E3S Web Conf.
Volume 22, 2017International Conference on Advances in Energy Systems and Environmental Engineering (ASEE17)
|Number of page(s)||8|
|Published online||07 November 2017|
Application of regression tress for prediction of water conduits failure rate
Wrocław University of Science and Technology, Faculty of Environmental Engineering, Wybrzeże Wyspiańskiego 27, 50-370 Wroclaw, Poland
* Corresponding author: firstname.lastname@example.org
This paper presents the results of predicting the failure rate of water distribution pipes and house connections in a selected city in Poland by means of regression trees. Several regression tree models were built as part of modelling. Optimal models were selected (separately for each of the water conduit types) via an analysis of the so–called costs. The regression tree structure comprised independent variables, i.e. predictors (length, diameter, year of construction and material). The failure rates of the two types of water conduits were the dependent variable. The optimal models were characterized by the lowest costs and a relatively simple tree architecture. Operational data from the years 2001–2012 were used to determine the experimental (real) values of the failure rate and to build regression tree models. The optimal models included eight divided nodes and nine end nodes. The ranking of the significance of the parameters showed that length was the predictor responsible for division on the successive tree levels. The obtained high consistency (0.99) of the real data with the predicted ones indicates that the regression tree method can be used to analyze and assess the failure rate of water conduits.
© The Authors, published by EDP Sciences, 2017
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. (http://creativecommons.org/licenses/by/4.0/).
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.