Ferdowsi University of Mashhad Journal of Research and Rural Planning 2783-2791 15 2 2026 06 01 Ranking the Drivers of Agricultural Machinery Maintenance Costs and Their Impacts on Rural Economic Sustainability in Western Iran 95 130 48543 10.22067/jrrp.v15i2.2507-1172 EN Khalil Jaidari Qa, C., Islamic Azad University, Qazvin, Iran. 0009-0001-0550-3243 Sadegh Feizollahi Il, C., Islamic Azad University, Ilam, Iran. 0009-0000-4051-7950 Alireza Irajpour Qa, C., Islamic Azad University, Qazvin, Iran. 0009-0003-2666-7233 Journal Article 2026 06 09 <strong>Purpose-</strong> This study aims to identify and rank the key drivers affecting agricultural machinery maintenance and repair costs and to examine their implications for strengthening the economic sustainability of rural areas in western Iran.<br /><strong>Design/methodology/approach-</strong> This study employs a mixed-methods research design integrating qualitative and quantitative approaches. In the qualitative phase, a systematic literature review and semi-structured interviews with 24 agricultural mechanization experts were conducted to identify and validate the main factors influencing maintenance and repair costs. In the quantitative phase, real maintenance records of 80 agricultural machines operating in the provinces of Kermanshah, Ilam, and Kurdistan were collected over a three-year period, resulting in 240 machine-year observations. The Extreme Gradient Boosting (XGBoost) algorithm was then applied to model the relationship between explanatory variables and maintenance costs and to rank the relative importance of factors based on the Gain criterion.<br /><strong>Findings- </strong>The findings demonstrate that maintenance and repair costs are influenced by the simultaneous interaction of technical, managerial, operational, human, and environmental factors. The XGBoost model achieved satisfactory predictive performance on the testing dataset, with an R² value of 0.86, RMSE of 39.4 million IRR, and MAE of 28.7 million IRR. The feature importance analysis revealed that machine age (18.9%), lack of preventive maintenance implementation (17.2%), and unfavorable operating conditions (15.1%) were the three most influential determinants of maintenance and repair costs, jointly accounting for more than half of the model’s total importance. These findings indicate that machinery aging, insufficient maintenance planning, and improper operating practices represent the primary sources of increased maintenance expenditures. Furthermore, the results demonstrate that effective preventive maintenance programs, improved maintenance management systems, optimized spare parts management, and enhanced operator practices can reduce life-cycle machinery costs and contribute to rural economic sustainability.<br /><strong>Research limitations/implications-</strong> This study is limited by the geographical scope of the investigated provinces and the availability of historical maintenance records from agricultural machinery. Future research may expand the analysis to other regions and incorporate larger longitudinal datasets, economic variables, and advanced predictive approaches to further improve the understanding of machinery maintenance cost dynamics.<br /><strong>Originality/value-</strong> This study contributes to the agricultural mechanization and rural sustainability literature by integrating expert-based knowledge, real machine-year maintenance records, and machine-learning techniques within a unified analytical framework. Unlike previous studies that mainly relied on conventional statistical methods or examined individual factors separately, this research applies the XGBoost algorithm to quantify the relative importance of maintenance cost drivers and links machinery maintenance management with rural economic sustainability. https://jrrp.um.ac.ir/article_48543_7a770b232ea56afe9641bb93fb3c2c6b.pdf