Ilorin metropolitan water supply infrastructure (WSI) asset management under limited data

ABSTRACT Traditional water supply infrastructure management relies heavily on manual inspection methods that often fall short, particularly for monitoring deeply buried and inaccessible systems. These challenges are exacerbated by the limited data availability, such as historical pipe failure records, material degradation rates, and soil composition details, which are vital for infrastructure assessment. To overcome these limitations, this study employs Monte Carlo simulations to generate a comprehensive dataset that incorporates variables such as pipe breaks, diameter, age, soil type, and material. Advanced statistical techniques, including logistic regression and Bayesian updating, are then used to refine predictions dynamically by integrating simulation results with sparse real-world observations. This approach not only identified key risk factors like pipe age and break frequency but also demonstrated strong accuracy in classifying asset conditions. The study's findings support the development of a prioritized asset management plan, emphasizing routine data collection and predictive maintenance strategies that enhance infrastructure resilience and performance.