Journal of Korean Society of Transportation (대한교통학회지)

Korean Society of Transportation (대한교통학회)
권호 표지
DOI QR코드

DOI QR Code

Determination of the Optimal Aggregation Interval Size of Individual Vehicle Travel Times Collected by DSRC in Interrupted Traffic Flow Section of National Highway

국도 단속류 구간에서 DSRC를 활용하여 수집한 개별차량 통행시간의 최적 수집 간격 결정 연구

  • PARK, Hyunsuk (ICT Convergence and Integration Research Institute, Korea Institute of Civil Engineering and Building Technology) ;
  • KIM, Youngchan (Department of Transportation Engineering, The University of Seoul)
  • 박현석 (한국건설기술연구원 ICT융합연구소) ;
  • 김영찬 (서울시립대학교 교통공학과)
  • Received : 2016.10.06
  • Accepted : 2017.02.28
  • Published : 2017.02.28
Download PDF
⟨ Previous Next ⟩

Abstract

The purpose of this study is to determine the optimal aggregation interval to increase the reliability when estimating representative value of individual vehicle travel time collected by DSRC equipment in interrupted traffic flow section in National Highway. For this, we use the bimodal asymmetric distribution data, which is the distribution of the most representative individual vehicle travel time collected in the interrupted traffic flow section, and estimate the MSE(Mean Square Error) according to the variation of the aggregation interval of individual vehicle travel time, and determine the optimal aggregation interval. The estimation equation for the MSE estimation utilizes the maximum estimation error equation of t-distribution that can be used in asymmetric distribution. For the analysis of optimal aggregation interval size, the aggregation interval size of individual vehicle travel time was only 3 minutes or more apart from the aggregation interval size of 1-2 minutes in which the collection of data was normally lost due to the signal stop in the interrupted traffic flow section. The aggregation interval that causes the missing part in the data collection causes another error in the missing data correction process and is excluded. As a result, the optimal aggregation interval for the minimum MSE was 3~5 minutes. Considering both the efficiency of the system operation and the improvement of the reliability of calculation of the travel time, it is effective to operate the basic aggregation interval as 5 minutes as usual and to reduce the aggregation interval to 3 minutes in case of congestion.

연구는 국도 단속류 구간에서 DSRC로 수집한 개별차량 통행시간의 대푯값 산정 시 신뢰도를 높이는 최적 수집 간격을 결정하는데 목적이 있다. 이를 위하여, 단속류 구간에서 수집되는 가장 대표적인 개별차량 통행시간의 분포인 양봉형태의 비대칭 분포를 따르는 수집데이터를 활용하고 개별차량 통행시간의 수집 간격 크기를 변화시켜 MSE(Mean Square Error)를 추정함으로 오차가 최소가 되는 최적 수집 간격 크기를 결정한다. MSE 산정을 위한 편의 추정식은 비대칭 분포에서도 활용이 가능한 t-분포의 최대 추정 오차식을 활용하였다. 최적 수집 간격 분석을 위한 데이터 수집 간격은 단속류 구간에서 신호정지로 데이터 수집이 정상적으로 결측 되는 1-2분 수집 간격은 제외하고, 3분 이상의 수집 간격만을 대상으로 하였다. 데이터 수집 시 결측을 발생시키는 수집 간격은 결측 데이터 보정처리 과정에서 또 다른 오차를 유발하게 되어 배제하였다. 분석결과 MSE가 최소가 되는 최적 수집 간격은 3-5분이며, 통행시간 증가 시 최적 수집 간격은 3분으로 짧아짐을 확인하였다. 시스템 운영의 효율성과 통행시간 대푯값 산정의 신뢰도 향상을 모두 고려할 때 기본 수집 간격은 기존과 같이 5분으로 운영하고, 정체 시는 3분으로 수집 간격을 줄여 운영하는 것이 효과적일 것으로 사료된다.

Keywords

References

  1. Choi D. W. (2012), Determine Optimal Aggregation Interval for Travel Time Estimation on Rural Interrupted Traffic Flow Considering Traffic Flow Characteristics, University of Science and Technology Master's Thesis.
  2. Gajewski B.J., Turner S.M., Eisele W.L., Spiegelman O.H. (2001), ITS Data Archving: Statistical Technique for Inductance Loop Detector Speed Data, Transportation Research Record 1719.
  3. Kim H. J., Kim J. S. (2013), Lecture Statistics, Myung Jin.
  4. Korea Institute of Civil Engineering and Building Technology (2009), Final Report of Design Service for the National Highway ITS of Seoul Regional Construction Management Office 2009(2nd), Seoul Regional Construction Management Office.
  5. Korea Institute of Civil Engineering and Building Technology (2016), Final Report of ITS Operation Management of Seoul National Highway in 2016, Seoul Regional Construction Management Office.
  6. Lim H. S. (2005), A Study on the Optimal Aggregation Interval for Travel Time Estimation on the Rule Arterial Interrupted Traffic Flow, Ajou University Master's Thesis.
  7. Park D. J. (2000), Determining Optimal Aggregation Interval Size for Travel Time Estimation and Forecasting With Statistical Models, J. Korean Soc. Transp., 18(3), Korean Society of Transportation, 55-76.
  8. Park H. S., Kim Y. C. (2014), A Study on the Setting RSE Considering the Reliability of Traffic Information, Conference of Korean Society of ITS, Korean Society of ITS, 257-261.
  9. Yoo J. S., Oh C. S. (1999), Modern Statistics, Park Young Sa.
  10. Yoo S. Y., Rho J. H., Park D. J. (2004), Investigating Optimal Aggregation Interval Size of Loop Detector Data for Travel Time Estimation and Prediction, J. Korean Soc. Transp., 22(6), Korean Society of Transportation, 109-119.