Bayesian Belief Networks for predicting drinking water distribution system pipe breaks

Bayesian Belief Networks for Predicting Drinking Water Distribution System Pipe 1 Breaks 2 Royce A. Francis ([email protected]), Seth D. Guikema ([email protected]), Lucas Henneman 3 ([email protected]) 4 a Department of Engineering Management and Systems Engineering, 1776 G St., NW #159, The George Washington 5 University, Washington, DC, USA 20006 6 b Department of Geography and Environmental Engineering, The Johns Hopkins University, 3400 N. Charles Street, 7 Baltimore, MD, USA 21218 8 Abstract: In this paper, we use Bayesian Belief Networks (BBNs) to construct a knowledge model for pipe breaks in 9 a water zone. To the authors’ knowledge, this is the first attempt to model drinking water distribution system pipe 10 breaks using BBNs. Development of expert systems such as BBNs for analyzing drinking water distribution system 11 data is not only important for pipe break prediction, but is also a first step in preventing water loss and water quality 12 deterioration through the application of machine learning techniques to facilitate data-based distribution system 13 monitoring and asset management. Due to the difficulties in collecting, preparing, and managing drinking water 14 distribution system data, most pipe break models can be classified as “statistical-physical” or “hypothesis-generating.” 15 We develop the BBN with the hope of contributing to the “hypothesis-generating” class of models, while 16 demonstrating the possibility that BBNs might also be used as “statistical-physical” models. Our model is learned 17 from pipe breaks and covariate data from a mid-Atlantic United States (U.S.) drinking water distribution system 18 network. BBN models are learned using a constraint-based method, a score-based method, and a hybrid method. 19 Model evaluation is based on log-likelihood scoring. Sensitivity analysis using mutual information criterion is also 20 reported. While our results indicate general agreement with prior results reported in pipe break modeling studies, they 21 also suggest that it may be difficult to select among model alternatives. This model uncertainty may mean that more 22 research is needed for understanding whether additional pipe break risk factors beyond age, break history, pipe 23 material, and pipe diameter might be important for asset management planning. 24