Evaluation of Various Machine Learning Methods to Predict Istanbul’s Freshwater Consumption

Abstract

Planning, organizing, and managing water resources is crucial for urban areas and metropolitans. Istanbul is one of the largest megacities, with a population of over 15 million. The large volume of water demand and increasing scarcity of clean water resources make long-term planning necessary for this city, as sustained water supply requires large-scale investment projects. Successful investment plans require accurate projections and forecasting for freshwater demand. This study considers different machine learning methods for freshwater demand forecasting for Istanbul. Using monthly consumption data provided by the municipality since 2009, we compare forecasting accuracies of ARIMA, Holt-Winters, Artificial Neural Networks, Recursive Neural Networks, Long-Short Term Memory, and Simple Recurrent Neural Network models. We find that the monthly freshwater demand of Istanbul is best predicted by Multi-Layer Perceptron and Seasonal ARIMA. From the predictive modeling perspective, this result is another indication of the combined usage of conventional forecasting models and novel machine learning techniques to achieve the highest forecasting accuracy.

Keywords

• Water Management
• Machine Learning
• Neural Networks
• Autoregressive Models

References

1. Adamowski, J. F. (2008). Peak daily water demand forecast modeling using artificial neural networks. Journal of Water Resources Planning and Management, 134(2), 119-128.
2. Altunkaynak, A., Özger, M., Çakmakci, M. (2005). Water consumption prediction of Istanbul city by using fuzzy logic approach. Water resources management, 19, 641-654.
3. Bata, M. T. H., Carriveau, R., Ting, D. S. K. (2020). Short-term water demand forecasting using nonlinear autoregressive artificial neural networks. Journal of Water Resources Planning and Management, 146(3), 04020008.
4. Burak, ZS., Bilge, A.H., Ülker, D. (2021). Assessment and simulation of water transfer for the megacity Istanbul, Phys. Geogr., 43(6): 784-808.
5. Caiado, J. (2010). Performance of combined double seasonal univariate time series models for forecasting water demand. Journal of Hydrologic Engineering, 15(3), 215-222.
6. Celik, OI., Gazioglu, C. (2022). Coast type based accuracy assessment for coastline extraction from satellite image with machine learning classifiers, The Egyptian Journal of Remote Sensing and Space Science, 25 (1), 289-299.
7. Chatfield, C. (1978). The Holt‐winters forecasting procedure. Journal of the Royal Statistical Society: Series C (Applied Statistics), 27(3), 264-279.
8. Chatfield, C., Yar, M. (1988). Holt‐Winters forecasting: some practical issues. Journal of the Royal Statistical Society: Series D (The Statistician), 37(2), 129-140.

9. Chen, L. (2011). Genetic least squares support vector machine approach to hourly water consumption prediction. Journal of Zhejiang University (Engineering Science), 45(6), 1100-1103.
10. Cleveland, R. B., Cleveland, W. S., McRae, J. E., Terpenning, I. (1990). STL: A seasonal-trend decomposition. J. Off. Stat, 6(1), 3-73.
11. Contreras, J., Espínola, R. Member, S., Nogales, FJ (2003). ARIMA models to predict next-day electricity prices, 18(3), 1014-1020.
12. DiPietro, R., Hager, G. D. (2020). Deep learning: RNNs and LSTM. In Handbook of medical image computing and computer assisted intervention (pp. 503-519). Academic Press.
13. Essi̇en, E., Jesse, E., Igbokwe, J. (2019). Assessment of Water Level in Dadin Kowa Dam Reservoir in Gombe State Nigeria Using Geospatial Techniques, International Journal of Environment and Geoinformatics, 6(1), 115-130. doi.10.30897/ijegeo. 487885.
14. Falah, F., Rahmati, O., Rostami, M., Ahmadisharaf, E., Daliakopoulos, I. N., Pourghasemi, H. R. (2019). Artificial neural networks for flood susceptibility mapping in data-scarce urban areas. In Spatial modeling in GIS and R for Earth and Environmental Sciences (pp. 323-336). Elsevier.
15. Gazioğlu, C., Yücel, Z.Y., Doğan, E. (1998). Uydu Verileri İle İstanbul Boğazi ve Yakin Çevresindeki İçme Suyu Havzalarina Genel Bir Bakiş., Büyükşehirlerde atık su yönetimi ve deniz kirlenmesi kontrolu sempozyumu. 18-20 Kasım 1998,
16. Goksel, C., Musaoglu, N., Gurel, M., Ulugtekin, N., Tanik, A., Seker, D. Z. (2006). Determination of land-use change in an urbanized district of Istanbul via remote sensing analysis. Fresenius Environmental Bulletin, 15(8 A), 798–805. 4
17. Guo, G., Liu, S., Wu, Y., Li, J., Zhou, R., Zhu, X. (2018). Short-term water demand forecast based on deep learning method. Journal of Water Resources Planning and Management, 144(12), 04018076.
18. Hassoun, M. H. (1995). Fundamentals of artificial neural networks. MIT press.
19. Hekimoğlu, M. (2022). Markdown Optimization with Generalized Weighted Least Squares Estimation. International Journal of Computational Intelligence Systems, 15(1), 109.
20. Ho, S. L., Xie, M., Goh, T. N. (2002). A comparative study of neural network and Box-Jenkins ARIMA modeling in time series prediction. Computers & Industrial Engineering, 42(2-4), 371-375.
21. Hu, P., Tong, J., Wang, J., Yang, Y., de Oliveira Turci, L. (2019, June). A hybrid model based on NN and Bi-LSTM for urban water demand prediction. In 2019 IEEE Congress on evolutionary computation (CEC) (pp. 1088-1094). IEEE.
22. Jain, A. K. M., J., Mohiuddin, KM (1996). Artificial Neural Networks: A Tutorial. IEEE Computer Society (29), 31-44. Kingma, D. P., Adam, B. J. (2015). A method for stochastic optimization. CoRR. 2014; abs/1412.6980. ArXiv preprint arXiv:1412.6980.
23. Liu, J., Savenije, H. H., Xu, J. (2003). Forecast of water demand in Weinan City in China using WDF-ANN model. Physics and Chemistry of the Earth, Parts A/B/C, 28(4-5), 219-224.
24. McCulloch, W. S. y Pitts, W (1943), A logical calculus of the ideas immanent in nervous activity. Bull. of Math. Biophysics, 5, 116.
25. Mu, L., Zheng, F., Tao, R., Zhang, Q., Kapelan, Z. (2020). Hourly and daily urban water demand predictions using a long short-term memory based model. Journal of Water Resources Planning and Management, 146(9), 05020017.
26. Sak, H., Senior, A., Beaufays, F. (2014). Long short-term memory based recurrent neural network architectures for large vocabulary speech recognition. arXiv preprint arXiv:1402.1128.
27. Savun-Hekimoğlu, B., Erbay, B., Hekimoğlu, M., Burak, S. (2021). Evaluation of water supply alternatives for Istanbul using forecasting and multi-criteria decision making methods. Journal of Cleaner Production, 287, 125080.
28. Smolak, K., Kasieczka, B., Fialkiewicz, W., Rohm, W., Siła-Nowicka, K., Kopańczyk, K. (2020). Applying human mobility and water consumption data for short-term water demand forecasting using classical and machine learning models. Urban Water Journal, 17(1), 32-42.
29. Stevenson, S. (2007). A comparison of the forecasting ability of ARIMA models. Journal of Property Investment & Finance, 25(3), 223-240.
30. Tang, Z., Fishwick, P. A. (1993). Feedforward neural nets as models for time series forecasting. ORSA journal on computing, 5(4), 374-385.
31. Tiwari, M. K., Adamowski, J. (2013). Urban water demand forecasting and uncertainty assessment using ensemble wavelet‐bootstrap‐neural network models. Water Resources Research, 49(10), 6486-6507.
32. Xia, Y., Xiang, M., Li, Z., Mandic, D. P. (2018). Echo state networks for multidimensional data: Exploiting noncircularity and widely linear models. In Adaptive Learning Methods for Nonlinear System Modeling (pp. 267-288). Butterworth-Heinemann.
33. Yücel, Z.Y. Gazioğlu, C., Doğan, E., Kaya, H. (2002). Uzaktan Algilama ve CBS/B ile Ömerli Baraji ve Yakin Çevresinin İzlenmesi, Türkiye'nin Kıyı ve Deniz Alanları IV Ulusal Konferansı Bildiriler Kitabı