Pipe Break Rate Assessment While Considering Physical and Operational Factors: A Methodology based on Global Positioning System and Data-Driven Techniques

An accurate prediction of pipes failure rate plays a substantial role in the management Water Distribution Networks (WDNs). In this study, field study was conducted to register break and relevant causes WDN Yazd City, Iran. way, 851 water were incepted localized by Global Positioning System (GPS) apparatus. Then, 1033 cases reported eight zones understudy during March-December 2014. Pipes (BRP) calculated using depth pipe installation (hP), number failures (NP), pressure operation (P), age (AP). After completing database, robust Artificial Intelligence models, namely Multivariate Adaptive Regression Spline (MARS), Gene-Expression Programming (GEP), M5 Model Tree employed extract precise formulation for estimation. Results proposed relationships demonstrated that MARS model with Coefficient Correlation (R) 0.981 Root Mean Square Error (RMSE) 0.544 provided more satisfying efficiency than (R = 0.888 RMSE 1.096). Furthermore, statistical results indicated GEP models had comparatively at same accuracy level. Explicit equations (AI) satisfactorily comparable those obtained literature review terms various conditions: physical, operational, environmental factors complexity AI models. Through probabilistic framework rate, first-order reliability analysis technique highly performance when MARS-extracted-equation assigned as limit state function.