Maintenance Method of Logistics Vehicle Based on Data Science and Quality

With the changes in consumption around the world, the global logistics and logistics management has been developed, which has derived the business opportunities of freight logistics and the demand for vehicles, which has led to the increase of vehicles service and service parts . Therefore, effective remaining vehicle readiness and reduction in maintain costs have become an urging subject to be solved in the industry of today. However, as in the era of big data, it is important that the enterprise makes good use of data and information to save costs, increase revenue, and ensure competitive advantages. But if we could not ensure the quality of the data, it would easily lead the analysis to the wrong decisions. Therefore, this study is based on the predictive maintenance, taking the condition of data quality considerations, and using algorithms to construct a decision support model, and proposing optimal replacement cycles and rules for vehicle components, and analyzing the impact on brands and maintenance amounts. Therefore, this study is based on maintenance history, through systematic and manual analysis, to obtain good quality data, and then use chi-square test and algorithm analysis to establish a classification model for decision support. The research department analyzes and analyzes the 3.5 ton freight vehicle maintenance and repair history of a case company from 2008–2016. After the data is cleaned and sorted, it obtains 173,693 work orders and good data quality data for 23 types of maintenance items. And the results show that: the costs contains significant divergence among brands; service parts damage is related to particular environment; we can obtain appropriate service period through proper classification rules. The decision support model constructed by this study will be improved and integrated with the actual needs of the industry on the premise of taking into account the quality of data.