Decomposition and Holt-Winters Techniques Enhanced by Whale Optimization Algorithm: Case Study of PM2.5 Forecasting in 8 Northern Provinces of Thailand
Article Sidebar
Main Article Content
Abstract
The objective of this study was to evaluate the effectiveness of two forecasting models: the hybrid Whale Optimization Algorithm with Holt-Winters (WOA-HW) and hybrid Whale Optimization Algorithm with Decomposition (WOA-D), in forecasting weekly PM2.5 concentrations in 8 provinces in Northern Thailand. These models were compared to classical decomposition (Classic-D) and grid search Holt-Winters (Classic-HW) models using a training dataset of 130 weeks. The results show that WOA-HW and WOA-D outperformed the classical models, with WOA-D exhibiting significantly lower RMSE than Classic-D. Although WOA-HW had RMSE values comparable to Classic-HW, it required less time to find the optimal parameters. For long-term forecasts over two years, a test dataset of 105 weeks was used, with RMSE, MAE, and MAPE serving as evaluation metrics. The results indicated that the optimal model varied for each province: WOA-HW was best for Lampang and Chiang Rai, WOA-D for Mae Hong Son and Phayao, Box-Jenkins for Nan and Phrae, Classic-HW for Lamphun, and Classic-D for Chiang Mai. The two-year forecast for all provinces revealed a distinct seasonal pattern in PM2.5 concentrations, with levels exceeding health-impact thresholds from December to April.
Article Details
References
Dorigo, M., 1992, Optimization, Learning and Natural Algorithms, Ph.D. Thesis, Politecnico di Milano, Italy, 140 p.
Dorigo, M. and Stützle, T., 2004, Ant Colony Optimization, MIT Press, Cambridge, MA, USA, 305 p.
Karaboga, D., 2005, An Idea Based on Honey Bee Swarm for Numerical Optimization, Technical Report-TR06, Engineering Faculty, Computer Engineering Department, Erciyes University, 10 p.
Yang, X.S., and Deb, S., 2009, Cuckoo Search Via Lévy Flights, Proceedings of World Congress on Nature & Biologically Inspired Computing (NaBIC), pp. 210-214.
Pan, W.T., 2011, A New Evolutionary Computation Approach: Fruit Fly Optimization Algorithm, Proceedings of the Conference of Digital Technology and Innovation Management, Taipei, Taiwan, pp. 382–391.
Yang, X.S., 2012, Flower pollination algorithm for global optimization, Unconventional Computation and Natural Computation, Lecture Notes in Computer Science. 7445: 240-249.
Kaewpaengjuntra, S., Somhom, S. and Saenchan, L., 2010, Electricity consumption forecasting model using hybrid Holt-Winters exponential smoothing and artificial bee colony algorithm. Information Technology Journal. 6(1): 12-17. (in Thai)
Assis, M.V.O., Carvalho, L.F., Rodrigues, J.J.P.C. and Provençal, M.L., 2013, Holt-Winters statistical forecasting and ACO metaheuristic for traffic characterization. Proceeding of 2013 IEEE International Conference on Communications (ICC), Budapest, Hungary, pp. 2524-2528.
Janta, S., Thaninpong, P. and Minsan.W., 2018, Holt-Winters Exponential Smoothing Forecasting using Flower Pollination Algorithm based Parameter Estimation, Proceeding of The 6th Academic Science and Technology Conference 2018, Bang Phli, Samut Prakan, pp. BS-40-46. (in Thai)
Available Source: https://drive.google.com/file/d/1zVB-vqbNqfTqr-ZYUmiY8wS2uN1amtOy/view?usp=drive_link, July 1, 2023. (in Thai)
Jiang, W., Wu, X., Gong, Y., Yu, W. and Zhong, X., 2020, Holt–Winters smoothing enhanced by fruit fly optimization algorithm to forecast monthly electricity consumption, Energy. 193: 116779.
Mauricio, C.C, and Ostia, C.F., 2023, Cuckoo search algorithm optimization of Holt-Winter method for distribution transformer load forecasting. Proceeding of the 2023 9th International Conference on Control, Automation and Robotics (ICCAR), Beijing, China, pp. 36-42.
Mirjalili, S. and Lewis, A., 2016, The whale optimization algorithm, Advances in Engineering Software. 95: 51-67.
Nadimi‑Shahraki M.H., Zamani, H., Varzaneh Z.A. and Mirjalili, S., 2023, A systematic review of the whale optimization algorithm: theoretical foundation improvements and hybridizations, Archives of Computational Methods in Engineering. 30: 4113-4159.
Minsan, W., Saengngammuang, N., Taninpong, P. and Thumronglaohapun, S., 2021, Comparing Methods of Optimization in Solver of Excel 2019 and Whale Optimization Algorithm, UTK Research Journal. 15(2): 106-120. (in Thai)
Minsan, W. and Minsan, P., 2023, Incorporating Decomposition and the Holt-Winters Method into the Whale Optimization Algorithm for Forecasting Monthly Government Revenue in Thailand, Science & Technology Asia. 28(4): 38-53.
Minsan, W. and Minsan, P., 2024, Decomposition and Holt-Winters Enhanced by the Whale Optimization Algorithm for Forecasting the Amount of Water Inflow into the Large Dam Reservoirs in Southern Thailand, Journal of Current Science and Technology. 14(2): 1-16.
Ministry of Natural Resources and Environment, Pollution Control Department, Available Source: http://air4thai.pcd.go.th/webV2/history/# , July 30, 2023. (in Thai)
Box, G.E.P., Jenkins, G.M., Reinsel, G.C. and Ljung G.M., 2015, Time Series Analysis: Forecasting and Control, 5th ed., Wiley, 712 p.
Hochreiter, S. and Schmidhuber, J., 1997, Long short-term memory, Neural Computation. 9(8): 1735-1780.
Singhaworawong, P., (2020), Forecasting PM2.5 in Chiang Mai using long short-term memory models, Master Thesis, Srinakharinwirot University, Bangkok, 76 p. (in Thai)
Google Colab, Overview of Colaboratory Features, Available Source: https://colab.research.google.com/notebooks/intro.ipynb , May 1, 2023.