Urban Emergency Vehicle Deployment Under Dynamic Traffic Environment

Emergencies in cities often bring about a large number of uncertain and diverse rescue demands.Urban emergency rescue work has the characteristics of short rescue cycles and complicated road conditions.In order to ensure the healthy and stable development of the city,decision makers demand to arrange emergency response points in advance in certain locations of the city according to the actual situation to ensure that emergency vehicles are dispatched to rescue points quickly,efficiently and accurately,and provide emergency response service.When deploying sites and vehicles,it is necessary to provide urban residents with more efficient and fair service coverage based on the changing traffic conditions in the area.In the actual rescue process,the demand point that the emergency vehicle should be responsible for cannot simply be determined by the straight-line distance,because the shortest path in space between two points may not be the optimal path in time.Therefore,according to the characteristics of the traffic conditions in different time periods,this dissertation deploys emergency vehicles in time periods,with the purpose of making each demand point available for rapid rescue provided by emergency vehicles deployed around it.The dissertation first constructs a dynamic deployment strategy of urban emergency vehicles,and describes the deployment goals of urban emergency vehicles based on the dynamics of the transportation system.In particular,this dissertation systematically analyzes and summarizes related work at home and abroad,and proposes a data-driven dynamic deployment mechanism for urban emergency vehicles.Then the city dynamic traffic is divided,and the speed changes of vehicles in different directions on each street are obtained.After getting the travel time of the vehicles in the area,it is analyzed,mainly through the clustering algorithm to find out the regularity of the travel time of the vehicles in different time periods and on different roads.The slider algorithm is used to obtain the mode change of the road network,where the critical point is the demarcation point of the mode transition,and the time period distribution of the road network in different clusters is obtained.Using different vehicle deployments for emergency sites in different time periods.At the same time,the dissertation studies the emergency vehicle deployment model and solution algorithm based on the results of the division of urban traffic operation status,considering the two indicators of efficiency and fairness.This dissertation builds an emergency vehicle coverage planning model that minimizes the total rescue time.In the solution of the model,considering that a day is divided into different periods,this dissertation further considers the selection of a multi-chromosome genetic algorithm with enhanced elite retention based on the traditional genetic algorithm Solve.For the solution obtained,the coverage rate is used to evaluate the efficiency of vehicle deployment,and the Gini coefficient is used to evaluate the fairness of the deployment plan.The dynamic deployment plan and the static fixed emergency vehicle deployment plan are compared with site coverage and fairness.The results show that the emergency mechanism and mathematical model designed in this dissertation have certain superiority and effectiveness.The significance of this dissertation is to propose a data-driven dynamic deployment framework for emergency vehicles,which is suitable for multiple types of emergency scenarios and can be used to plan emergency response systems for different service groups and site levels,and efficiently improve the pertinence of emergency services.Compared with the traditional static fixed location selection,the dynamic deployment mechanism of emergency vehicles proposed in this dissertation combines real traffic information during emergency preparations and provides a theoretical reference for the actual organization and operation.