Research On Data Analysis Model Of Bus Maintenance Based On Sequential Pattern Mining

Bus as one of the important part of urban public transportation, plays a very important role, and it mutually complementary and interdependent with railway. In order to provide convenient, comfortable and efficient travel services for passengers, one of the important prerequisite is to ensure the vehicles are under a good quality and condition. Such a request, depends on the public transport service enterprises to transport vehicles have a very good maintenance and repair services. On the other hand, the bus vehicle maintenance in bus enterprise management cost as high as 25%, so in ensuring good vehicle quality condition at the same time, also had to consider the maintenance cost. Vehicle daily maintenance will produce a lot of data, by relying on data mining technology and the use of methods and algorithms, some of the hidden knowledge will be found to optimize the daily maintenance and develop differentiated maintenance programs, which will find a reasonable balance in the quality of maintenance, repair efficiency and maintenance costs so as to improve the quality of passenger service and improve the economic efficiency of the enterprises.In order to solve the problems, firstly, through analysis of the composition and characteristics of the bus vehicle maintenance information management system to generate data, combined with the actual needs of the bus vehicle maintenance enterprise management, and puts forward the bus vehicle maintenance data mining model. Then it introduces the related theories, methods and algorithms of sequential pattern mining, and analyzes the advantages and disadvantages of each algorithm. After comparison and analysis, this paper selected the Apriori algorithm and FP-Growth algorithm as the data mining algorithm of the above model, and in the fourth chapter, the two algorithms are described and implemented in detail. Then in the fifth chapter proposed the data mining model of technology roadmap, and definitions of technology in the circuit diagram, data preparation, data selection, data pre processing, data conversion, data mining, visualization model rules and knowledge base generation of each step are described in detail, introduces the bus vehicle maintenance data mining analysis model of the implementation process. Then, according to the data mining model, the corresponding knowledge database is generated, which can be used in the information management system of bus maintenance. The operation of the algorithm results were detailed comparative analysis, indicating that the FP-Growth algorithm compared with Apriori algorithm both in the efficiency of the algorithm, or overhead compared, better than a lot of. Finally, the thesis on the bus vehicle maintenance data mining model is described. That means the model not only the vehicle repair data can dig out the reliable knowledge database, but also on the data of vehicle maintenance and spare parts using data mining. The conclusion part of the thesis, from the hierarchy of the model design, mining algorithm, technical route map and a knowledge database using were evaluated, and get the data mining model can basically meet the bus service enterprises use the conclusion.