Hierarchical Fuzzy Expert System for Risk of Failure of Water Mains

Abstract

In Canada and the United States, there have been 700 water main breaks per day costing more than CAD 6 billion since 2000. Risk of failure is defined as the combination of probability and impact severity of a particular circumstance that negatively impacts the ability of infrastructure assets to meet municipal objectives. The presented research in this paper assists in designing a framework to evaluate the risk of water main failure using hierarchical fuzzy expert system (HFES). This system considers 16 risk-of-failure factors within four main categories representing both probability and negative consequences of failure. Results show that pipe age confers a strong impact on risk of failure followed by pipe material and breakage rate. They also show that damage to surroundings has the most negative consequence of a failure event. A set of municipal water network data are collected and used to examine the developed HFES. According to the proposed scale of risk of failure, about 8.4% (13 km) of the network's pipelines are risky and require mitigation actions in the short term. © 2010 ASCE.