Hydraulic Deterioration Models for Storm-Water Drainage Pipes: Ordered Probit versus Probabilistic Neural Network

A decrease in discharge capacity of storm-water drainage pipes is the result of the so-called hydraulic deterioration which reduces the cross sectional area of pipes and increases the pipe roughness. Hydraulic deterioration is caused by tree root intrusion, sediment accumulation, and encrustation, and is affected by many influential factors such as pipe size and pipe location. Predicting hydraulic deterioration is important for effective management of drainage pipes. An ordered probit deterioration model (OPDM) and a probabilistic neural network deterioration model (PNNDM) were developed in this study using the influential factors as model inputs and the hydraulic condition as model output. Their predictive performances were compared against each other using a case study from Melbourne, Australia with a sample of 417 storm-water drainage pipes subjected to closed circuit television inspection. The results show that the PNNDM is more suitable for predicting the hydraulic deterioration and outperforms the OPDM. Several input factors such as pipe size and pipe age are found significant to the hydraulic deterioration.

[1]  G. Kuczera,et al.  Markov Model for Storm Water Pipe Deterioration , 2002 .

[2]  Samer M. Madanat,et al.  Robust Maintenance Policies for Markovian Systems under Model Uncertainty , 2005, Comput. Aided Civ. Infrastructure Eng..

[3]  David B. Dunson,et al.  Bayesian Data Analysis , 2010 .

[4]  Samer Madanat,et al.  Computation of Infrastructure Transition Probabilities using Stochastic Duration Models , 2002 .

[5]  PingSun Leung,et al.  Predicting shrimp disease occurrence: artificial neural networks vs. logistic regression , 2000 .

[6]  T. Cacoullos.  Estimation of a multivariate density , 1966 .

[7]  Kerry J. McManus,et al.  Practical review: a discussion of deterioration models for stormwater pipe systems in Victoria, Australia , 2005 .

[8]  Anne Ng,et al.  Modelling serviceability deterioration of concrete stormwater pipes using genetic algorithm trained neural network , 2006 .

[9]  Naim Haie,et al.  Neural Networks in Water Resources Management , 2001 .

[10]  Ferenc Szidarovszky,et al.  Artificial Neural Network Approach for Predicting Transient Water Levels in a Multilayered Groundwater System under Variable State, Pumping, and Climate Conditions , 2003 .

[11]  Stewart Burn,et al.  Prediction models for serviceability deterioration of stormwater pipes , 2008 .

[12]  Alberto Tesi,et al.  On the Problem of Local Minima in Backpropagation , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[13]  George Morcous,et al.  Case-Based Reasoning System for Modeling Infrastructure Deterioration , 2002 .

[14]  Tarek Zayed,et al.  Condition Rating Model for Underground Infrastructure Sustainable Water Mains , 2006 .

[15]  Huu Dung Tran,et al.  Investigation of deterioration models for stormwater pipe systems , 2007 .

[16]  Samuel T. Ariaratnam,et al.  Assessment of Infrastructure Inspection Needs Using Logistic Models , 2001 .

[17]  M.H. Hassoun,et al.  Fundamentals of Artificial Neural Networks , 1996, Proceedings of the IEEE.

[18]  Theodore Johnson,et al.  Exploratory Data Mining and Data Cleaning , 2003 .

[19]  Mahesh D. Pandey,et al.  Probabilistic Neural Network for Reliability Assessment of Oil and Gas Pipelines , 2002 .

[20]  Kerry J. McManus,et al.  The effect of Thornthwaite Moisture Index Changes in ground movement predictions in Australian soils , 2004 .

[21]  Michael K Stenstrom,et al.  Identification of land use with water quality data in stormwater using a neural network. , 2003, Water research.

[22]  Donald F. Specht,et al.  Probabilistic neural networks , 1990, Neural Networks.

[23]  B. J. C. Perera,et al.  Neural networks deterioration models for serviceability condition of buried stormwater pipes , 2007, Eng. Appl. Artif. Intell..

[24]  Anne Ng,et al.  Selection of genetic algorithm operators for river water quality model calibration , 2003 .

[25]  Stephen Thomas,et al.  Modelling sovereign credit ratings: Neural networks versus ordered probit , 2006, Expert Syst. Appl..

[26]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[27]  Jamshid Mohammadi,et al.  Analysis of Bridge Condition Rating Data Using Neural Networks , 1997 .

[28]  Vasilios Katos,et al.  Network intrusion detection: Evaluating cluster, discriminant, and logit analysis , 2007, Inf. Sci..

[29]  Tarek Hegazy,et al.  Predicting cost deviation in reconstruction projects: Artificial neural networks versus regression , 2003 .

[30]  Manjriker Gunaratne,et al.  Application of Neural Network Model to Forecast Short-Term Pavement Crack Condition: Florida Case Study , 2001 .

[31]  Mohamed Abdel-Aty,et al.  Predicting Injury Severity Levels in Traffic Crashes: A Modeling Comparison , 2004 .

[32]  Duc Truong Pham,et al.  Intelligent Optimisation Techniques: Genetic Algorithms, Tabu Search, Simulated Annealing and Neural Networks , 2011 .

[33]  Paul Davis,et al.  Application of probabilistic neural networks in modelling structural deterioration of stormwater pipes , 2006 .

[34]  B. Tabachnick,et al.  Using Multivariate Statistics , 1983 .

[35]  Samer Madanat,et al.  Estimation of infrastructure transition probabilities from condition rating data , 1995 .

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

[37]  S. Manel,et al.  Evaluating presence-absence models in ecology: the need to account for prevalence , 2001 .

[38]  Seongkyu Chang,et al.  Application of Probabilistic Neural Networks for Prediction of Concrete Strength , 2005 .

[39]  Dulcy M. Abraham,et al.  Estimating Transition Probabilities in Markov Chain-Based Deterioration Models for Management of Wastewater Systems , 2006 .

[40]  Maha N. Hajmeer,et al.  Comparison of logistic regression and neural network-based classifiers for bacterial growth , 2003 .

[41]  Guru Kulandaivel,et al.  Pipeline Condition Prediction Using Neural Network Models , 2005 .