DOI QR코드

DOI QR Code

Long-Term Arrival Time Estimation Model Based on Service Time

버스의 정차시간을 고려한 장기 도착시간 예측 모델

  • 박철영 (순천대학교 전기.전자.정보통신공학과) ;
  • 김홍근 (순천대학교 전기.전자.정보통신공학과) ;
  • 신창선 (순천대학교 정보통신공학과) ;
  • 조용윤 (순천대학교 정보통신공학과) ;
  • 박장우 (순천대학교 정보통신공학과)
  • Received : 2017.02.27
  • Accepted : 2017.04.03
  • Published : 2017.07.31

Abstract

Citizens want more accurate forecast information using Bus Information System. However, most bus information systems that use an average based short-term prediction algorithm include many errors because they do not consider the effects of the traffic flow, signal period, and halting time. In this paper, we try to improve the precision of forecast information by analyzing the influencing factors of the error, thereby making the convenience of the citizens. We analyzed the influence factors of the error using BIS data. It is shown in the analyzed data that the effects of the time characteristics and geographical conditions are mixed, and that effects on halting time and passes speed is different. Therefore, the halt time is constructed using Generalized Additive Model with explanatory variable such as hour, GPS coordinate and number of routes, and we used Hidden Markov Model to construct a pattern considering the influence of traffic flow on the unit section. As a result of the pattern construction, accurate real-time forecasting and long-term prediction of route travel time were possible. Finally, it is shown that this model is suitable for travel time prediction through statistical test between observed data and predicted data. As a result of this paper, we can provide more precise forecast information to the citizens, and we think that long-term forecasting can play an important role in decision making such as route scheduling.

버스정보 시스템을 이용하는 시민들은 더 정확한 예측 정보를 원한다. 하지만 평균 기반 단기간 예측 알고리즘을 사용하는 대부분의 버스정보시스템에서는 교통흐름, 신호주기, 정차시간 등의 영향이 고려되지 않기 때문에 많은 오차를 포함하고 있는 실정이다. 따라서 본 논문에서는 오차의 영향요인 분석을 통해 예측정보의 정밀도를 향상시켜 시민들의 편의를 도모하고자 한다. 이에 현재 운영되고 있는 버스정보 시스템의 자료를 토대로 오차의 영향요인을 분석했다. 분석 데이터에서 시간대별 특성과 지리적 여건에 의한 영향이 복합적으로 나타나고, 정차시간과 단위구간속도에 미치는 영향도가 다름을 보였다. 이에 따라 정차시간은 일반화 가법 모형을 사용하여 시간, GPS 좌표, 통과 노선수의 설명변수로 패턴을 구축하고, 단위구간에 대해 은닉 마르코프 모델을 사용하여 교통흐름에 따른 영향도를 고려한 패턴을 구축했다. 패턴 구축의 결과로 정밀한 실시간예측이 가능하고, 노선 통행속도의 장기간 예측이 가능했다. 마지막으로 관측 데이터와 예측 데이터의 통계적 검정 과정을 통해 전구간 예측에 적합한 모델임을 보였다. 본 논문의 결과로 시민들에게 더 정확한 예측 정보를 제공하고, 장기간 예측은 배차시간 등의 의사결정에 중요한 역할을 수행할 수 있으리라 생각한다.

Keywords

References

  1. H. G. Kim, C. Y. Park, D. C. Shin, C. S. Shin, Y. Y. Cho, and J. W. Park, "A Study on Traffic Analysis Using Bus Information System," The KIPS Transactions on Computer and Communication Systems, Vol.5, No.9, pp.261-267, 2016. https://doi.org/10.3745/KTCCS.2016.5.9.261
  2. H. G. Kim, C. Y. Park, C. S. Shin, Y. Y. Cho, and J. W. Park, "Time Series Analysis for Traffic Flow Using Dynamic Linear Model," The KIPS Transactions on Computer and Communication Systems, Vol.6, No.4, pp.179-188, 2017. https://doi.org/10.3745/KTCCS.2017.6.4.179
  3. C. Y. Park, H. G. Kim, C. S. Shin, Y. Y. Cho, and J. W. Park, "Arrival Time Estimation for Bus Information System Using Hidden Markov Model," The KIPS Transactions on Computer and Communication Systems, Vol.6, No.4, pp. 189-196, 2017. https://doi.org/10.3745/KTCCS.2017.6.4.189
  4. S. H. Lee, B. S. Moon, and B. J. Park, "The Bus Arrival Time Prediction using Bus Delay Time," Journal of Korean Society of Transportation, Vol.28, No.1, pp.125-134, Feb., 2010.
  5. Liu. H, "Generalized Additive Model. Ph.D." Thesis, University of Minnesota Duluth, Duluth, MN, USA, 2008.
  6. M. J. Choo, "A Study on the Application of Generalized Additive Model in predicting customer churn of the mobile phone company," Master thesis, Ewha Womans Univ., Korea, 2011.
  7. L. R. RABINER, "A tutorial on Hidden Markov Models and Selected Applications in Speech Recognition," Proceedings of the IEEE, Vol.77, No.2, Feb., 1989.
  8. I. Visser. "Seven things to remember about hidden Markov models: A tutorial on Markovian models for time series," Journal of Mathematical Psychology, Vol.55, pp.403-415, Jul., 2011. https://doi.org/10.1016/j.jmp.2011.08.002
  9. S. H. Lee, B. S. Moon, and B. J. Park, "The Bus Arrival Time Prediction using Markov Chain," The Journal of The Korea Institute of Intelligent Transport Systems, Vol.8, No.3, pp. 1-10, Jun., 2009.
  10. B. S. Choi, H. C. Kang, S, K, Lee, and S. T. Han, "A Study for Traffic Forecasting Using Traffic Statistic Information," The Korean Journal of Applied Statistics, Vol.22, No.6, pp. 1177-1190, Oct., 2009. https://doi.org/10.5351/KJAS.2009.22.6.1177
  11. T. G. Kim, H. C. Ahn, and S. G. Kim, "Predictive Modeling of the Bus Arrival Time on the Arterial using Real-Time BIS Data," Journal of The Korean Society of Civil Engineers, Vol.29, No.1, pp.1-9, Jan., 2009.
  12. Y. Y. Lee, "A Study on Estimate to Link Travel Time Using Traveling Data of Bus Information System," Journal of Korean Society of Transportation, Vol.30, No.3, pp.241-246, 2010.
  13. B. Portugais and M. Khanal, "State-Space Models With Kalman Filtering for Freeway Traffic Forecasting," International Journal of Modern Engineering, Vol.15, No.1, pp.11-14, 2014.
  14. G. Petris, and S. Petrone, "State Space Models in R," Journal of Statistical Software, Vol.41, No.4, pp.1-25, 2011,
  15. G. Petris, S. Petrone, and P. Campagnoli, "Dynamic Linear Models with R," Springer Science Business Media, 2009.
  16. Indrabayu, R. Y. Bakti, I. S. Areni, and A. A. Prayogi, "Vehicle detection and tracking using Gaussian Mixture Model and Kalman Filter," 2016 International Conference on Computational Intelligence and Cybernetics, pp.115-119, Nov., 2016.
  17. C. Zeng and W. Li, "Application of Extended Kalman Filter for tracking high dynamic GPS signal," 2016 IEEE International Conference on Signal and Image Processing (ICSIP), pp.503-507, Aug., 2016.