Research On The Cooperative Optimization Of Freeway Patrol Vehicle Route And Vehicle Configuration

With the rapid growth of China's economy and the improvement of people's living standards,the number of per capita vehicles is also increasing day by day.Traffic accidents lead to traffic congestion,casualties and secondary accidents on freeways,which bring tremendous pressure and severe challenges to the freeway traffic management departments.The increase of traffic accidents and the lack of traffic police resources pose great challenges to the traditional extensive traffic service mode.Most of the previous studies on patrol site selection,route selection and vehicle configuration consider the optimization of the three factors separately,and the combinatorial studies are rarely.At the same time,the invalidation problem is not considered.How to set facility location and allocate police resources scientifically and rationally,schedule police quickly and efficiently under the condition of limited police resources,to shorten the accident response time,minimize traffic delay and loss of life.It is a main goal which the freeway traffic management department to improve their management level and work efficiency.At the same time,it is the research direction that relevant scientific research personnel studies and explores hard.This paper conducts an in-depth study on the collaborative optimization of patrol route and vehicle configuration in the context of traffic accidents in the freeway network,especially in the case of multiple traffic accidents.The specific research contents are as follows:(1)The two interrelated components in a freeway patrol service program,namely beat design and fleet allocation,are addressed in a single integrated model,with the objective of minimizing the overall average incident response time.And the traditional expressway patrol service is further studied.On the basis of reviewing the modeling of vehicle route optimization problem,the section of fressway network is taken as the demand point,and the freeway patrol route problem is transformed into an m TSP problem,then a traditional patrol model is built according to m TSP.The model is highly nonlinear and non-convex mixed integer optimization problem.This paper attempts to transform the model into a linear problem.Take Sioux Falls the classic road network as an example,the patrolling route is encoded by the orderly integer coding method,and the highly nonlinear and non-convex mixed integer optimization model is solved by the improved genetic algorithm.At the same time,the result with scenario algorithm,large-scale neighborhood search algorithm,simulated annealing algorithm and double-base greedy algorithm are compared,and the programming performance and efficiency of genetic algorithm are analyzed.(2)The model and method of freeway overlapping patrol mode path selection and vehicle configuration collaborative optimization are studied.In this paper,the restriction that each road section can only be patrolled by one route is relaxed,and multiple routes are allowed to patrol a certain road section or some road sections at the same time.In accordance with the fairness principle,a nonlinear mixed integer complete coverage model is established,with the objective of minimizing the overall average incident response time.The genetic algorithm is coded by combining ordinal code with matrix code to achieve crossover,mutation and other genetic operations.The route connectivity is ensured by using the split technology,and the vehicle allocation algorithm is solved by LINGO.The algorithm was applied based on the Sioux Falls classic directed road network,and the results were compared with the results of non-overlapping patrol mode solved by genetic algorithm and simulated annealing algorithm.(3)A patrol in a rescue mission state,the demand on other sections which is patrolled by the patrol vehicle will be not met under multiple traffic accident scenarios.For this problem,this paper put forward one part of the patrol vehicle is parked in the optimized area on standby,another part of the patrol vehicle is responsible for patrol back and forth.The joint patrol model of freeway patrol route and vehicle configuration collaborative optimization in case of multiple accidents is studied in order to minimize the overall average incident response time under the optimal route and minimum manpower material resources.Based on Lagrange relaxation algorithm,branch and bound algorithm,shortest circuit algorithm,dynamic programming algorithm and genetic algorithm,the model is solved.And the feasibility of patrol model and method is discussed through the case study,and the analysis results are given.(4)Taking part of the freeway network in Jilin province as an example,a case study on the cooperative optimization of patrol route and vehicle configuration based on the overall average incident response time was carried out.The overall average incident response time is compared and analyzed,by combining the current jurisdiction with the status of vehicle configuration of highway administration division or traffic police department.Based on the research results,the strategy proposals for related freeway management departments were provided.Based on relevant research results at home and abroad,this paper takes investigation statistics and optimization modeling as research methods to deeply dig and study the rapid response of traffic accidents in freeway network.On one hand,at the theoretical level,it has certain innovation and continuity significance,and on the other hand,at the practical application level,it has important practical significance to improve the response speed of traffic accidents and the scientific practice level of freeway management.