Application of Artificial Neural Networks (ANN) to model the failure of urban water mains

  • Authors:

  • Raed Jafar;Isam Shahrour;Ilan Juran

  • Affiliations:

  • Laboratoire de mécanique de Lille, University of Sciences and Technologies of Lille, 59650 Villeneuve d'Ascq, France;Laboratoire de mécanique de Lille, University of Sciences and Technologies of Lille, 59650 Villeneuve d'Ascq, France;Urban Utility Center, Polytechnic University, Brooklyn, NY 11201, United States

  • Venue:
  • Mathematical and Computer Modelling: An International Journal
  • Year:
  • 2010

Quantified Score

Hi-index 0.98

Visualization

Abstract

This paper presents an application of Artificial Neural Networks (ANN) to model the failure rate and estimate the optimal replacement time for the individual pipes in an urban water distribution system. The performances of the ANN are examined using a 14-year data set collected in a city in the north of France. The first part of the paper presents the collected data. The second part describes the construction and validation of six ANN models. After a discussion of the performances of these models, they are used for the prediction of water mains failure and the determination of the benefit index, which allows optimization of investment for the rehabilitation and maintenance of urban water mains. The spatial repartition of the risk of degradation is illustrated using a geographic information system, which constitutes an effective tool for the elaboration of strategies of rehabilitation of water distribution systems.