Short-Term Forecasting of Household Water Demand in the UK Using an Interpretable Machine Learning Approach
Publication: Journal of Water Resources Planning and Management
Volume 147, Issue 4
Abstract
This study utilizes a rich UK data set of smart demand metering data, household characteristics, and weather data to develop a demand forecasting methodology that combines the high accuracy of machine learning models with the interpretability of statistical methods. For this reason, a random forest model is used to predict daily demands 1 day ahead for groups of properties (mean of ) with homogenous characteristics. A variety of interpretable machine learning techniques [variable permutation, accumulated local effects (ALE) plots, and individual conditional expectation (ICE) curves] are used to quantify the influence of these predictors (temporal, weather, and household characteristics) on water consumption. Results show that when past consumption data are available, they are the most important explanatory factor. However, when they are not, a combination of household and temporal characteristics can be used to produce a credible model with similar forecasting accuracy. Weather input has overall a mild to no effect on the model’s output, although this effect can become significant under certain conditions.
Get full access to this article
View all available purchase options and get full access to this article.
Data Availability Statement
Some or all data, models, or code used during the study were provided by a third party. Direct requests for these materials may be made to the provider as indicated in the Acknowledgments.
Acknowledgments
This study was funded as part of the Water Informatics Science and Engineering Centre for Doctoral Training (WISE CDT) under a grant from the Engineering and Physical Sciences Research Council (EPSRC) (Grant No. EP/L016214/1). The data for this study were made available by Wessex Water.
References
Adadi, A., and M. Berrada. 2018. “Peeking inside the black-box: A survey on explainable artificial intelligence (XAI).” IEEE Access 6: 52138–52160. https://doi.org/10.1109/ACCESS.2018.2870052.
Anele, A. O., Y. Hamam, A. M. Abu-Mahfouz, and E. Todini. 2017. “Overview, comparative assessment and recommendations of forecasting models for short-term water demand prediction.” Water 9 (11): 887. https://doi.org/10.3390/w9110887.
Apley, D. W., and J. Zhu. 2020. “Visualizing the effects of predictor variables in black box supervised learning models.” J. R. Stat. Soc., Ser. B 82 (4): 1059–1086. https://doi.org/10.1111/rssb.12377.
Bellfield, S. 2001. “Short-term domestic water demand: Estimation, forecasting and management.” Ph.D. thesis, Faculty of Environment, School of Geography, Univ. of Leeds.
Berk, R. A., S. B. Sorenson, and G. Barnes. 2016. “Forecasting domestic violence: A machine learning approach to help inform arraignment decisions.” J. Empirical Legal Stud. 13 (1): 94–115. https://doi.org/10.1111/jels.12098.
Billings, R. B., and C. V. Jones. 2008. Forecasting urban water demand. 2nd ed. Denver: American Water Works Association.
Breiman, L. 2001. “Random forests.” Mach. Learn. 45 (1): 5–32.
Butler, D., and F. A. Memon. 2006. Water demand management. London: International Water Association Publishing.
CACI (Consolidated Analysis Center, Incorporated). 2014. “The acorn user guide.” Accessed November 29, 2020. https://acorn.caci.co.uk.
Carmichael, I., et al. 2019. “Joint and individual analysis of breast cancer histologic images and genomic covariates.” Accessed January 15, 2019. https://arxiv.org/abs/1912.00434.
Carvalho, D. V., E. M. Pereira, and J. S. Cardoso. 2019. “Machine learning interpretability: A survey on methods and metrics.” Electronics 8 (8): 832. https://doi.org/10.3390/electronics8080832.
Chen, G., T. Long, J. Xiong, and Y. Bai. 2017. “Multiple random forests modelling for urban water consumption forecasting.” Water Resour. Manage. 31 (15): 4715–4729. https://doi.org/10.1007/s11269-017-1774-7.
Choi, E. 2018. “Doctor AI: Interpretable deep learning for modeling electronic health records.” Ph.D. thesis, School of Computational Science and Engineering, Georgia Institute of Technology.
Clarke, G. P., A. Kashti, A. McDonald, and P. Williamson. 1997. “Estimating small area demand for water: A new methodology.” Water Environ. J. 11 (3): 186–192. https://doi.org/10.1111/j.1747-6593.1997.tb00114.x.
Cremona, F., L. Tuvikene, J. Haberman, P. Nõges, and T. Nõges. 2018. “Factors controlling the three-decade long rise in cyanobacteria biomass in a eutrophic shallow lake.” Sci. Total Environ. 621 (Apr): 352–359. https://doi.org/10.1016/j.scitotenv.2017.11.250.
Culos, A., et al. 2020. “Integration of mechanistic immunological knowledge into a machine learning pipeline improves predictions.” Nat. Mach. Intell. 2: 619–628. https://doi.org/10.1038/s42256-020-00232-8.
Doshi-Velez, F., and B. Kim. 2017. “Towards a rigorous science of interpretable machine learning.” Preprint, submitted February 28, 2017. https://arxiv.org/abs/1702.08608.
Dos Santos, C. C., and F. A. J. Pereira. 2014. “Water demand forecasting model for the metropolitan area of São Paulo, Brazil.” Water Resour. Manage. 28 (13): 4401–4414. https://doi.org/10.1007/s11269-014-0743-7.
Dresner, S., and P. Ekins. 2006. “Design of environmentally and socially conscious water metering tariffs for the UK.” J. Environ. Plann. Manage. 49 (6): 909–928. https://doi.org/10.1080/09640560600947048.
Duerr, I., H. R. Merrill, C. Wang, R. Bai, M. Boyer, M. D. Dukes, and N. Bliznyuk. 2018. “Forecasting urban household water demand with statistical and machine learning methods using large space-time data: A Comparative study.” Environ. Modell. Software 102 (Apr): 29–38. https://doi.org/10.1016/j.envsoft.2018.01.002.
Edwards, K., and L. Martin. 1995. “A methodology for surveying domestic water consumption.” Water Environ. J. 9 (5): 477–488. https://doi.org/10.1111/j.1747-6593.1995.tb01486.x.
Gato, S., N. Jayasuriya, and P. Roberts. 2007. “Temperature and rainfall thresholds for base use urban water demand modeling.” J. Hydrol. 337 (3–4): 364–376. https://doi.org/10.1016/j.jhydrol.2007.02.014.
Goldstein, A., A. Kapelner, J. Bleich, and E. Pitkin. 2015. “Peeking inside the black box: Visualizing statistical learning with plots of individual conditional 166 expectation.” J. Comput. Graphical Stat. 24 (1): 44–65. https://doi.org/10.1080/10618600.2014.907095.
Goodchild, C. 2003. “Modelling the impact of climate change on domestic water demand.” Water Environ. J. 17 (1): 8–12. https://doi.org/10.1111/j.1747-6593.2003.tb00423.x.
Hartley, J. A. 1995. “A neural network and rule based system application in water demand forecasting.” Ph.D. thesis, Dept. of Electronic and Computer Engineering, Brunel Univ.
Herrera, M., L. Torgo, J. Izquierdo, and R. Pérez-García. 2010. “Predictive models for forecasting hourly urban water demand.” J. Hydrol. 387 (1–2): 141–150. https://doi.org/10.1016/j.jhydrol.2010.04.005.
Jorgensen, B., M. Graymore, and K. O’Toole. 2009. “Household water use behavior: An integrated model.” J. Environ. Manage. 91 (1): 227–236. https://doi.org/10.1016/j.jenvman.2009.08.009.
Jorgensen, B. S., J. F. Martin, M. W. Pearce, and E. Willis. 2014. “Predicting household water consumption with individual-level variables.” Environ. Behav. 46 (7): 872–897. https://doi.org/10.1177/0013916513482462.
Kowalski, M., and D. Marshallsay. 2005. “Using measured microcomponent data to model the impact of water conservation strategies on the diurnal consumption profile.” Water Sci. Technol. Water Supply 5 (3–4): 145–150. https://doi.org/10.2166/ws.2005.0094.
Liaw, A., and M. Wiener. 2002. “Classification and regression by randomForest.” R News 2 (3): 18–22.
Met Office. 2006a. MIDAS: UK daily rainfall data. Oxford, UK: NCAS British Atmospheric Data Centre.
Met Office. 2006b. MIDAS: UK daily temperature data. Oxford, UK: NCAS British Atmospheric Data Centre.
Met Office. 2006c. MIDAS: UK daily weather observation data. Oxford, UK: NCAS British Atmospheric Data Centre.
Met Office. 2006d. MIDAS: UK hourly weather observation data. Oxford, UK: NCAS British Atmospheric Data Centre.
Met Office. 2006e. MIDAS: UK soil temperature data. Oxford, UK: NCAS British Atmospheric Data Centre.
Molnar, C. 2019. “Interpretable machine learning.” Accessed November 29, 2020. https://christophm.github.io/interpretable-ml-book/.
Parker, J. M. 2014. “Assessing the sensitivity of historic micro-component household water-use to climatic drivers.” Ph.D. thesis, Geography and Environment, Loughborough Univ.
Parker, J. M., and R. L. Wilby. 2013. “Quantifying household water demand: A review of theory and practice in the UK.” Water Resour. Manage. 27 (4): 981–1011. https://doi.org/10.1007/s11269-012-0190-2.
Russac, D. A. V., K. R. Rushton, and R. J. Simpson. 1991. “Insights into domestic demand from a metering trial.” Water Environ. J. 5 (3): 342–351. https://doi.org/10.1111/j.1747-6593.1991.tb00628.x.
Schleich, J., and T. Hillenbrand. 2008. “Determinants of residential water demand in Germany.” Ecol. Econ. 68 (6): 1756–1769. https://doi.org/10.1016/j.ecolecon.2008.11.012.
Scornet, E. 2017. “Tuning parameters in random forests.” In Vol. 60 of ESAIM: Proc., and Surveys, 144–162. Les Ulis, France: EDP Sciences. https://doi.org/10.1051/proc/201760144.
Stanley, N., et al. 2020. “VoPo leverages cellular heterogeneity for predictive modeling of single-cell data.” Nat. Commun. 11: 3738. https://doi.org/10.1038/s41467-020-17569-8.
Steinley, D. 2006. “K-means clustering: A half-century synthesis.” Br. J. Math. Stat. Psychol. 59 (1): 1–34. https://doi.org/10.1348/000711005X48266.
Williamson, P., G. Mitchell, A. McDonald, and P. Wattage. 2002. “Domestic water demand forecasting: A static microsimulation approach.” Water Environ. Manage. 16 (4): 243–248. https://doi.org/10.1111/j.1747-6593.2002.tb00410.x.
Wong, J., Q. Zhang, and Y. D. Chen. 2010. “Statistical modeling of daily urban water consumption in Hong Kong: Trend, changing patterns, and forecast.” Water Resour. Res. 46 (3). https://doi.org/10.1029/2009WR008147.
Xenochristou, M. 2019. “Water demand forecasting using machine learning on weather and smart metering data.” Ph.D. thesis, College of Engineering, Mathematics, and Physical Sciences, Centre for Water Systems, Univ. of Exeter.
Xenochristou, M., M. Blokker, I. Vertommen, J. Urbanus, and Z. Kapelan. 2018. “Investigating the influence of weather on water consumption: A Dutch case study.” In Proc., WDSA/CCWI 2018 Joint Conf., 26–35. Sheffield, UK: Univ. of Sheffield.
Xenochristou, M., C. Hutton, J. Hofman, and Z. Kapelan. 2020a. “Water demand forecasting accuracy and influencing factors at different spatial scales using a Gradient Boosting Machine.” Water Resour. Res. 56 (8): e2019WR026304. https://doi.org/10.1029/ 2019WR026304.
Xenochristou, M., and Z. Kapelan. 2020. “An ensemble stacked model with bias correction for improved water demand forecasting.” Urban Water J. 17 (3): 212–223. https://doi.org/10.1080/1573062X.2020.1758164.
Xenochristou, M., Z. Kapelan, and C. Hutton. 2020b. “Using smart demand metering data and customer characteristics to investigate the influence of weather on water consumption in the UK.” J. Water Resour. Plann. Manage. 146 (2): 04019073. https://doi.org/10.1061/(ASCE)WR.1943-5452.0001148.
Zubaidi, S. L., S. K. Gharghan, J. Dooley, R. M. Alkhaddar, and M. Abdellatif. 2018. “Short-term urban water demand prediction considering weather factors.” Water Resour. Manage. 32 (14): 4527–4542. https://doi.org/10.1007/s11269-018-2061-y.
Information & Authors
Information
Published In
Journal of Water Resources Planning and Management
Volume 147 • Issue 4 • April 2021
Copyright
© 2021 American Society of Civil Engineers.
History
Received: Jul 5, 2019
Accepted: Sep 6, 2020
Published online: Jan 20, 2021
Published in print: Apr 1, 2021
Discussion open until: Jun 20, 2021
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.