Journal of Water and Wastewater Science and Engineering

Journal of Water and Wastewater Science and Engineering

Pipe Failure Prediction Using Weighted Logistic Regression for Managing Imbalanced and Recurring Events (Case Study: District 2 of Tehran)

Document Type : Research Paper

Authors
Mohsen Sattari 1
Homayoun Motiee 2
1 Ph.D. Candidate, Civil, Water and Environmental Faculty, Shahid Beheshti University, Tehran, Iran.
2 Associate Professor, Civil, Water and Environmental Faculty, Shahid Beheshti University, Tehran, Iran
10.22112/jwwse.2026.560844.1470
Abstract
Predicting pipe failures in water distribution networks is crucial for transitioning from reactive maintenance to proactive management. However, developing accurate predictive models faces challenges posed by imbalanced data (failures vs. non-failures) and the recurring nature of these events, which are often overlooked. This research addresses these challenges by developing a weighted logistic regression model, a machine learning technique. The study utilized 12 years of real-world pipe failure data from the water distribution network of District 2 in Tehran. The data were randomly split into two subsets: 70% for training and 30% for validation, to verify the model’s performance on unseen data. The model’s performance was evaluated using the Receiver Operating Characteristic (ROC) curve. Results demonstrated high predictive power with an Area Under the Curve (AUC) of 0.88 on the validation set. Pipe length was identified as the most significant predictor, with an odds ratio of 1.02. A negative coefficient for pipe age (-0.104) was a key finding, attributed to survivor bias stemming from premature failures due to improper installation. This model can serve as a reliable tool for water utilities to prioritize pipe inspections and replacements, emphasizing the critical role of installation quality alongside the pipe’s physical characteristics.
Keywords
Pipe Failure
Water Distribution Network
Machine Learning
Imbalanced Data
Recurrent Events
نوذری، ن، مستوری، ر، میرحسینی، و س.م.، (1402)، "بررسی کج شدن ناحیه‌ای درختان و سازه‌ها ناشی از نشست خاک در اثر شکستگی لوله شبکه توزیع آب در تهران"، علوم و مهندسی آب و فاضلاب، 8(2)، 37-46، https://doi.org/10.22112/jwwse.2022.348911.1319
مصلحی، ا.، و جلیلی قاضی‎زاده، م.ر.، (1395)، "مروری بر روابط فشار-شکستگی در شبکه‌های آب‌رسانی"، علوم و مهندسی آب و فاضلاب، 1(1)، 11-19، https://doi.org/10.22112/jwwse.2017.51053
مطیعی، ه.، و  قاسم‎نژاد، س.، (1396)، "کاربرد وتوسعه مدل‎های رگرسیونی برای پیش‎بینی میزان شکست لوله‎های شبکه توزیع آب شهری، مورد مطالعاتی ناحیه یک منطقه یک تهران"، علوم و مهندسی آب و فاضلاب، 2(2)، 48-58، ‎ https://doi.org/10.22112/jwwse.2017.89374.1024
Asadi, Y., (2024), “Employing machine learning in water infrastructure management: Predicting pipeline failures for improved maintenance and sustainable operations”, Industrial Artificial Intelligence, 2(1), 8, https://doi.org/10.1007/s44244-024-00022-w
Barton, N.A., Hallett, S.H., Jude, S.R. and Tran, T.H., (2022), “An evolution of statistical pipe failure models for drinking water networks: A targeted review”, Water Supply, 22(4),  3784-3813, https://doi.org/10.2166/ws.2022.019
Burnham, K.P., and Anderson, D.R., (2002), Model selection and multimodel inference: A practical information-theoretic approach, 2nd Edition, Springer-Verlag, https://doi.org/10.1007/b97636
Chang, P.W., and Newman, T.B., (2024), “Receiver operating characteristic (ROC) curves: The basics and beyond”, Hospital Pediatrics, 14(7), e330-e334, https://doi.org/10.1542/hpeds.2023-007462
Chen, R., Wang, Q., and Javanmardi, A., (2025), “A review of the application of Machine Learning for pipeline integrity predictive analysis in water distribution networks”, Archives of Computational Methods in Engineering,. 32(6), 3821-3849, https://doi.org/10.1007/s11831-025-10251-6
Farmani, R., Kakoudakis, K., Behzadian, K., and Butler, D., (2017), “Pipe failure prediction in water distribution systems considering static and dynamic factors”, Procedia Engineering, 186, 117-126, https://doi.org/10.1016/j.proeng.2017.03.217
Gross, R.E., and Harris, S.P., (2013), “Statistical performance evaluation of soft (elastomer) seat pressure relief valves”, In: Pressure Vessels and Piping Conference, (Vol. 55706, p. V06AT06A077), American Society of Mechanical Engineers, https://doi.org/10.1115/PVP2013-97031
He, H., and Garcia, E.A., (2009), “Learning from imbalanced data”, IEEE Transactions on Knowledge and Data Engineering, 21(9), 1263-1284, https://doi.org/10.1109/TKDE.2008.239
Hosmer, Jr., D.W., Lemeshow, S., and Sturdivant, R.X., (2013), Applied logistic regression, John Wiley & Sons, https://doi.org/10.1002/0471722146
Kakoudakis, K., Behzadian, K., Farmani, R., and Butler, D., (2017), “Pipeline failure prediction in water distribution networks using evolutionary polynomial regression combined with K-means clustering”, Urban Water Journal, 14(7), 737-742, https://doi.org/10.1080/1573062X.2016.1253755
Karadirek, I.E., Kaya-Basar, E., and Akdeniz, T., (2024), “A study on pipe failure analysis in water distribution systems using logistic regression”, Water Supply, 24(1), 176-186, https://doi.org/10.2166/ws.2023.335
Karimian, F., Kaddoura, K., Zayed, T., Hawari, A., and Moselhi, O., (2021), “Prediction of breaks in municipal drinking water linear assets”, Journal of Pipeline Systems Engineering and Practice, 12(1), 04020060, https://doi.org/10.1061/(ASCE)PS.1949-1204.0000511
Kleiner, Y., and Rajani, B., (2001), “Comprehensive review of structural deterioration of water mains: Statistical models”, Urban Water, 3(3),131-150, https://doi.org/10.1016/S1462-0758(01)00033-4
Latifi, M., Beig Zali, R., Javadi, A.A., and Farmani, R., (2024), “Customised-sampling approach for pipe failure prediction in water distribution networks”, Scientific Reports, 14(1),182, https://doi.org/10.1038/s41598-024-69109-9
Leštáková, M., Logan, K.T., Rehm, I.S., Pelz, P.F., and Friesen, J., (2024), “Do resilience metrics of water distribution systems really assess resilience? A critical review”, Water Research, 248, 120820, https://doi.org/10.1016/j.watres.2023.120820
Li, C., Chen, J., and Qin, G., (2019), “Partial Youden index and its inferences”, Journal of Biopharmaceutical Statistics, 29(2), 385-399,‏ https://doi.org/10.1080/10543406.2018.1535502
Omar, A., Delnaz, A., and Nik-Bakht, M., (2023), “Comparative analysis of machine learning techniques for predicting water main failures in the City of Kitchener”, Journal of Infrastructure Intelligence and Resilience, 2(3), 100044,‏ https://doi.org/10.1016/j.iintel.2023.100044
Rifaai, T.M., Abokifa, A.A., and Sela, L., (2022), “Integrated approach for pipe failure prediction and condition scoring in water infrastructure systems”, Reliability Eineering & System Safety, 220, 108271, https://doi.org/10.1016/j.ress.2021.108271
Robles-Velasco, A., Cortés, P., Muñuzuri, J., and Onieva, L., (2020), “Prediction of pipe failures in water supply networks using logistic regression and support vector classification”, Reliability Engineering & System Safety, 196, 106754, https://doi.org/10.1016/j.ress.2019.106754
Salehinasab, S., and Goshtasbi Rad, E., (2025), “Enhancing pipe failure prediction using cluster-specific classification via K-means clustering”, Sustainable and Resilient Infrastructure, 1-30, https://doi.org/10.1080/23789689.2025.2470504
Scheidegger, A., Leitao, J.P. and Scholten, L. (2015). Statistical failure models for water distribution pipes–A review from a unified perspective. Water research. 83. 237-247, https://doi.org/10.1016/j.watres.2015.06.027
Shirzad, A., and Safari, M.J.S., (2019), “Pipe failure rate prediction in water distribution networks using multivariate adaptive regression splines and random forest techniques”, Urban Water Journal, 16(9), 653-661, https://doi.org/10.1080/1573062X.2020.1713384
Snider, B., and McBean, E.A., (2020), “Improving urban water security through pipe-break prediction models: Machine learning or survival analysis”, Journal of Environmental Engineering, 146(3), 04019129, https://doi.org/10.1061/(ASCE)EE.1943-7870.0001657
Solomatine, D.P., and Ostfeld, A., (2008), “Data-driven modelling: Some past experiences and new approaches”, Journal of Hydroinformatics, 10(1), 3-22, https://doi.org/10.2166/hydro.2008.015
Wang, X., Mazumder, R.K., Salarieh, B., Salman, A.M., Shafieezadeh, A., and Li, Y., (2022), “Machine learning for risk and resilience assessment in structural engineering, Progress and future trends”, Journal of Structural Engineering, 148(8), 03122003, https://doi.org/10.1061/(ASCE)ST.1943-541X.0003392
Wu, Y., and Liu, S., (2017), "A review of data-driven approaches for burst detection in water distribution systems", Urban Water Journal, 14(9), 972-983, https://doi.org/10.1080/1573062X.2017.1279191
Yamijala, S., Guikema, S.D., and Brumbelow, K., (2009), “Statistical models for the analysis of water distribution system pipe break data”, Reliability Engineering & System Safety, 94(2), 282-293, https://doi.org/10.1016/j.ress.2008.03.011
Journal of Water and Wastewater Science and Engineering
Volume 11, Issue 2
Summer 2026
Pages 17-30

  • Receive Date 20 November 2025
  • Revise Date 08 June 2026
  • Accept Date 22 June 2026