Evaluation of Foundation Models for Water Demand Forecasting in a Zero-Shot Scenario
Resumo
Deploying Time Series Foundation Models (TSFMs) in domain-specific applications without retraining remains an open challenge, particularly in systems characterized by high seasonal volatility. In this work we evaluate Chronos and Moirai for zero-shot daily water demand forecasting using data from a coastal city in Southern Brazil, where seasonal tourism drives significant population fluctuations. Results are compared against statistical and machine learning baselines from the literature. Our findings support that Transformer-based FMs achieve competitive performance, highlighting their viability for water demand forecasting in data-scarce or rapid-deployment scenarios.
Referências
Adamowski, J. et al. (2012). Comparison of multiple linear and nonlinear regression, autoregressive integrated moving average, artificial neural network, and wavelet artificial neural network methods for urban water demand forecasting in montreal, canada. Water Resources Research, 48(1):W01528.
Ansari, A. F. et al. (2024). Chronos: Learning the language of time series. Transactions on Machine Learning Research.
Ansari, A. F. et al. (2025). Chronos-2: Scaling time series foundation models. arXiv preprint arXiv:2510.15821.
Bergmeir, C. and Benítez, J. M. (2012). On the use of cross-validation for time series predictor evaluation. Information Sciences, 191:192–213.
Bommasani, R. et al. (2022). On the opportunities and risks of foundation models.
Brown, T. et al. (2020). Language models are few-shot learners. In Advances in Neural Information Processing Systems, volume 33, pages 1877–1901.
Brownlee, J. (2020). Introduction to Time Series Forecasting with Python: How to Prepare Data and Develop Models to Predict the Future. Machine Learning Mastery, 1.9 edition.
de Souza Groppo, G., Costa, M. A., and Libânio, M. (2019). Predicting water demand: a review of the methods employed and future possibilities. Water Supply, 19(8):2179–2198.
Devlin, J. et al. (2019). BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL), pages 4171–4186, Minneapolis, Minnesota.
Fang, X., Liu, J., Zhou, M., Zhang, H., and Zhao, J. (2024). Review of the mechanism and methodology of water demand forecasting in the socio-economic system. Water, 16(11).
González Perea, R., Camacho Poyato, E., and Rodríguez Díaz, J. (2024). Attention is all water need: Multistep time series irrigation water demand forecasting in irrigation disctrics. Computers and Electronics in Agriculture, 218:108723.
Hyndman, R. J. and Athanasopoulos, G. (2021). Forecasting: Principles and Practice. OTexts, Melbourne, 3 edition.
Karaouli, N., Coquenet, D., Fromont, E., Mermillod, M., and Reyboz, M. (2025). How foundational are foundation models for time series forecasting? In NeurIPS 2025 Workshop on Recent Advances in Time Series Foundation Models (BERT2S), San Diego, CA, USA.
Li, Z. et al. (2025). TSFM-Bench: A comprehensive and unified benchmark of foundation models for time series forecasting. In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining, volume 2, pages 5595–5606, Toronto, Canada.
Liang, Y. et al. (2024). Foundation models for time series analysis: A tutorial and survey. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pages 6555–6565, Barcelona, Spain.
Liao, Y. (2025). LLM4TS: Large language & foundation models for time series. GitHub repository.
Liu, C., Aksu, T., Liu, J., Liu, X., Yan, H., Pham, Q., Savarese, S., Sahoo, D., Xiong, C., and Li, J. (2026). Moirai 2.0: When less is more for time series forecasting.
Nabaei, S. H., Zheng, Z., Chen, D., and Heydarian, A. (2025). Multimodal foundation models for zero-shot water stress forecasting in precision agriculture. In 2025 International Conference on Machine Learning and Applications (ICMLA), pages 1330–1337.
Niknam, A., Zare, H. K., Hosseininasab, H., Mostafaeipour, A., and Herrera, M. (2022). A critical review of short-term water demand forecasting tools—what method should i use? Sustainability, 14(9).
Ramesh, A. et al. (2021). Zero-shot text-to-image generation. In Proceedings of the 38th International Conference on Machine Learning (ICML), volume 139, pages 8821–8831.
Rasul, K. et al. (2024). Lag-Llama: Towards foundation models for probabilistic time series forecasting. In R0-FoMo: Robustness of Few-Shot and Zero-Shot Learning in Foundation Models, NeurIPS Workshop.
Salloom, T., Kaynak, O., and He, W. (2025). A novel deep neural network architecture for real-time water demand forecasting. arXiv preprint arXiv:2512.06714.
Shu, J., Xia, X., Han, S., He, Z., Pan, K., and Liu, B. (2024). Long-term water demand forecasting using artificial intelligence models in the tuojiang river basin, china. PLOS ONE, 19(5):e0302558.
Sit, M. et al. (2020). A comprehensive review of deep learning applications in hydrology and water resources. Water Science and Technology, 82(12):2635–2670.
Stefaniak, A. K., Jaskowiak, P. A., and Weihmann, L. (2024). A case study on water demand forecasting in a coastal tourist city. In Proceedings of the Brazilian Conference on Intelligent Systems (BRACIS), pages 3–17, Belém, Brazil.
Sun, A. Y. and Sun, A. A. (2026). Zero-shot forecasting of streamflow using time series foundation models: are we there yet? Machine Learning: Earth, 2(1):010501.
Vaswani, A. et al. (2017). Attention is all you need. In Proceedings of the 31st Conference on Neural Information Processing Systems (NIPS), volume 30, Long Beach, CA, USA.
Vieira Filho, J., Scortegagna, A., Vieira, A. P. S. D., and Jaskowiak, P. A. (2024). Machine learning for water demand forecasting: Case study in a brazilian coastal city. Water Practice and Technology, 19(5):1586–1602.
Wen, Q. et al. (2023). Transformers in time series: A survey. In Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence (IJCAI-23), number 759, pages 6778–6786.
Woo, J., Liu, C., Kumar, A., Xiong, C., Savarese, S., and Sahoo, D. (2024). Unified training of universal time series forecasting transformers. arXiv preprint arXiv:2402.02592.
Ye, J. et al. (2024). Empowering time series analysis with foundation models: A comprehensive survey.
Ansari, A. F. et al. (2024). Chronos: Learning the language of time series. Transactions on Machine Learning Research.
Ansari, A. F. et al. (2025). Chronos-2: Scaling time series foundation models. arXiv preprint arXiv:2510.15821.
Bergmeir, C. and Benítez, J. M. (2012). On the use of cross-validation for time series predictor evaluation. Information Sciences, 191:192–213.
Bommasani, R. et al. (2022). On the opportunities and risks of foundation models.
Brown, T. et al. (2020). Language models are few-shot learners. In Advances in Neural Information Processing Systems, volume 33, pages 1877–1901.
Brownlee, J. (2020). Introduction to Time Series Forecasting with Python: How to Prepare Data and Develop Models to Predict the Future. Machine Learning Mastery, 1.9 edition.
de Souza Groppo, G., Costa, M. A., and Libânio, M. (2019). Predicting water demand: a review of the methods employed and future possibilities. Water Supply, 19(8):2179–2198.
Devlin, J. et al. (2019). BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL), pages 4171–4186, Minneapolis, Minnesota.
Fang, X., Liu, J., Zhou, M., Zhang, H., and Zhao, J. (2024). Review of the mechanism and methodology of water demand forecasting in the socio-economic system. Water, 16(11).
González Perea, R., Camacho Poyato, E., and Rodríguez Díaz, J. (2024). Attention is all water need: Multistep time series irrigation water demand forecasting in irrigation disctrics. Computers and Electronics in Agriculture, 218:108723.
Hyndman, R. J. and Athanasopoulos, G. (2021). Forecasting: Principles and Practice. OTexts, Melbourne, 3 edition.
Karaouli, N., Coquenet, D., Fromont, E., Mermillod, M., and Reyboz, M. (2025). How foundational are foundation models for time series forecasting? In NeurIPS 2025 Workshop on Recent Advances in Time Series Foundation Models (BERT2S), San Diego, CA, USA.
Li, Z. et al. (2025). TSFM-Bench: A comprehensive and unified benchmark of foundation models for time series forecasting. In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining, volume 2, pages 5595–5606, Toronto, Canada.
Liang, Y. et al. (2024). Foundation models for time series analysis: A tutorial and survey. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pages 6555–6565, Barcelona, Spain.
Liao, Y. (2025). LLM4TS: Large language & foundation models for time series. GitHub repository.
Liu, C., Aksu, T., Liu, J., Liu, X., Yan, H., Pham, Q., Savarese, S., Sahoo, D., Xiong, C., and Li, J. (2026). Moirai 2.0: When less is more for time series forecasting.
Nabaei, S. H., Zheng, Z., Chen, D., and Heydarian, A. (2025). Multimodal foundation models for zero-shot water stress forecasting in precision agriculture. In 2025 International Conference on Machine Learning and Applications (ICMLA), pages 1330–1337.
Niknam, A., Zare, H. K., Hosseininasab, H., Mostafaeipour, A., and Herrera, M. (2022). A critical review of short-term water demand forecasting tools—what method should i use? Sustainability, 14(9).
Ramesh, A. et al. (2021). Zero-shot text-to-image generation. In Proceedings of the 38th International Conference on Machine Learning (ICML), volume 139, pages 8821–8831.
Rasul, K. et al. (2024). Lag-Llama: Towards foundation models for probabilistic time series forecasting. In R0-FoMo: Robustness of Few-Shot and Zero-Shot Learning in Foundation Models, NeurIPS Workshop.
Salloom, T., Kaynak, O., and He, W. (2025). A novel deep neural network architecture for real-time water demand forecasting. arXiv preprint arXiv:2512.06714.
Shu, J., Xia, X., Han, S., He, Z., Pan, K., and Liu, B. (2024). Long-term water demand forecasting using artificial intelligence models in the tuojiang river basin, china. PLOS ONE, 19(5):e0302558.
Sit, M. et al. (2020). A comprehensive review of deep learning applications in hydrology and water resources. Water Science and Technology, 82(12):2635–2670.
Stefaniak, A. K., Jaskowiak, P. A., and Weihmann, L. (2024). A case study on water demand forecasting in a coastal tourist city. In Proceedings of the Brazilian Conference on Intelligent Systems (BRACIS), pages 3–17, Belém, Brazil.
Sun, A. Y. and Sun, A. A. (2026). Zero-shot forecasting of streamflow using time series foundation models: are we there yet? Machine Learning: Earth, 2(1):010501.
Vaswani, A. et al. (2017). Attention is all you need. In Proceedings of the 31st Conference on Neural Information Processing Systems (NIPS), volume 30, Long Beach, CA, USA.
Vieira Filho, J., Scortegagna, A., Vieira, A. P. S. D., and Jaskowiak, P. A. (2024). Machine learning for water demand forecasting: Case study in a brazilian coastal city. Water Practice and Technology, 19(5):1586–1602.
Wen, Q. et al. (2023). Transformers in time series: A survey. In Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence (IJCAI-23), number 759, pages 6778–6786.
Woo, J., Liu, C., Kumar, A., Xiong, C., Savarese, S., and Sahoo, D. (2024). Unified training of universal time series forecasting transformers. arXiv preprint arXiv:2402.02592.
Ye, J. et al. (2024). Empowering time series analysis with foundation models: A comprehensive survey.
Publicado
19/10/2026
Como Citar
SANDLER, Luana Rathunde; JASKOWIAK, Pablo Andretta.
Evaluation of Foundation Models for Water Demand Forecasting in a Zero-Shot Scenario. In: ENCONTRO NACIONAL DE INTELIGÊNCIA ARTIFICIAL E COMPUTACIONAL (ENIAC), 23. , 2026, Cuiabá/MT.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 2631-2642.
ISSN 2763-9061.
DOI: https://doi.org/10.5753/eniac.2026.30068.
