A Predictive Model for PDAM Water Pipe Network Repair Costs Due to Corrosion: Statistical Correlation between Raw Water Quality (pH, Total Dissolved Solids) Data from the Ministry of PUPR and Non-Revenue Water (NRW) Index

Abstract

This research establishes a predictive cost-estimation framework for Municipal Water Utilities (PDAM) in Indonesia by quantifying the nexus between raw water hydrochemical profiles and infrastructural integrity. The study focuses on pH levels and Total Dissolved Solids (TDS) as primary catalysts for internal pipe degradation, utilizing high-fidelity secondary datasets from the Ministry of Public Works and Housing (PUPR) and Statistics Indonesia (BPS) for the period 2020–2024. A multiple linear regression model was developed using the Ordinary Least Squares (OLS) method to measure the sensitivity of Non-Revenue Water (NRW) indices and repair expenditures to chemical fluctuations. Empirical findings reveal a robust positive correlation () between TDS concentrations and repair costs, with every 100 mg/L increase in TDS associated with a surge in maintenance burden of approximately 12.4 million IDR/km annually. Furthermore, water acidity (pH < 6.5) was identified as a critical threshold for the acceleration of internal tuberculation and pressure-induced fractures. The study demonstrates that chemical-induced physical leakage accounts for approximately 64% of total technical water losses in the most aggressive regional clusters. These results provide a strategic tool for PDAM management to transition from reactive maintenance to Condition-Based Maintenance (CBM), offering significant implications for national water security and budget optimization.