Browse > Article

Development and Application of Imputation Technique Based on NPR for Missing Traffic Data  

Jang, Hyeon-Ho (서울대학교 환경대학원)
Han, Dong-Hui (한국도로공사 도로교통연구원)
Lee, Tae-Gyeong (한양대학교 도시공학과)
Lee, Yeong-In (서울대학교 환경대학원)
Won, Je-Mu (한양대학교 도시공학과)
Publication Information
Journal of Korean Society of Transportation / v.28, no.3, 2010 , pp. 61-74 More about this Journal
Abstract
ITS (Intelligent transportation systems) collects real-time traffic data, and accumulates vest historical data. But tremendous historical data has not been managed and employed efficiently. With the introduction of data management systems like ADMS (Archived Data Management System), the potentiality of huge historical data dramatically surfs up. However, traffic data in any data management system includes missing values in nature, and one of major obstacles in applying these data has been the missing data because it makes an entire dataset useless every so often. For these reasons, imputation techniques take a key role in data management systems. To address these limitations, this paper presents a promising imputation technique which could be mounted in data management systems and robustly generates the estimations for missing values included in historical data. The developed model, based on NPR (Non-Parametric Regression) approach, employs various traffic data patterns in historical data and is designated for practical requirements such as the minimization of parameters, computational speed, the imputation of various types of missing data, and multiple imputation. The model was tested under the conditions of various missing data types. The results showed that the model outperforms reported existing approaches in the side of prediction accuracy, and meets the computational speed required to be mounted in traffic data management systems.
Keywords
Data Management System; Historical Data; Missing Data; NPR; Multiple Imputation;
Citations & Related Records
Times Cited By KSCI : 5  (Citation Analysis)
연도 인용수 순위
1 Mulhern, F. J. and R. J. Caprara (1994), "A nearest neighbor model for forecasting market response", International Journal of Forecasting, Vol. 10, No. 2, pp.191∼207.   DOI   ScienceOn
2 Oswald R. K., Scherer W.T. and Smith B.L.(2000), "Traffic flow forecasting using approximate Nearest Neighbor Nonparametric regression", Research project report for U.S. DOT University transportation center.
3 Qi, Y., and Smith, B.L. (2004), "Identifying nearest-neighbors in a large-scale incident data archive", TRR, 1879, pp.89∼98.   DOI
4 Robinson, P. (1983), "Nonparametric estimators for time series", Journal of Time Series Analysis, Vol. 4, No 3, pp.185∼207.   DOI
5 Smith, B.L. (1995), "Forecasting freeway traffic flow for intelligent transportation system applications", Doctorial dissertation. Department of civil engineering, University of Virginia, Charlottesville.
6 Smith B.L., and Conklin J.H. (2002), "The use of local lane distribution patterns to estimate missing data values from traffic monitoring systems", TRR, Vol. 1811, pp.50∼56.   DOI
7 Smith, B.L., B.M. Williams, and R.K. Oswald (2002), "Comparison of parametric and nonparametric models for traffic flow forecasting", TR Part C, Vol. 10. pp.303∼321.   DOI   ScienceOn
8 Sun, H. et al. (2003), "Use of local linear re- gression for short-term traffic forecasting", TRR, 1936, pp.143∼150.
9 김정연․이영인․백승걸․남궁성(2006), "차량 검지자료 결측 보정처리에 관한 연구(이력자료 활용방안을 중심으로)", 대한교통학회지, 제24권 제7호, 대한교통학회, pp.27∼40.   과학기술학회마을
10 김현석․남두희․임강원․이영인(2007), "순환확률분포를 이용한 교통량 결측자료 보정 모형", 대학교통학회지, 제25권 제4호, 대한교통학회, pp.109∼121.   과학기술학회마을
11 Chen, C., Kwon, J., Rice, J., Skabardonis, A., and Varaiya, P. (2003), "Detecting errors and imputing missing data for single loop surveillance systems", TRR Vol. 1855, pp.160∼167.   DOI   ScienceOn
12 하정아․박재화․김성현(2007), "일반국도 상시 교통량자료를 이용한 교통량 결측자료 추정", 대학교통학회지, 제25권 제1호, 대한교통학회, pp.121∼132.   과학기술학회마을
13 Altman, N.S. (1992), "An introduction to kernel and nearest-neighbor nonparametric regression", The American Statistician, Vol. 46, No. 3, pp.175∼185.
14 Chang, H. et al. (2010), "Dynamic multi- interval bus travel time prediction using bus transit data", Transportmetrica, Vol. 6. No. 1, pp.19∼38.   DOI
15 Ni, D., Leonard, J.D., Guin, A., Feng, C. (2005), "Multiple imputation scheme for overcoming the missing values and variability issues in ITS data", JTE, Vol. 131, Issue 12, pp.931∼938.
16 Davis, G. and Nihan, N. (1991), "Nonparametric regression and short-term freeway traffic fore- casting", JTE, Vol. 117, No. 2, pp.178∼188.
17 Disbro, J.E., Frame, M. (1989), "Traffic flow theory and chaotic behavior", New York State Department of Transportation Report FHWA/NY/SR-98/91, New York.
18 Karlsson, M. and S. Yakowitz (1987), "Rainfall-runoff forecasting methods, old and new", Stochastic Hydrology and Hydraulics, Vol. 1, No. 4, pp.303∼318.   DOI
19 이승재․백남철․권희정(2002), "단기조사 교통량을 이용한 AADT 추정연구", 대한교통학회지, 제20권 제6호, 대한교통학회, pp.59∼68.   과학기술학회마을
20 이승재․백남철․권희정․최대순․도명식(2001), "불규칙변동 분해 시계열분석 기법을 사용한 AADT 추정", 대한교통학회지, 제19권 제6호, 대한교통학회, pp.65∼73.
21 강원의․백남철․윤해경(2004), "AIC(AKaike's Information Criterion)을 이용한 교통량 예측 모형", 대한교통학회지, 제22권 제7호, 대한교통학회, pp.155∼159.   과학기술학회마을