All Issue

2025 Vol.58, Issue 3 Preview Page

Research Article

A study on city-scale monthly water demand forecasting methodology using artificial intelligence

인공지능을 활용한 시·군 단위 월별 생·공용수 예측 방법 연구

YoungDon Choia*
,
Ji Hwan Ohb
,
Sung Hoon Kimc
,
Yong Kyu Limd
,
Kyungwon Jooe
,
Dong Jin Limf
최 영돈a*
,
오 지환b
,
김 성훈c
,
임 용규d
,
주 경원e
,
임 동진f
aPrinciple Researcher, K-water Research Institute, AI Research Lab, Daejeon, Korea
bSenior Specialist, Water Resources Business Department, K-water, Daejeon, Korea
cHead Researcher, K-water Research Institute, AI Research Lab, Daejeon, Korea
dGeneral Manager, Water Resources Planning Department, K-water, Daejeon, Korea
eSenior Researcher, K-water Research Institute, AI Research Lab, Daejeon, Korea
fGeneral Director, Han River Basin Management Department, K-water, Daejeon, Korea
aK-water연구원 연구관리처 AI연구센터 책임연구원
bK-water 수자원사업처 물수요공급분석부 선임위원
cK-water연구원 연구관리처 AI연구센터 센터장
dK-water 수자원기획처 물관리기획부 부장
eK-water연구원 연구관리처 AI연구센터 선임연구원
fK-water 한강유역관리처 처장
*Corresponding Author
31 March 2025. pp. 209-223
PDF XML Citation
Abstract

The current method of water demand forecasting in the National Water Management Plan of South Korea and many developed countries still uses Liter Per Capita Day (LPCD) and future population. In this study, we explored and evaluated city-scale monthly water demand forecasting methods using artificial intelligence to prepare for uncertainty of water demand forecasting due to climate change. First, we collected data (12 years, ’10~21 years) on 22 independent variables, which were used in previous studies for water demand forecasting. Next, 6 feature selection methods were applied to select more influential independent variables among all 22 independent variables for 10 representative cities. Finally, we evaluated three water demand forecasting methods (CASE-1: Random Forest and XGBoost model with all collected variables, CASE-2: Random Forest and XGBoost model with selected more influential variables, and CASE-3: Prophet model with univariate analysis) to present an appropriate water demand forecasting method for each cities.

Keywords
Artificial intelligence
Water demand forecast
Feature selection
Water management plan

우리나라뿐만 아니라 많은 선진국에서는 현재까지 1인 1일당 급수량(Liter Per Capita Day, LPCD)과 장래인구를 이용하여 연단위 장래 용수수요 추정을 해오고 있다. 본 연구에서는 월별 용수수요의 변화와 기후변화로 인한 장래 용수공급의 불확실성에 대응하기 위하여 최근 시계열예측에 많이 적용되고 있는 인공지능 기법을 활용한 적절한 생공용수 용수추정방법을 탐색하여 제시하였다. 먼저 (1) 10년 이상의 용수공급 데이터가 있는 10개 시군을 대상으로 과거 용수 수요예측 연구에 활용된 22개 영향변수와 종속변수인 용수공급량에 대한 데이터를 수집(‘05~21년내 12~17년 데이터)하고, (2) 조사된 22개의 영향변수를 6가지 영향력이 큰 변수선택 기법을 적용하여 빈도수가 높은 5개(동일 빈도로 인해 5개 초과시 동일 빈도변수는 모두 선택)의 영향변수를 선정하였다. 마지막으로 (3) 3가지 CASES (CASE-1: 모든 22개 영향변수를 Random Forest와 XGBoost모델에 적용, CASE-2: 영향력이 큰 영향변수를 Random Forest와 XGBoost모델에 적용, CASE-3: 용수공급량 변수를 Prophet모델의 입력자료로 단변량 분석)을 적용하여 데이터의 신뢰도와 영향변수의 영향력에 따른 시군별 적절한 미래 용수수요 추정방법론을 제시하였다.

키워드
인공지능
용수수요추정
영향변수선정
물관리기본계획
References
1

Breiman, L. (2001). "Random forests." Machine Learning, Vol. 45, pp. 5-32. doi: 10.1023/A:1010933404324

10.1023/A:1010933404324
2

Buck, S., Auffhammer, M., Soldati, H., and Sunding, D. (2020). "Forecasting residential water consumption in California: Rethinking model selection." Water Resources Research, Vol. 56, No. 1, e2018WR023965. doi: 10.1029/2018WR023965.

10.1029/2018WR023965
3

Capt, T., Mirchi, A., Kumar, S., and Walker, W.S. (2021). "Urban water demand: statistical optimization approach to modeling daily demand." Journal of Water Resources Planning and Management, Vol. 147, No. 2. doi: 10.1061/(asce)wr.1943-5452.0001315.

10.1061/(ASCE)WR.1943-5452.0001315
4

Chandrashekar, G., and Sahin, F. (2014). "A survey on feature selection methods." Computers and Electrical Engineering, Vol. 40, No. 1, pp. 16-28. doi: 10.1016/j.compeleceng.2013.11.024.

10.1016/j.compeleceng.2013.11.024
5

Chen, T., and Guestrin, C. (2016). "XGBoost: A scalable tree boosting system." KDD '16: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, U.S., pp. 785-794. doi: 10.1145/2939672.2939785.

10.1145/2939672.2939785
6

Choi, S.H., and Kim, S.H. (2006). "Trend analysis of domestic water consumption depending upon the characteristics of using tap water and economical parameters." Proceeding of the Korea Water Resources Association Conference, pp.121-125.

7

Cruz-Nájera, M.A., Treviño-Berrones, M.G., Ponce-Flores, M.P., Terán-Villanueva, J.D., Castán-Rocha, J.A., Ibarra-Martínez, S., Santiago, A., and Laria-Menchaca, J. (2022). "Short time series forecasting: Recommended methods and techniques." Symmetry, Vol. 14, No. 6, 1231. doi: 10.3390/sym14061231.

10.3390/sym14061231
8

Fattah, J., Ezzine, L., Aman, Z., El Moussami, H., and Lachhab, A., (2018). "Forecasting of demand using ARIMA model." International Journal of Engineering Business Management, Vol. 10. doi: 10.1177/1847979018808673.

10.1177/1847979018808673
9

Hariharan, J., Izadi Moud, H., Sands, K.. Capano, C., Stockinger, K., and Vowels, J. (2022). "Benchmarking drinking water consumption during construction phase." Construction Research Congress 2022, Arlington, VA, U.S., Vol. 1-A, pp. 560-568. doi: 10.1061/9780784483954.058.

10.1061/9780784483954.058
10

Kim, H.S., Lee, D.J., Park, N.S., and Jung, K.S. (2008). "Analysis on statistical characteristics of household water end-uses." Journal of the Korean Society of Civil Engineers, B, Vol. 28, No. 5B, pp. 603-14. doi: 10.12652/KSCE.2008.28.5B.603.

11

Maußner, C., Oberascher, M., Autengruber, A., Kahl, A., and Sitzenfrei, R. (2025). "Explainable artificial intelligence for reliable water demand forecasting to increase trust in predictions." Water Research, Vol. 268, 122779. doi: 10.1016/j.watres.2024.122779.

10.1016/j.watres.2024.12277939546974
12

Ministry of Environment (ME) (2014). Manual for forecasting water demand in water supply systems.

13

Nunes Carvalho, T.M., de Souza Filho, F. de A., and Porto, V.C. (2021). "Urban water demand modeling using machine learning techniques: Case study of Fortaleza, Brazil." Journal of Water Resources Planning and Management, Vol.147, No. 1, 05020026. doi: 0.1061/(ASCE)WR.1943-5452.0001310.

10.1061/(ASCE)WR.1943-5452.0001310
14

Oh, J.H., Lim, D.J., Kim, I.K., Shin, J.B., and Ryu, J.S. (2022). "Evaluation and comparison of water balance and budget forecasts considering the domestic and industrial water usage pattern." Journal of Korea Water Resources Association, Vol. 55, No. 11, pp. 941-953. doi: 10.3741/JKWRA.2022.55.11.941.

10.3741/JKWRA.2022.55.11.941
15

Pourmousavi, M., Nasrollahi, H., Najafabadi, A.A., and Kalhor, A. (2022). "Evaluating the performance of feature selection techniques and machine learning algorithms on future residential water demand." Water Supply, Vol. 22, No. 8, pp. 6833-6854. doi: 10.2166/ws.2022.243.

10.2166/ws.2022.243
16

Runge, J., Bathiany, S., Bollt, E., Camps-Valls, G., Coumou, D., Deyle, E., Glymour, C., Kretschmer, M., Mahecha, M.D., Muñoz-Marí, J. et al. (2019). "Inferring causation from time series in Earth system sciences." Nature Communications, Vol. 10, No. 1, 2553. doi: 10.1038/s41467-019-10105-3.

10.1038/s41467-019-10105-331201306PMC6572812
17

VanBerlo, B., Ross, M.A.S., and Hsia, D. (2021). "Univariate long-term municipal water demand forecasting." arXiv abs/ 2105.08486.

18

Xenochristou, M., Hutton, C., Hofman, J., and Kapelan, Z. (2020). "Water demand forecasting accuracy and influencing factors at different spatial scales using a gradient boosting machine." Water Resources Research, Vol. 56, No. 8, e2019WR026304. doi: 10.1029/2019WR026304.

10.1029/2019WR026304
19

Yang, M., Gao, E., Wang, G., Li, D., Zhou, W. and Zhou, X. (2024). "Medium and long-term regional water demand prediction using Harris hawks optimisation - backpropagation neural network model." Scientific Reports, Vol. 14, No. 1, 27763.

10.1038/s41598-024-78980-539533063PMC11557846
20

Yoo, S.H., and Yang, C.Y. (2005). "Analyzing the determinants of the piped water consumption in the large cities with consideration of non-response data." Economic Research, Vol. 23, No. 1, pp. 223-246.

21

Zanfei, A., Brentan, B.M., Menapace, A., Righetti, M., and Herrera, M. (2022). "Graph convolutional recurrent neural networks for water demand forecasting." Water Resources Research, Vol. 58, No. 7, e2022WR032299. doi: 10.1029/2022WR032299.

10.1029/2022WR032299
Information
  • Publisher :KOREA WATER RESOURECES ASSOCIATION
  • Publisher(Ko) :한국수자원학회
  • Journal Title :Journal of Korea Water Resources Association
  • Journal Title(Ko) :한국수자원학회 논문집
  • Volume : 58
  • No :3
  • Pages :209-223
  • Received Date : 2023-10-26
  • Revised Date : 2025-01-06
  • Accepted Date : 2025-01-13
  • DOI :https://doi.org/10.3741/JKWRA.2025.58.3.209
Journal Informaiton Journal of Korea Water Resources Association Journal of Korea Water Resources Association
Online Submission
Archives
  • scopus
  • KCI-수자원
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close