Study On Key Technologies Of Urban Traffic Police Deployment And Dispatch

With the rapid development of urban road traffic system,the contradiction between the lack of traffic police and traffic congestion are becoming more and more serious.It brings great pressure and severe challenges to traffic management departments.It is the improvement target of management level and work efficiency,and the main content studied and explored by scientists,to reform the commonly extensive traffic police duty mode,and realize scientific,standardized and meticulous management.The problems and solutions about identifying designated post and dividing responsible zones of traffic police deployment,and dynamically dispatching traffic police for emergency are studied,by using the complex network,modem comprehensive evaluation method,linear programming,artificial intelligence algorithm and other related theories and methods.The thesis focus on the method to identify police designated post,the mathematical models and algorithm for grid service management,and the key technologies for dispatching dynamic traffic police based on police duty behavior.The following are the research results.(1)The evaluation index system and comprehensive evaluation method for critical intersection nodes of traffic network are proposed.According to the dynamic and static factors influencing the identification of police designated post,two kinds of traffic network model about topological structure and running status are built with urban traffic data.Then different kinds of key indexes of intersection nodes are put forward for the practical significance of the indexes in the models,and a set of comprehensive evaluation index system of intersection node critical is established,which reflects various features to identify police designated post.Subsequently,TOPSIS-RBF combination evaluation method is studied to comprehensively evaluate the multiple indexes to get the intersection node critical results.Finally,the experiments build the topological structure and running status traffic network model by using Beijing's traffic network data,prove the feasibility of index calculation methods of the index system,and validate the effectiveness and advantages of the two evaluation method.(2)The method to divide responsible zones is put forward base on response time in the police grid duty management.At first,two kinds of basic binary covering models(Set Covering Model,SCM and Maximum Covering Model,MCM)are described to solve the problems of traffic police deployment with limited and unlimited resource in the police grid duty management,while the Extended Set Covering Model(ESCM)based on coverage degree are established.Secondly,the relevance is revealed that the number of police limited in MCM should be less than the amount of full coverage of the SCM.Then,MCM and ESCM are the models for solving the coverage problem under different conditions,which can be combined to form the Two-Objective Linear Programming Model(TOLPM).Based on the extension of binary covering models,the dynamic covering models of MCM and ESCM are constructed with temporal and spatial characteristics,in which the temporal variables are discontinuous.Next,the binary covering models and dynamic covering models based on deployed polices are put forward.Afterward,the linear programming and genetic algorithm for solving the models above are described,and the optimization method of key steps of genetic algorithm is discussed to solve different MCM and ESCM quickly.At last,the implementation of genetic algorithm and the calculation experiments of traffic data validate the content of the research.(3)A traffic police dispatch method based on duty status is presented,in which the key problem of duty behavior analysis and dynamic resource dispatch are studied.The traffic police dispatch method makes use of the duty behavior state of the police,which is obtained by the analysis of police duty and position information,to search and dispatch police effectively,combined with the event location and traffic information.The next is presenting a method to analyze the traffic police duty behavior based on location information,in which the automatic identification of the duty behavior improves the effectiveness of dynamic resource dispatch.Then the shortest path optimization algorithm with the ability of error control is proposed to adapt to the characteristics of dynamic traffic network,which can quickly search the most effective police and plan the fastest rescue route.The concrete implementation steps and error control methods of the algorithm are described,and the method of calculating error control parameters by using Nonlinear Interpolation is proposed.And the inverse Dijkstra algorithm is put forward to search for recent police resources.Finally,the experiments of the optimization algorithm validate the efficient increase of the algorithm.The experimental results show that the error is caused by the limited search range,and can be controlled by the amplification factor of the algorithm.To summarize,the results of this thesis can enrich the basic theory and technologies of traffic police management in our country,that provide a theoretical basis for the optimization and reformation of urban traffic police duty mode.It affords us technical means for the reasonable deployment and emergency dispatch of traffic police,which is a technical foundation for the construction of intelligent traffic duty management system.This work has great certain theoretic value and practical significance to establish a scientific,efficient and intelligent traffic duty management mechanism.