Abstract
This paper presents some fundamentals of the dynamic lane grouping (DLG) concept, the aim of which is to improve lane utilization under variations in traffic demand. The problem was formulated as a mathematical programming model to determine the optimal lane allocation of an isolated intersection in relation to minimizing the maximum lane flow ratio (defined as the assigned flow divided by the saturation rate). Shared lanes and user equilibrium status for lane use were also considered in the model. Numerical analyses were conducted for four cases with different numbers of lanes, saturation levels, and methods for setting signal timing (adaptive and fixed timing). The results indicated that DLG might provide significant performance improvement in reducing the maximum-flow ratio under the spatial variation in demand. The performance of DLG was also compared with real-time adaptive signal timing by means of the average delay; improvements to the average delay were also observed when the spatial variation was large. The maximum-flow ratio and the average delay of DLG remained almost unchanged for the full spectrum of spatial demand variation when total demand was constant. The results also implied that the more significantly the demand varied, the more benefits the dynamic lane grouping method could potentially provide.
Get full access to this article
View all access options for this article.