Developing a Decision-Support System to Optimize Rehabilitation and Replacement Programs for Ferrous Distribution Mains in Municipal Water Systems

Given the importance of watermains in supplying safe and quality drinking water to customers, municipalities are required to ensure that the distribution mains are performing at certain pre-defined levels of service (LoS). In North America, the majority of watermains are made of ferrous materials, typically ductile iron (DI), and cast iron (CI) pipelines. These assets are prone to deterioration and breaks due to aging and other influencing factors, including corrosion. Annually, municipalities confront several constraints related to budgets and the proper time to intervene to preserve water distribution mains. With the significant municipal budgets required to rehabilitate or replace assets with newer pipelines, municipalities should be focusing on prioritizing interventions based on the likelihood and consequence of failures. Besides risk assessment frameworks, optimized programs need to be developed to maximize performance while minimizing the overall funding requirements in order to maintain sustainable funding levels and infrastructure in both the short- and long-term. Therefore, the main objective of this paper was to develop a decision-support system based on the genetic algorithm (GA) optimization tool. The tool considered maximizing the network performance and minimizing the total costs during a 5-year study period. After implementing the model on part of the Municipality of Thames Centre’s ferrous network, the total required cost attained was $434,282, and the performance of the network was restored to approximately 28 out of 100. The total costs of the major intervention (structural lining) and the minor intervention (cathodic protection) were $402,603 and $31,678, respectively. This study will benefit municipalities in developing optimized rehabilitation/replacement programs considering typical municipal constraints while maximizing the existing performance of the network and minimizing the total costs.