"background": "Transport depot maintenance systems are critical for operational efficiency and economic yield in developing economies. However, systematic methodological evaluations of these systems, particularly those quantifying their impact on yield, are scarce in the literature. ", "purpose and objectives": "This study aims to methodologically evaluate existing maintenance systems in Tanzanian transport depots and quantify their causal effect on yield improvement through a multilevel regression framework. ", "methodology": "A cross-sectional design was employed, collecting primary data from a stratified sample of depots. The core analysis used a three-level mixed-effects model: Y{ij = \0 + \1 Xij + uj +, where Yij is yield for depot i in region j, Xij is a vector of maintenance system predictors, uj are region-level random effects, and is the error term. Inference was based on robust standard errors. ", "findings": "The multilevel analysis revealed that a depot's adoption of a computerised maintenance management system (CMMS) was positively associated with yield, with a coefficient of 0. 15 (95% CI: 0. 09, 0. 21). This indicates a statistically significant improvement, accounting for regional heterogeneity. The methodological evaluation identified spare parts inventory control as the most inconsistently applied subsystem. ", "conclusion": "The methodological framework confirms that structured maintenance systems, particularly CMMS integration, are significant drivers of yield improvement. The multilevel approach effectively captured the nested data structure inherent in depot operations. ", "recommendations": "Depot managers should prioritise investment in CMMS technologies. Policymakers are advised to develop guidelines for standardising inventory control procedures to reduce systemic variance identified in the evaluation. ", "key words": "Maintenance systems, multilevel regression, yield improvement, transport depots, infrastructure management", "contribution statement": "This paper provides a novel methodological framework
Amina Mwinyi (Sun,) studied this question.