并行蚁群算法 路径优化 

分 类 号：U111[交通运输工程]