Development of Data Mining Approaches to Planning Predictive Maintenance of Defensive Light Vehicles

Vehicles are considered as one of the most demanded equipment in a defense organization, if they are not serviced in a timely and efficient manner, they are considered as a threat to that organization. On the other hand, due to the significant increase in data related to maintenance of these devices in the organization under study, in this study, data mining technique has been used to increase the data conversion into information rate and knowledge discovery in order to prevent cease of missions .The purpose of this paper is to present the architecture of classifying the data generated from the reference and breakdowns of passenger cars of the studied organization and to predict the distances between their breakdowns through data mining method in order to plan their maintenance and repairs.

Based on this, the required data for a sample including 150 light vehicles were extracted. In this research, year of construction of the car, distance traveled and cause of failure as input variables and the intervals between failures as output variables were defined.

Based on data mining principles and using SPSSModeler software, vehicles were grouped based on the distances between breakdowns (km) with the C&R tree algorithm.

according to the outputs analysis, the maintenance and repair group of the studied organization should adjust the service and inspection plan of the vehicles based on the grouping done in the research and by considering the suggestions provided.