Research And Implementation Of Public Service Platform For Motor Vehicle Maintenance Industry Based On Hadoop Technology

With the rapid devedopment of China's economy and society,the current manufacturing industry and industry are also showing rapid development momenturn,especially in the motor vehicle manufacturing industry.In addition to traditional foreign-funded vehicles and joint-venture vehicles,many domestically produced vehicles have emerged.The popularity of automobiles has brought conveniencc to the transportation industry.It has gradually become the main means of transportation for peoples daily travel.The continuous growth of car ownership has brought huge aftermarket services.With the increasing demand from the user groups for vehicle maintenance and repair services,the resource docking between vehicle owners and maintenance companies is becoming more and more difficult.In order to better adapt to the development,the automotive after-market service is continuously transformed and upgraded,and the publie service for motor vehicle maintenance industry is developed.The plalform has very practical implications.This thesis develops a public service platform that based on Hadoop technology for motor vehicle maintenance industry to provide users with more convenient,reliable and efficient vehicle maintenance and repair services.The main work of the thesis is as follows:(1)The thesis designs a hierarchical structure of the public service platform for the motor vehicle maintenance industry that based on the relevant theoretical and practical research on traditional automotive aftermarket services in the automotive industry,which including infrastructure layer,data acquisition layer,service support layer and system application layer.Finally,a brief introduction the key technologies and related technology theory involved in the system.(2)The thesis designs and develops a high performance communication service that based on the research of related principles such as network communictation principle and high concurrency principle for the communication resource bottleneck causced by high data and high concurrent demand in the system.The communication server involves technologies such as pooling technology,dynamic buffer technology and message queuing.Through the cooperation these key technologies,it' s verified has good performance at the some relatively indicators such as the network throughput and response time by Apache's JMeter.(3)A smart recommendation algorithm has been added to the platform to enable resource docking for vehicle owners and maintenance companies.The thesis researches and improves a recommendation algorithm based on ItemRank which based on the research of collaborative filtering recommendation algorithm.Through experimental comparison according to the strictly follows the recommendation algorithm evaluation standard.,the improved algorithm has better index parameters than CF recommendation algorithm it can provide personalized intelligent recommendation service for the owner group of vehicle more accurately.(4)Finally,the public service platform is designed and developed for the motor vehicle maintenance industry,which based on Hadoop technology and the research on the above related theories.The platform provides Web-side applications and APP-side applications according to different user roles and usage scenarios,both applications use the Spring Boot framework and strictly follow the Restful style.which use the front-end and split-side modes help to improve the application decoupling of the system,and basically able to meet the relevant functional needs of users of all types of roles.