Identification and Analysis of Factors Affecting Industrial Machinery Maintenance and Repair Costs Using the XGBoost Algorithm
Abstract
The primary objective of this study was to scientifically identify and rank the factors affecting industrial machinery maintenance and repair costs in Iran’s petrochemical industry. To achieve this objective, a mixed-methods approach comprising a systematic literature review, semi-structured interviews with 12 industry experts, and advanced statistical analyses was employed. The required quantitative data were collected from 10 industrial machines over a three-year period. During the analytical phase, statistical methods, including correlation coefficients and analysis of variance (ANOVA), were used to assess the relationships among the variables. Subsequently, the advanced XGBoost machine-learning algorithm and the Gain metric were employed to calculate the relative importance of the factors and establish their final ranking. The findings indicated that three factors—equipment age (18.9%), failure to implement preventive maintenance (17.2%), and adverse operating conditions (15.1%)—together accounted for more than 50% of the total importance. Managerial factors, including spare-parts procurement lead time (9.8%) and the level of CMMS utilization (8.5%), occupied the subsequent ranks.