State-of-the-art review of water pipe failure prediction models and applicability to large-diameter mains

Abstract

This paper provides an overview of the work performed in the last 13 years to predict the failure of largediameter trunk water mains. Large-diameter water mains, defined as water mains with a diameter greater than 500 mm, form the main transmission lines in most water distribution systems. The consequences of their failure can be severe and costly. In order for predictive models to be applicable to large-diameter water mains, all models reviewed are capable of analysing individual pipes or pipe segments and calculate either an absolute probability of failure or the hazard of failure relative to other pipes in the system. These models can be divided into two broad categories: physical and statistical models. This review provides a description and a brief critique for each model presented.

Comparison of Statistical and Machine Learning Models for Pipe Failure Modeling in Water Distribution Networks

“…The performance of the ML methods is evaluated using accuracy, confusion matrices, and receiver operating characteristic (ROC) curves. Accuracy is estimated as the fraction of correct predictions to the total predictions [9], as shown in Equation (16). The confusion matrix, presented in Table 2, provides more information on model performance because it categorizes the results according to predictions and observations.…”

“…Particularly, Alvisi and Franchini [15] used two probabilistic models, namely the Weibull Exponential model and the Weibull Proportional Hazard model, to predict the number of failures in a pipe over an observation period using pipe diameter, age, and length. The results showed that both models correctly estimated the number of failures.Wilson et al [16] reviewed different statistical models (e.g., Logistic Regression and Proportional Hazard Model) for failure prediction in large-diameter pipes (i.e., greater than 500 mm). The authors concluded that the models were able to predict the failures of individual pipes or pipe segments with acceptable accuracy.…”

“…Wilson et al [16] reviewed different statistical models (e.g., Logistic Regression and Proportional Hazard Model) for failure prediction in large-diameter pipes (i.e., greater than 500 mm). The authors concluded that the models were able to predict the failures of individual pipes or pipe segments with acceptable accuracy.…”

Aplicación de la regresión logística para la predicción de roturas de tuberías en redes de abastecimiento de agua

“…Esto hace que su integración en la industria sea mucho más factible. Una de las desventajas de este método es que no evalúa la evolución de las variables en el tiempo (Wilson, Filion, & Moore, 2017). Futuras líneas de investigación deberían explorar esta carencia.…”

Neural Network Approach for Availability Indicator Prediction

“…Still, besides determining the failure frequency level of water conduits on the basis of operating data, it is also necessary to supplement such studies with the results of the modelling, using typical models or artificial intelligence and Monte Carlo simulations: probability of the failure occurrence and values of failure rate [5][6][7][8] or the amount of water losses through pipe breaks [9]. Therefore one of the aims of this research was to show that mathematical modelling could be used to determine the availability indicator and reliability level of water conduits.…”