Research On The Matching Method Of Vehicle And Cargo Information In Streaming Data Environment

With the rapid development of social economy and science and technology,the modern logistics industry is moving towards an intelligent,systematic and efficient development path.The face of freight forwarding business due to the source of information asymmetry caused by the vehicle idle and empty driving and other issues,all kinds of logistics information platform has mushroomed came into being,to promote the car source and supply information circulation and sharing,But as a logistics information platform,one of the core functions of vehicle information matching method,its related research is still in its infancy,not yet perfect.Most vehicle information matching methods mainly focus on "active information retrieval"and "establishing a system for evaluation of offline vehicles to evaluate" to achieve the matching of vehicle information.However,on the application level,the existing matching methods only focus on large-scale Long-distance transport of goods vehicles and goods docked in the freight scene,failed to meet the diversity of modern logistics freight transport,but also failed to meet the new requirements of modern logistics intelligent,real-time,sophisticated,multi-scene adaptation.Therefore,under the background of the new demand of modern freight logistics,this paper studies on intelligent and real-time vehicle information matching method and has great theoretical and practical value to promote the modernization of freight logistics.On the basis of summarizing the research background and related research status,this paper,based on the theory of streaming data and Storm real-time computing framework,k-Means clustering algorithm and genetic algorithm as the basic technology and method,Vehicle Information Matching Method in Data Environment.The research of this dissertation mainly includes the following aspects:By studying the concept of vehicle information matching problem,this paper studies the combinatorial optimization mathematical model of vehicle information matching problem.Then based on the genetic algorithm to solve the vehicle information matching problem,the matrix coding method is used to carry out chromosome Designed and proposed a multi-level punishment-the restoration strategy to deal with constraints,the use of adaptive crossover operator for cross-operation,put forward Hadamard product mutation operator for chromosome variation,improve measures to improve the efficiency and accuracy of the results;based on the study of matching algorithms,the use of Storm set up the framework of vehicle information matching method in streaming data environment,and put forward the steps of method implementation.At last,we verify the validity and algorithm performance of improved genetic algorithm through Java programming experiment.Secondly,based on Java-Storm programming verification streaming data environment The method of vehicle information matching is effective,and the method maintains good efficiency in continuous flow data verification operation.