Prediction of container throughput at Thailand’s major ports with error correction and parameter selection techniques emphasizes optimal smoothing period
Thoranin Sujjaviriyasup
Maritime Business Review, 2025, vol. 10, issue 2, 166-186
Abstract:
Purpose - A combined approach of additive Holt–Winters, support vector regression, simple moving average and generalized simulated annealing with error correction and optimal parameter selection techniques emphasizing optimal smoothing period in residual adjustment is developed and proposed to predict datasets of container throughput at major ports. Design/methodology/approach - The additive Holt–Winters model describes level, trend and seasonal patterns to provide smoothing values and residuals. In addition, the fitted additive Holt–Winters predicts a future smoothing value. Afterwards, the residual series is improved by using a simple moving average with the optimal period to provide a more obvious and steady series of the residuals. Subsequently, support vector regression formulates a nonlinear complex function with more obvious and steady residuals based on optimal parameters to describe the remaining pattern and predict a future residual value. The generalized simulated annealing searches for the optimal parameters of the proposed model. Finally, the future smoothing value and the future residual value are aggregated to be the future value. Findings - The proposed model is applied to forecast two datasets of major ports in Thailand. The empirical results revealed that the proposed model outperforms all other models based on three accuracy measures for the test datasets. In addition, the proposed model is still superior to all other models with three metrics for the overall datasets of test datasets and additional unseen datasets as well. Consequently, the proposed model can be a useful tool for supporting decision-making on port management at major ports in Thailand. Originality/value - The proposed model emphasizes smoothing residuals adjustment with optimal moving period based on error correction and optimal parameter selection techniques that is developed and proposed to predict datasets of container throughput at major ports in Thailand.
Keywords: Maritime transportation; Port management; Optimal smoothing period; Combined approach (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:
Downloads: (external link)
https://www.emerald.com/insight/content/doi/10.110 ... d&utm_campaign=repec (text/html)
https://www.emerald.com/insight/content/doi/10.110 ... d&utm_campaign=repec (application/pdf)
Access to full text is restricted to subscribers
Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
HTML/Text
Persistent link: https://EconPapers.repec.org/RePEc:eme:mabrpp:mabr-09-2024-0068
DOI: 10.1108/MABR-09-2024-0068
Access Statistics for this article
Maritime Business Review is currently edited by Chin-Shan Lu and Tsz Leung Yip
More articles in Maritime Business Review from Emerald Group Publishing Limited
Bibliographic data for series maintained by Emerald Support ().