Journal of Navigation and Port Research (한국항해항만학회지)

Korean Institute of Navigation and Port Research (한국항해항만학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on Application of ARIMA and Neural Networks for Time Series Forecasting of Port Traffic

항만물동량 예측력 제고를 위한 ARIMA 및 인공신경망모형들의 비교 연구

  • Shin, Chang-Hoon (Department of Logistics Engineering, National Korea Maritime University) ;
  • Jeong, Su-Hyun (Graduate school of National Korea Maritime University)
  • 신창훈 (한국해양대학교 물류시스템공학과) ;
  • 정수현 (한국해양대학교 대학원)
  • Received : 2011.01.24
  • Accepted : 2011.02.25
  • Published : 2011.02.28
Download PDF
⟨ Previous Next ⟩

Abstract

The accuracy of forecasting is remarkably important to reduce total cost or to increase customer services, so it has been studied by many researchers. In this paper, the artificial neural network (ANN), one of the most popular nonlinear forecasting methods, is compared with autoregressive integrated moving average(ARIMA) model through performing a prediction of container traffic. It uses a hybrid methodology that combines both the linear ARIAM and the nonlinear ANN model to improve forecasting performance. Also, it compares the methodology with other models in performance for prediction. In designing network structure, this work specially applies the genetic algorithm which is known as the effectively optimal algorithm in the huge and complex sample space. It includes the time delayed neural network (TDNN) as well as multi-layer perceptron (MLP) which is the most popular neural network model. Experimental results indicate that both ANN and Hybrid models outperform ARIMA model.

예측의 정확성은 비용의 감소나 고객서비스의 제고를 위해 필수적으로 선행되어야 하기에 현재까지도 많은 연구자들에 의해 연구되고 있는 분야이다. 본 연구에서는 국내 항만의 컨테이너 물동량 예측에 있어 대표적인 비선형예측모형인 인공신경망모형과 ARIMA모형에 대한 비교연구를 수행하는데 목적을 두었고, 컨테이너 물동량 예측력 제고를 위해 ARIMA모형과 인공신경망(ANN)모형을 결합한 하이브리드모형을 사용해 다른 모형들과 예측성과를 비교하고자 한다. 특히 인공신경망모형의 네트워크 구조 설계에 부분에 있어 방대하며 복잡한 탐색공간에서도 전역해 찾기에 효과적인 기법으로 알려져 있는 유전알고리즘을 사용함과 동시에 인공신경망의 대표적인 모형으로 알려진 다층 퍼셉트론(MLP)뿐만 아니라 시간지연네트워크(TDNN)를 사용해 예측성과를 비교하였다. 그 결과 ANN모형과 하이브리드모형이 ARIMA모형보다 더 뛰어난 예측성과를 보이는 것으로 나왔다.

Keywords

References

  1. 모수원, 김창범(2003), “해상물동량의 추정과 예측,” 해운물류연구, 제37호, pp.1-18.
  2. 신창훈, 강정식, 박수남, 이지훈(2008), “하이브리드 ARIMA-신경망 모델을 통한 컨테이너물동량 예측에 관한 연구,” 항해항만학회지, 제32권, 제1호, pp.81-88. https://doi.org/10.5394/KINPR.2008.32.1.081
  3. 전찬영(1999), “교차검증을 통한 우리나라 중장기 항만물동량 예측," 해양정책연구, 제14권, pp.127-157.
  4. 전찬영, 송주미(2007), “인공신경망모형의 항만물동량 예측 적용에 관한 연구," 해운물류연구 제53호, pp.65-82.
  5. 지원철(1995), “신경망을 이용한 시계열 분석: M1 - Competition Data에 대한 예측성과 분석," 한국전문가시스템학회지 창간호, pp.135-148.
  6. Box, G. E. P. and Jenkins, G. M.(1976), “Time SeriesAnalysis Forecasting and Control,” Holden-Day, SanFrancisco.
  7. Chatfield, C.(1993), “Neural Networks: ForecastingBreakthrough or Passing Fad?,” International Journalof Forecasting Vol. 9, No.1, pp.1-3. https://doi.org/10.1016/0169-2070(93)90043-M
  8. Fishwick, P. A.(1989), “Neural Network Models inSimulation: A Comparison with Traditional ModelingApproaches,” Proceedings of Winter SimulationConference, pp.702–710.
  9. Geman, G., Bienenstock, E. and Doursat, R.(1992),“Neural Networks and Bias/Variance Dilemma,” NeuralComputation, Vol. 4, No. 1, pp.1-58. https://doi.org/10.1162/neco.1992.4.1.1
  10. Ghysels, E., Granger, C. W. J and Siklos, P. L.(1996),“Is Seasonal Adjustment a Linear or Nonlinear DataFiltering Process?,” Journal of Business & EconomicStatistics, Vol. 14, No. 3, pp.374-386. https://doi.org/10.2307/1392449
  11. Hansen, J. V., McDonald, J. B and Nelson, R. D(1999),“Time Series Prediction with Genetic-AlgorithmDesigned Neural Networks: An Empirical ComparisonWith Modern Statistical Models,” ComputationalIntelligence, Vol. 15, No. 3, pp.171-184. https://doi.org/10.1111/0824-7935.00090
  12. Ittig, P. T.(1997), “A Seasonal Index for Business,”Decision Science, Vol. 28, No.2, pp.335-355. https://doi.org/10.1111/j.1540-5915.1997.tb01314.x
  13. Jhee, W. C. and Lee, J. K.(1993), “Performance ofNeural Networks in Managerial Forecasting,”International Journal of Intelligent Systems inAccounting and Finance Management, Vol. 2, No. 1,pp.55-71. https://doi.org/10.1002/j.1099-1174.1993.tb00034.x
  14. Kim, H. J. and Shin, K. S.(2007), “A Hybrid Approachbased on Neural Networks and Genetic Algorithms forDetecting Temporal Patterns in Stock Markets,”Applied Soft Computing Journal, Vol. 7, No. 2,pp.569-576. https://doi.org/10.1016/j.asoc.2006.03.004
  15. Lapedes, A. and Farber, R.(1987), “Nonlinear SignalProcessing using Neural Networks: Prediction andSystem Modeling,” Los Almos National Laboratory,Technical Report, LA-UR-87-2662.
  16. Miller, D. M and Williams, D.(2004), “DampingSeasonal Factors: Shrinkage Estimators for theX-12-ARIMA Program,” International Journal ofForecasting, Vol. 20, No. 4, pp.529-549. https://doi.org/10.1016/j.ijforecast.2004.03.002
  17. Nelson, M., Hill, T., Remus, T. and O’Connor, M.(1999), “Time Series Forecasting using NNs: Should the Databe Deseasonalized First?,” Journal of Forecasting, Vol.18, No. 5, pp.359-367. https://doi.org/10.1002/(SICI)1099-131X(199909)18:5<359::AID-FOR746>3.0.CO;2-P
  18. Ripley, B. D.(1993), “Statistical Aspects of NeuralNetworks," In Networks and Chaos : Statistical andProbabilistic Aspects ed Barndorff-Nielsen, O. E.,Jensen, J. L. and Kendall, W. S., London.
  19. Sharda, R. and Patil, R. B.(1992), “ConnectionistApproach to Time Series Prediction: An EmpiricalTest,” Journal of Intelligent Manufacturing, Vol. 3, No.5, pp.317-323. https://doi.org/10.1007/BF01577272
  20. Taskaya-Temizel, T. and Casey, M. C.(2005), “AComparative Study of Autoregressive Neural NetworkHybrids,” Neural Networks, Vol.18 pp.781-789. https://doi.org/10.1016/j.neunet.2005.06.003
  21. Waibel, A., Hanazawa, T., Hinton, G., Shikano, K. andLang, K.(1989), “Phoneme Recognition using TimeDelay Neural Networks,” IEEE Transactions onAcoustics, Speech and Signal Process, Vol. 37, No. 3,pp.328-339. https://doi.org/10.1109/29.21701
  22. Weigend, A. S., Rumelhart, D. E. and Huberman, B.A.(1990), “Backpropagation, Weight-elimination andTime Series Prediction,” Connectionist Models:proceedings of the 1990 Summer School, pp.105-116.
  23. White, H.(1988), “Economic Prediction using NeuralNetworks: The Case of IBM Daily Stock Returns,”Proceedings of the IEEE International Conference onNeural Networks, 2, pp.451-458. https://doi.org/10.1109/ICNN.1988.23959
  24. Zhang, G. P.(2003), “Time Series Forecasting using aHybrid ARIMA and Neural Network Model,”Neurocomputing, Vol. 50, No. 2, pp.159-175. https://doi.org/10.1016/S0925-2312(01)00702-0
  25. Zhang, G. P. and Qi, M.(2005), “Neural NetworkForecasting for Seasonal and Trend Time Series,”European Journal of Operational Research, Vol. 160,No. 2, pp.501-514. https://doi.org/10.1016/j.ejor.2003.08.037

Cited by

  1. A Prediction of Marine Traffic Volume using Artificial Neural Network and Time Series Analysis vol.20, pp.1, 2014, https://doi.org/10.7837/kosomes.2014.20.1.033
  2. A Forecast Method of Marine Traffic Volume through Time Series Analysis vol.19, pp.6, 2013, https://doi.org/10.7837/kosomes.2013.19.6.612
  3. GMDH-based hybrid model for container throughput forecasting: Selective combination forecasting in nonlinear subseries vol.62, 2018, https://doi.org/10.1016/j.asoc.2017.10.033
  4. A Study on Estimating Container Throughput in Korean Ports using Time Series Data vol.40, pp.2, 2016, https://doi.org/10.5394/KINPR.2015.40.2.57