A Decision-making Framework for Water Distribution Systems using Fuzzy Inference and Centrality Analysis

Abstract

Water Distribution Systems (WDSs) are among the most important infrastructures that are critical for the smooth functioning of communities. However, age-old existing WDSs are progressively at risk in the United States. Since failure in a WDS may affect other interdependent infrastructure and result in high economic consequences, water utilities are more interested in preventing rather than reacting to failure. The current study proposes a decision support framework that employs fuzzy hierarchical inference and network graph analysis to rank the most vulnerable water pipelines considering a set of risk factors and their negative consequences. Fourteen (14) risk factors are identified considering water and road network interdependence. These factors are classified into four main vulnerability indices (strength, hydraulic, environmental, road) and one consequence class in order to evaluate the integrated risk of water mains. Fuzzy analytical hierarchy process is used to quantify the uncertainty in the risk factors to aid the decision-making process. Network centrality analysis is used to identify the most critical components of the WDS. The final decision is made by combining the outputs from the fuzzy inference and the network centrality analysis. The WDS of Modena, Italy is used to demonstrate the proposed approach.