The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

A Study of Travel Time Prediction using K-Nearest Neighborhood Method

K 최대근접이웃 방법을 이용한 통행시간 예측에 대한 연구

  • Lim, Sung-Han (Highway & Transportation Research Division, Korea Institute of Construction Technology) ;
  • Lee, Hyang-Mi (Highway & Transportation Research Division, Korea Institute of Construction Technology) ;
  • Park, Seong-Lyong (ITS & Road Environment Division, Ministry of Land, Infrastructure and Transport) ;
  • Heo, Tae-Young (Department of Information and Statistics, Chungbuk National University)
  • 임성한 (한국건설기술연구원 도로교통연구실) ;
  • 이향미 (한국건설기술연구원 도로교통연구실) ;
  • 박성룡 (국토교통부 첨단도로환경과) ;
  • 허태영 (충북대학교 자연과학대학 정보통계학과)
  • Received : 2013.09.06
  • Accepted : 2013.10.15
  • Published : 2013.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

Travel-time is considered the most typical and preferred traffic information for intelligent transportation systems(ITS). This paper proposes a real-time travel-time prediction method for a national highway. In this paper, the K-nearest neighbor(KNN) method is used for travel time prediction. The KNN method (a nonparametric method) is appropriate for a real-time traffic management system because the method needs no additional assumptions or parameter calibration. The performances of various models are compared based on mean absolute percentage error(MAPE) and coefficient of variation(CV). In real application, the analysis of real traffic data collected from Korean national highways indicates that the proposed model outperforms other prediction models such as the historical average model and the Kalman filter model. It is expected to improve travel-time reliability by flexibly using travel-time from the proposed model with travel-time from the interval detectors.

통행시간은 교통정보 중에서 가장 대표적이고 이용자 선호도가 높은 정보이다. 본 연구에서는 일반국도를 대상으로 실시간 시스템에 적용 가능한 통행시간 예측 방법을 개발하고자 하였다. 통행시간 예측방법으로 비모수적 접근 방법인 K 최대근접이웃 방법을 적용하였다. K 최대근접이웃 방법은 데이터에 대한 특별한 가정이 필요 없고, 모수 추정 과정이 필요 없어 실시간 교통관리시스템에 적합하다. K 최대근접이웃 방법의 우수성을 평가하기 위해 교통 분야에서 많이 적용되고 있는 이력자료 평균방법과 칼만 필터방법을 선정하여 평균절대백분율오차와 변동계수를 통해 평가하였다. 평가 결과 K 최대근접이웃 방법이 이력자료 평균방법과 칼만 필터방법에 비해 우수한 것으로 분석되었다. 통행시간 정보 제공 시 본 연구에서 개발된 방법을 통해 도출된 통행시간과 구간검지기로부터 관측된 통행시간을 탄력적으로 적용함으로써 통행시간 정보의 신뢰도를 향상시킬 수 있을 것으로 기대된다.

Keywords

References

  1. Cha, W. O. and Huh, M. Y. (2008). k-nearest neighbor-based approach for the estimation of mutual information, Communications for Statistical Applications and Methods, 15, 17-26.
  2. Jang, J., Beak, N., Kim, S. and Byun, S. (2004). Dynamic travel time prediction using AVI data, Journal of Korean Society of Transportation, 22, 169-175.
  3. Kim, J., Rho, J., Park, D. and Namkoong, S. (2006). Comparative analysis of link travel times: Departure time based vs arrival time based, Journal of Korea Spatial Planning Review, 48, 71-86.
  4. Lam, W. H. K., Tang, Y. F., Chan, K. S. and Tam, M. L. (2006a). Short-term hourly traffic forecasts using Hong Kong annual Traffic Census, Transportation, 33, 291-310. https://doi.org/10.1007/s11116-005-0327-8
  5. Lam, W. H. K., Tang, Y. F. and Tam, M. L. (2006b). Comparison of two non-parametric models for daily traffic forecasting in Hong Kong, Journal of Forecasting, 25, 173-192. https://doi.org/10.1002/for.984
  6. Lee, Y. and Lee. J. (2002). A study on link travel time estimation methodology for traffic information service (Determination of an adequate sample size), Journal of Korean Society of Transportation, 20, 55-67.
  7. Lim, S. (2011). Travel Time Prediction Simultaneously using Point and Interval Detector Date, A dissertation submitted for the degree of Doctor of Philosophy, The University of Seoul.
  8. Lim, S. and Lee, C. (2011). Data fusion algorithm improves travel time predictions, IET Intelligent Transport Systems, 5, 302-309. https://doi.org/10.1049/iet-its.2011.0014
  9. Ministry of Land, Infrastructure and Transport (2005). Highway Capacity Manual.
  10. Park, C. S. and Huh, K. (2010). Optimal k-nearest neighborhood classifier using genetic algorithm, Communications for Statistical Applications and Methods, 17, 17-27. https://doi.org/10.5351/CKSS.2010.17.1.017
  11. Park, S.-H., Bang, S.-W. and Jhun, M.-S. (2011). On the use of sequential adaptive nearest neighbors for missing value imputation, The Korean Journal of Applied Statistics, 24, 1249-1257. https://doi.org/10.5351/KJAS.2011.24.6.1249
  12. Smith, B. L. and Demetsky, M. J. (1997). Traffic flow forecasting: Comparison of modeling approaches, Journal of Transportation Engineering ASCE, 123, 261-266. https://doi.org/10.1061/(ASCE)0733-947X(1997)123:4(261)
  13. Tam, M. L. and Lam, W. H. K. (2009). Short-term Travel Time Prediction for Congested Urban Road Networks, Transportation Research Board 88th Annual Meeting 2009 Paper 9-2313.
  14. You, J. and Kim, T. J. (2005). Towards Developing a Travel Time Forecasting Model for Location-based Services: A Review, In Reggiani, A., and L. A. Schintler, Eds. Methods and Models in Transport and Telecommunications, Springer, Berlin, Heidelberg, 45-61.

Cited by

  1. Development of an Evaluation Index for Identifying Freeway Traffic Safety Based on Integrating RWIS and VDS Data vol.32, pp.5, 2014, https://doi.org/10.7470/jkst.2014.32.5.441
  2. Expressway Travel Time Prediction Using K-Nearest Neighborhood vol.34, pp.6, 2014, https://doi.org/10.12652/Ksce.2014.34.6.1873