Forecasting watermain failure using artificial neural network modelling

After rapid urban expansion in Ontario, post-World War II, there followed a lengthy period of time where only minimal infrastructure maintenance occurred. Now, however, most of that infrastructure is approaching the end of its predicted life expectancy, and has started failing at an unprecedented rate. The combination of low maintenance and the increasing age of water distribution infrastructure has resulted in increasing rates of pipe failures. To assign priorities for repair/replacement, artificial neural network modelling is employed. Eight independent variables are employed, namely pipe length, diameter, age, break category, soil type, pipe material, the year of Cement Mortar Lining (if implemented), and the year of Cathodic Protection (if implemented), to determine the importance of different factors influencing the pipe failure rate. The results in application to the distribution system in Etobicoke, Ontario demonstrate that ANN models have very strong predictive capabilities (R2=0.94) when compared with the multiple linear regression method (R2=0.75) to assist rehabilitation planning. Après la rapide expansion urbaine qui suivi la seconde guerre mondiale, l’Ontario connu une longue période pendant laquelle on ne porta attention qu’à l’entretien des petites infrastructures. Maintenant, la plupart des infrastructures approchent de leur fin de vie, et ont commencé à se détériorer à un rythme sans précédent. La combinaison du faible entretien et du vieillissement des infrastructures de distribution de l’eau a entraîné une augmentation des taux de bris des conduites. Pour constituer un outil d’aide à la décision, essentiel dans le choix du réseau à réhabiliter en priorité, on cartographie la prévision des défaillances du réseau de distribution d’eau à l’aide du système de modélisation des réseaux neuronaux artificiels (RNA). Cette approche a été appliquée au réseau de la ville de Etobicoke dans l’Ontario. Le modèle comporte huit variables indépendantes, notamment: longueur de la conduite, diamètre, âge, matériau, catégorie des défaillances, type de sol, plus deux facteurs de travaux de réhabilitation. Aux canalisations, on inclut l'année de mortier du ciment de revêtement (s’il a été appliqué), et l'année de la protection cathodique (si elle est appliquée). Afin de déterminer l'importance des différents facteurs qui influencent les défaillances des conduites. Les résultats obtenus pour le réseau d’eau à Etobicoke, démontrent que les modèles RNA ont de très fortes capacités de prévision (R2 = 0.94) pour faciliter les stratégies de réhabilitation, par rapport à la méthode de régression linéaire multiple (R2 = 0.75).

[1]  Thomas M. Walski,et al.  Economic analysis of water main breaks , 1982 .

[2]  Uri Shamir,et al.  An Analytic Approach to Scheduling Pipe Replacement , 1979 .

[3]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1989, Math. Control. Signals Syst..

[4]  anonymous,et al.  Comprehensive review , 2019 .

[5]  F Despagne,et al.  Neural networks in multivariate calibration. , 1998, The Analyst.

[6]  Isam Shahrour,et al.  Application of Artificial Neural Networks (ANN) to model the failure of urban water mains , 2010, Math. Comput. Model..

[7]  Neil S. Grigg,et al.  Failure Assessment Modeling to Prioritize Water Pipe Renewal: Two Case Studies , 2009 .

[8]  Holger R. Maier,et al.  Forecasting chlorine residuals in a water distribution system using a general regression neural network , 2006, Math. Comput. Model..

[9]  Jayong Koo,et al.  Predicting water pipe breaks using neural network , 2005 .

[10]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[11]  D. Needham,et al.  Why gas mains fail. Part 2. Deep excavations and tunneling , 1981 .

[12]  Osama Moselhi,et al.  Classification of Defects in Sewer Pipes Using Neural Networks , 2000 .

[13]  Yacoub M. Najjar,et al.  A Neural Network Approach for Site Characterization and Uncertainty Prediction , 1996 .

[14]  B. S. Lochbaum.  PSE&G develops models to predict main breaks , 1993 .

[15]  Holger R. Maier,et al.  Artificial Neural Network based Settlement Prediction Formula for Shallow Foundations on Granular Soils , 2002 .

[16]  Ahmad Habibian,et al.  Effect of Temperature Changes on Water‐Main Breaks , 1994 .

[17]  Isam Shahrour,et al.  Prediction of watermain failure frequencies using multiple and Poisson regression. , 2009 .

[18]  R. Hecht-Nielsen,et al.  Theory of the Back Propagation Neural Network , 1989 .

[19]  Shang-Lien Lo,et al.  Use of a GIS-based hybrid artificial neural network to prioritize the order of pipe replacement in a water distribution network , 2010, Environmental monitoring and assessment.

[20]  I A Basheer,et al.  Computational neural networks for predictive microbiology: I. Methodology. , 1997, International journal of food microbiology.

[21]  Scott McDonald.  Cathodic protection of water mains in Ottawa: analysis and planning , 2003 .

[22]  Edward A. McBean,et al.  Impacts of cathodic protection on pipe break probabilities: a Toronto case study , 2008 .

[23]  Prabhat Hajela,et al.  Neurobiological computational models in structural analysis and design , 1991 .

[24]  Hecht-Nielsen.  Theory of the backpropagation neural network , 1989 .

[25]  E. Mine Cinar,et al.  Neural Networks: A New Tool for Predicting Thrift Failures , 1992 .

[26]  I. C. Goulter,et al.  Spatial and temporal groupings of water main pipe breakage in Winnipeg , 1988 .

[27]  D. Kelly O'Day,et al.  Organizing and analyzing leak and break data for making main replacement decisions , 1982 .

[28]  Alfred S. Ciottoni.  Updating the New York City Water System , 1985 .

[29]  Holger R. Maier,et al.  Neural networks for the prediction and forecasting of water resource variables: a review of modelling issues and applications , 2000, Environ. Model. Softw..

[30]  Tarek Zayed,et al.  Infrastructure Management : Integrated AHP/ANN Model to Evaluate Municipal Water Mains' Performance , 2008 .

[31]  Balvant Rajani,et al.  Comprehensive review of structural deterioration of water mains: statistical models , 2001 .

[32]  Robert M. Clark,et al.  Costs for Water Supply Distribution System Rehabilitation , 2002 .

[33]  Graeme C. Dandy,et al.  Optimal Scheduling of Water Pipe Replacement Using Genetic Algorithms , 2001 .

[34]  Balvant Rajani,et al.  Forecasting Variations and Trends in Water-Main Breaks , 2002 .

[35]  Maureen Caudill,et al.  Neural networks primer, part III , 1988 .

[36]  Jean-Pierre Villeneuve,et al.  Modeling Water Pipe Breaks—Three Case Studies , 2003 .

[37]  Dragan Savic,et al.  Assessing pipe failure rate and mechanical reliability of water distribution networks using data-driven modeling , 2009 .