Abstract
A motorway weaving section connects a pair of closely spaced entry- and exit-ramps, where intensive lane-changings of merging and diverging vehicles take place over a relatively short space. A detailed trajectory data reveal that a significant proportion of the lane changing at the weaving section exhibits group lane-changing behaviour, in the forms of a lane-changing platoon and simultaneous weaving behaviour. The acceptable gaps are different in group lane-changing and weaving, compared to a single isolated lane-changing. This paper proposes a random utility formulation with explicit representation of these different lane changing mechanisms. The model parameters are calibrated using Maximum Likelihood Estimation technique and individual vehicle trajectory data extracted from video recordings. There are significant differences in behavioural parameter values for lane changing mechanisms. In particular, the results suggest that the relative speed with respect to the current and target lane leaders have varying impacts on the gap acceptance behaviour.
Original language | English |
---|---|
Pages (from-to) | 1-21 |
Number of pages | 21 |
Journal | Transportmetrica B |
Volume | 8 |
Issue number | 1 |
DOIs | |
Publication status | Published - 2 Jan 2020 |
Keywords
- group behaviour
- Lane changing
- leader effect
- random utility maximisation
- weaving section