Pipe failure rate prediction in water distribution networks using multivariate adaptive regression splines and random forest techniques

ABSTRACT This paper presents the results of a comparison between multivariate adaptive regression splines (MARS) and random forest (RF) techniques in pipe failure prediction in two water distribution networks. In this regard, pipe diameter, pipe length, pipe installation depth, pipe age and average hydraulic pressure are considered as input variables. Results show that the RF outperforms the MARS which is found as an accurate pipe failure rate predictor. The proposed models are further evaluated through dividing the data into three parts of lower, medium and higher pipe failure rate values. According to the equations produced by MARS technique, three variables of pipe diameter, pipe age and average hydraulic pressure are distinguished as the most effective variables in predicting pipe failure rate in the first case study. Four variables of pipe diameter, pipe length, pipe age and average hydraulic pressure are determined as the most effective variables in the second case study.

[1]  Pao-Shan Yu,et al.  Comparison of random forests and support vector machine for real-time radar-derived rainfall forecasting , 2017 .

[2]  Andrew Wood,et al.  Using Water Main Break Data to Improve Asset Management for Small and Medium Utilities: District of Maple Ridge, B.C. , 2009 .

[3]  Edward A. McBean,et al.  IMPROVING TIME-TO-FAILURE PREDICTIONS FOR WATER DISTRIBUTION SYSTEMS USING GRADIENT BOOSTING ALGORITHM , 2018 .

[4]  Balvant Rajani,et al.  Prioritising Individual Water Mains for Renewal , 2008 .

[5]  Małgorzata Kutyłowska,et al.  Comparison of Two Types of Artificial Neural Networks for Predicting Failure Frequency of Water Conduits , 2016 .

[6]  Małgorzata Kutyłowska,et al.  Application of MARSplines Method for Failure Rate Prediction , 2018, Periodica Polytechnica Civil Engineering.

[7]  Mohammad Rezaie-Balf,et al.  Wavelet coupled MARS and M5 Model Tree approaches for groundwater level forecasting , 2017 .

[8]  V. N. Sharda,et al.  Performance of Multivariate Adaptive Regression Splines (MARS) in predicting runoff in mid-Himalayan micro-watersheds with limited data / Performances de régressions par splines multiples et adaptives (MARS) pour la prévision d'écoulement au sein de micro-bassins versants Himalayens d'altitudes inte , 2008 .

[9]  Shelley E. McDonald,et al.  Failure modes and mechanisms in gray cast iron pipe , 2001 .

[10]  A. Vanrenterghem-Raven Risk Factors of Structural Degradation of an Urban Water Distribution System , 2007 .

[11]  N. Sánchez,et al.  A spatial downscaling approach for the SMAP passive surface soil moisture product using random forest regression , 2018 .

[12]  Guoqiang Chen,et al.  Multiple Random Forests Modelling for Urban Water Consumption Forecasting , 2017, Water Resources Management.

[13]  Ahmad Sharafati,et al.  Development of multivariate adaptive regression spline integrated with differential evolution model for streamflow simulation , 2019, Journal of Hydrology.

[14]  Junfeng Zhu,et al.  Applying a weighted random forests method to extract karst sinkholes from LiDAR data , 2016 .

[15]  Edward E. Gbur,et al.  Statistical modeling techniques , 1983 .

[16]  Solomon Tesfamariam,et al.  Performance indicators for small- and medium-sized water supply systems: a review , 2014 .

[17]  J. Friedman Multivariate adaptive regression splines , 1990 .

[18]  Regression methods for predicting rate and type of failures of water conduits , 2017 .

[19]  Dragan Savic,et al.  The Use of Data-Driven Methodologies for Prediction of Water and Wastewater Asset Failures , 2009 .

[20]  S. M. Bateni,et al.  Estimation of soil cation exchange capacity using Genetic Expression Programming (GEP) and Multivariate Adaptive Regression Splines (MARS) , 2015 .

[21]  Kerry J. McManus,et al.  Prediction of Water Pipe Asset Life Using Neural Networks , 2007 .

[22]  Ozgur Kisi,et al.  Modelling daily dissolved oxygen concentration using least square support vector machine, multivariate adaptive regression splines and M5 model tree , 2018 .

[23]  Symeon E. Christodoulou,et al.  A Risk Analysis Framework for Evaluating Structural Degradation of Water Mains in Urban Settings, Using Neurofuzzy Systems and Statistical Modeling Techniques , 2003 .

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

[25]  Mir Jafar Sadegh Safari,et al.  Decision tree (DT), generalized regression neural network (GR) and multivariate adaptive regression splines (MARS) models for sediment transport in sewer pipes. , 2019, Water science and technology : a journal of the International Association on Water Pollution Research.

[26]  A. M. Al-Juboori Generating Monthly Stream Flow Using Nearest River Data: Assessing Different Trees Models , 2019, Water Resources Management.

[27]  Murat Kankal,et al.  Estimating suspended sediment load with multivariate adaptive regression spline, teaching-learning based optimization, and artificial bee colony models. , 2018, The Science of the total environment.

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

[29]  Jalal Shiri,et al.  Improving the performance of the mass transfer-based reference evapotranspiration estimation approaches through a coupled wavelet-random forest methodology , 2018, Journal of Hydrology.

[30]  Seyed Farzad Karimian Failure Rate Prediction Models of Water Distribution Networks , 2015 .

[31]  Hongwei Zhang,et al.  An assessment model of water pipe condition using Bayesian inference , 2010 .

[32]  J. M. Makar,et al.  A preliminary analysis of failures in grey cast iron water pipes , 2000 .

[33]  Luigi Berardi,et al.  Development of pipe deterioration models for water distribution systems using EPR , 2008 .

[34]  M. Kutyłowska Forecasting failure rate of water pipes , 2019 .

[35]  Tarek Mabrouki,et al.  On the Modeling of Surface Roughness and Cutting Force when Turning of Inconel 718 Using Artificial Neural Network and Response Surface Methodology: Accuracy and Benefit , 2017 .

[36]  David H. Marks,et al.  Performance of water distribution networks: integrated approach , 1990 .

[37]  Eric R. Ziegel,et al.  The Elements of Statistical Learning , 2003, Technometrics.

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

[39]  Scott Gould,et al.  Understanding how the Australian climate can affect pipe failure , 2009 .

[40]  Y Hu,et al.  Factors contributing to the failure of asbestos cement water mains , 2007 .

[41]  Solomon Tesfamariam,et al.  Estimating time to failure of cast-iron water mains , 2007 .

[42]  Massoud Tabesh,et al.  A comparison between performance of support vector regression and artificial neural network in prediction of pipe burst rate in water distribution networks , 2014, KSCE Journal of Civil Engineering.

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

[44]  Balvant Rajani,et al.  Comprehensive review of structural deterioration of water mains: physically based models , 2001 .

[45]  Puligandla Viswanadham Failure Modes and Mechanisms , 1998 .

[46]  R. Peralta,et al.  Transport modeling and multivariate adaptive regression splines for evaluating performance of ASR systems in freshwater aquifers , 2017 .

[47]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[48]  Jerome H. Friedman Multivariate adaptive regression splines (with discussion) , 1991 .

[49]  S. Sen,et al.  First Report of Alternaria dianthicola Causing Leaf Blight on Withania somnifera from India. , 2007, Plant disease.

[50]  Mohamed M. Morsy,et al.  Modeling urban coastal flood severity from crowd-sourced flood reports using Poisson regression and Random Forest , 2018 .

[51]  Małgorzata Kutyłowska,et al.  Neural network approach for failure rate prediction , 2015 .