The Time Prediction For Complete Maintenance Of Commercial Vehicle Basic Clustering Analysis

Complete maintenance of commercial vehicle is one of the motor vehicle maintenance management system in China.In the daily maintenance,primary maintenance and complete maintenance,the complete maintenance is one of the highest level.It is to check and adjust the braking system,steering control system,suspension and other safety components,removal of tires,tire transposition,check the adjustment Engine technology and vehicle emissions-related systems-based maintenance operations.Complete maintenance must be carried out by a qualified service technician from a vehicle maintenance company licensed by the road transport authority.The ministry of communications first order in 2016,the maintenance of secondary operations by the vehicle every three months or 15,000 km for a road transport operators based on the specific circumstances of the vehicle to determine their own maintenance cycle.Therefore,how to determine the two-stage maintenance cycle through the specific situation of the vehicle has become an economic value of the research.Data mining through a large number of historical data for calculation and analysis to find hidden in the data which is not easy to find the law and knowledge.Data mining commonly used methods with classification prediction,clustering,association rules and time series.As one of the common methods of data mining,clustering analysis is a process of partitioning data objects into clusters.Data objects within clusters are as similar as possible,while data objects between clusters are as different as possible.With many years of research and development,clustering analysis has been widely used in business decision-making,image recognition,data compression and other fields.Clustering analysis techniques can be further divided into subdivided,density,hierarchical,grid and constraint methods.In order to solve the problem of forecasting the complete maintenance time point of the dynamic and effective operating vehicle,this paper uses the k-means clustering algorithm to classify the operational vehicle data objects by the complete maintenance time point forecasting of the operating vehicles,and divides the data into five classes,and then through vehicle quality cluster and time mapping to form the next complete maintenance vehicles operating point of time for maintenance forecast.In this process,200 sets of experimental data and 20 sets of test data used in the experiment were provided by some transport enterprises and some maintenance enterprises with complete maintenance qualification by one city in Jiangxi Province.