Estimation of traffic stream space mean speed from time aggregations of double loop detector data

In one of the very first papers on traffic flow theory back in 1952, Wardrop presented the difference between the space mean speed (SMS) and the time mean speed (TMS) of a group of traveling vehicles, and derived a relationship suitable for estimating TMS, given SMS and the speed variance over SMS. As time goes by, traffic practitioners have tended towards computing TMS instead of SMS, mainly when using double loop detectors, and nowadays this is the usual practice in traffic management centers. Therefore, the useful relationship between TMS and SMS should go the other way around in relation to Wardrop's. Recently, the complementary relationship, suitable for estimating SMS from TMS and the speed variance over TMS, has been proved. However this is not enough, as speed variance is usually not available. The present paper develops a probabilistic method to estimate SMS from TMS without the previous knowledge of speed variance and only using the usual time aggregations of double loop detector data. The main assumption of the method - the normality of vehicle speed distribution - is discussed and a formulation to obtain the expected error of the estimation is derived. The results obtained with test data from the AP-7 highway, near Barcelona in Spain, show that the developed methodology is able to estimate SMS with an average relative error as low as 0.5%.

[1]  M. Cassidy.  BIVARIATE RELATIONS IN NEARLY STATIONARY HIGHWAY TRAFFIC , 1998 .

[2]  Michael J. Cassidy,et al.  Relation Among Average Speed, Flow, and Density and Analogous Relation Between Density and Occupancy , 1997 .

[3]  Joy Dahlgren,et al.  Enhancement of Vehicle Speed Estimation with Single Loop Detectors , 2004 .

[4]  Nicholas J. Garber,et al.  Traffic and Highway Engineering , 1988 .

[5]  Hesham A Rakha,et al.  Estimating Traffic Stream Space Mean Speed and Reliability from Dual- and Single-Loop Detectors , 2005 .

[6]  Nancy L. Nihan,et al.  Can Single-Loop Detectors Do the Work of Dual-Loop Detectors? , 2003 .

[7]  J. Wardrop.  ROAD PAPER. SOME THEORETICAL ASPECTS OF ROAD TRAFFIC RESEARCH. , 1952 .

[8]  J. M. D. Castillo,et al.  ON THE FUNCTIONAL FORM OF THE SPEED-DENSITY RELATIONSHIP--II: EMPIRICAL INVESTIGATION , 1995 .

[9]  Adolf D. May,et al.  Traffic Flow Fundamentals , 1989 .

[10]  Carlos F. Daganzo,et al.  Fundamentals of Transportation and Traffic Operations , 1997 .

[11]  D. L. Gerlough,et al.  Traffic flow theory : a monograph , 1975 .

[12]  Yinhai Wang,et al.  Freeway Traffic Speed Estimation with Single-Loop Outputs , 2000 .

[13]  Benjamin Coifman,et al.  Improved velocity estimation using single loop detectors , 2001 .

[14]  J. M. D. Castillo,et al.  On the functional form of the speed-density relationship—I: General theory , 1995 .

[15]  Frank A. Haight,et al.  Mathematical Theories of Traffic Flow , 2012 .

[16]  Benjamin Coifman,et al.  ESTIMATING MEDIAN VELOCITY INSTEAD OF MEAN VELOCITY AT SINGLE LOOP DETECTORS , 2003 .

[17]  Bruce Hellinga,et al.  Improving freeway speed estimates from single-loop detectors , 2002 .

[18]  C. Jotin Khisty,et al.  Transportation Engineering: An Introduction , 1990 .

[19]  Donald J Dailey,et al.  A statistical algorithm for estimating speed from single loop volume and occupancy measurements , 1999 .

[20]  Baibing Li,et al.  On the recursive estimation of vehicular speed using data from a single inductance loop detector: A Bayesian approach , 2009 .

[21]  J. G. Wardrop,et al.  Some Theoretical Aspects of Road Traffic Research , 1952 .