Prediction of Water Pipe Failure Using Fuzzy Inference System

Recurrent incidences of pipeline failure bring about serious physical, economical, and environmental consequences. Therefore, developing pragmatic approaches to model the water distribution infrastructure is crucial for preserving these assets. The objective of this paper is to develop a pipe failure prediction model grounded in fuzzy inference system and the probability of failure analysis to estimate the rate of failure in water infrastructure. First, the attributes that contribute to water pipelines deterioration are identified. Second, the fuzzy logic engine is designed to simulate the defined codes and functions. Third, probability of failure schemes are generated to address the uncertainties and predict the risk associated with water pipes’ failure. The developed model was applied to the water network of the city of El Pedregal in Peru. This paper developed an automated tool, expected to improve the quality of decision making, as it can assist water utility managers and infrastructure engineers in optimizing their future plans.