COMBINED TECHNIQUES FOR FORECASTING THE VOLUME OF PACKAGES IN INTERNAL POSTAL TRAFFIC OF SERBIA

Ivana Rogan, Olivera Pronić-Rančić

DOI Number
https://doi.org/10.22190/FUACR220330006R
First page
059
Last page
075

Abstract


The main goal of time series analysis is to explain the main features of the data in a chronological order and in the general case to predict future processes, products, service requirements, etc., using appropriate statistical models. In this paper, time series prediction was performed using a seasonal autoregressive integrated moving average model (SARIMA) in the XLSTAT add-in for Excel environment, as well as two artificial neural network (ANN) models - long short-term memory (LSTM) network and relatively new machine learning technique - extreme learning machines (ELM).  The proposed approaches were used for forecasting the volume of packages in the internal postal traffic of Serbia for the period 2014-2020. A comparison of the obtained modeling results with the original data was made and it was shown that the best modelling results were achieved by using ELM.

Keywords

Time series analysis, forecasting, ANN, SARIMA, LSTM, ELM

Full Text:

PDF

References


Taeho Jo, Machine Learning Foundations: Supervised, Unsupervised, and Advanced Learning. Springer Nature Switzerland AG, 2021.

Manjusha Pandey; Siddharth Swarup Rautaray, Machine Learning: Theoretical Foundations and Practical Applications. Springer Nature Singapore Pte Ltd. 2021.

https://ai.stanford.edu/people/nilsson/mlbook.html

arXiv:2012.03854v2 [stat.AP] 3 Jun 2021. Available: https://arxiv.org/pdf/2012.03854v1.pdf

G.E. Box, G.M. Jenkins, G.C. Reinsel, G. M. Ljung, Time Series Analysis, Forecasting and Control. New Jersey, John Wiley and Sons, 2016.

R. J. Hyndman, G. Athanasopoulos, Forecasting: principles and practice. 3rd edition, OTexts: Melbourne, Australia. OTexts.com/fpp3. Accessed on 26 September 2021.

D. C. S. Bisht (eds.), M. Ram (eds.), Recent Advances in Time Series Forecasting. Boca Raton, CRC Press, 2021.

A. Ticherahine, A. Boudhaouia, P. Wira and A. Makhlouf, “Time series forecasting of hourly water consumption with combinations of deterministic and learning models in the context of a tertiary building,” Int.Conf. on Decision Aid Sciences and Application, 2020.

H. Yu, L. Jingming, R. Sumei, Z. Shuping, “Hybrid Model for Financial Time Series Forecasting – Integration of EWT, ARIMA with The Improved ABC Optimized ELM,” IEEE Access, 1–1. doi:10.1109/access.2020.2987547.

B. Predić, N. Radosavljević, A. Stojčić, “Time Series Analysis: Forecasting Sales Periods in Wholesale Systems,” Facta universitatis, Series: Automatic Control and Robotics, vol. 18, no 3, pp. 177–188, 2019.

M. Paluszek, S. Thomas, MATLAB Machine Learning Recipes: A Problem-Solution Approach. Second Edition, New Jersey, Apress, 2019.

M. Paluszek, S. Thomas, Practical MATLAB Deep Learning: A Project-Based Approach. New Jersey, Apress, 2020.

S. Hochreiter, J. Schmidhuber, "Long short-term memory", Neural Computation. 9 (8): 1735–1780, 1997.

arXiv:1303.5778v1 [cs.NE] 22 Mar 2013.

https://www.mathworks.com/help/deeplearning/ref/sgdmupdate.html .

A. Voelker, I. Kajić, C. Eliasmith, “Legendre Memory Units: Continuous-Time Representation in Recurrent Neural Networks,” 33rd Conference on Neural Information Processing Systems NeurIPS Vancouver, Canada, 2019.

P. Poonia, V. K. Jain, “Short-Term Traffic Flow Prediction: Using LSTM”, IEEE, International Conference on Emerging Trends in Communication, Control and Computing, 2020.

Guang-Bin Huang, Qin-Yu Zhu, Chee-Kheong Siew, "Extreme learning machine: theory and applications," Neurocomputing. 70 (1): 489–501, 2006.

Guang-Bin Huang, Qin-Yu Zhu, K. Z. Mao, Chee-Kheong Siew, P. Saratchandran and N. Sundararajan, "Can threshold networks be trained directly?" in IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 53, no. 3, pp. 187-191, March 2006, doi: 10.1109/TCSII.2005.857540.

G. -B. Huang, H. Zhou, X. Ding and R. Zhang, "Extreme Learning Machine for Regression and Multiclass Classification," in IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), vol. 42, no. 2, pp. 513-529, April 2012, doi: 10.1109/TSMCB.2011.2168604.

Guang-Bin Huang, “What are Extreme Learning Machines? Filling the Gap Between Frank Rosenblatt's Dream and John von Neumann's Puzzle,” Cognitive Computation, volume 7, issue 3 2015.

J. Wang, S. Lu, Shui-Hua Wang, Yu-Dong Zhang, “A review on extreme learning machine,” Multimed Tools Appl, 2021. [Online]. Available: https://doi.org/10.1007/s11042-021-11007-7 .

https://elmorigin.wixsite.com/originofelm .

https://www.programmersought.com/article/231014377/

https://www.xlstat.com/en/.

https://www.mathworks.com/help/econ/estimate-multiplicative-arimamodel-using-econometric-modeler.html .

N. Knežević, N. Glišović, M. Milenković, N. Bojović, “Neural Networks Based on Metaheuristics for Forecasting the Revenue of Postal Services,” (in Serbian), PosTel 2018, pp.33-42, Belgrade, Serbia, 2018.

I. D. Rogan, O.R. Pronić-Rančić, “Forecasting the volume of postal services using Savitzky-Golay filter modification,” Sozopol, Bulgaria, 56th International Scientific Conference on Information, Communication and Energy Systems and Technologies (ICEST), 2021. DOI: 10.1109/ICEST52640.2021.9483459.

I. D. Rogan, O.R. Pronić-Rančić, “SARIMA and ANN Approaches in Forecasting the Volume of Postal Services,” Niš, 15th International Conference on Advanced Technologies, Systems and Services in Telecommunications, TELSIKS, 2021.

C. M. Hurvich, Chih-Ling Tsai, “Regression and time series model selection in small samples”, Biometrika, vol. 76, no. 2, pp. 297-307, 1989.

https://www.posta.rs/eng/stanovnistvo/usluga.aspx?usluga=postal-services/parcel-services-serbia/sending-parcels-within-serbia




DOI: https://doi.org/10.22190/FUACR220330006R

Refbacks

  • There are currently no refbacks.


Print ISSN: 1820-6417
Online ISSN: 1820-6425