Citation

Ma T, Ahn S. Transp. Res. Rec. 2008; 2088: 138-147.

Copyright

(Copyright © 2008, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

10.3141/2088-15

PMID

unavailable

Abstract

Data from two freeway locations show that absent lane changing, the relationships between congested speed and spacing are linear in most ranges of congested speed. The relationships are also found to be statistically indifferent across lanes, indicating that a single general model would suffice for all lanes. Moreover, speed-spacing relationships for different types of lane changers (discretionary versus mandatory due to merging and diverging) are also found to be statistically indifferent. Anticipation and relaxation periods, defined as durations of perturbed car-following behavior before and after a lane-change maneuver, were measured for lane changers and the vehicles immediately behind the followers. Lane changers exhibited anticipation and relaxation periods of 8 and 14 s, respectively. The followers in the initial lanes exhibited anticipation and relaxation periods of 4 and 8 s, which are smaller than the periods for the followers in the target lanes (12 and 16 s) as well as the periods for the lane changers. The findings suggest that car-following behavior on a freeway can be predicted according to Newell's simplified car-following theory, and a single model can be used without much loss of accuracy for different lanes. The findings also suggest that the anticipation and relaxation processes are independent of lane-changing types but are different for lane changers and followers.