The Dodoma Urban Water Supply and Sanitation Authority (DUWASA) faces critical maintenance challenges in efficiently managing water pumping systems across the region, resulting in reactive maintenance decisions that reduce system availability to 65% compared to the industry standard of 95% (World Bank, 2023). This study developed a comprehensive maintenance management model to address optimal resource distribution and enhance water pumping system availability through evidence-based decision-making frameworks (Ahmad & Kamaruddin, 2012). The research employed a mixed-methods design utilising stratified random sampling of 65 pumping systems from 77 total systems across multiple operational zones. Comprehensive data collection involved questionnaires, structured interviews, data logger monitoring, and detailed system condition assessments focusing on seven critical parameters: ambient temperature variations, voltage fluctuations, equipment age, water mineral content (salinity), availability of spare parts, suspension system failures, and seasonal changes in water levels (Kiliç et al., 2017). Relative Importance Index (RII) analysis revealed ambient temperature variations as the most significant factor (RII = 0.874), followed by voltage fluctuations (RII = 0.846) and water mineral content (RII = 0.837). Multiple regression analysis generated a robust predictive model with strong statistical performance (R² = 0.699), indicating that the seven technical factors explain approximately 70% of system availability performance variance (Sharma & Srivastava, 2018). The resulting regression equation: Water Pumping System Availability = 0.990 - 0.220(Ambient Temperature) - 0.010(Voltage Fluctuations) - 0.410(Equipment Age) - 0.110(Water Mineral Content) + 0.210(Spare Parts Availability) + 0.670(Suspension System Management) + 0.080(Seasonal Water Level Management) provides a quantitative framework for maintenance decision-making. The developed Water Pumping System Maintenance Management Model (WPSMM) represents a computerised application featuring system inventory management, condition monitoring protocols, maintenance team coordination, and performance monitoring capabilities (Poór et al., 2020). System validation across 12 months of operational data demonstrated effective performance prediction with an overall accuracy of 89.48% compared to the actual availability of 90%, confirming the model's practical applicability. This research contributes a context-specific, resource-constraint-aware maintenance framework that enables evidence-based maintenance decisions, facilitating the transition from reactive to proactive maintenance practices while enhancing system availability, extending asset life, and providing a replicable model for similar water utilities in developing countries (Pathirana et al., 2021).
Dawa et al. (Thu,) studied this question.