Development and testing of a constrained optimization model for traffic signal plan transition

In advanced traffic signal control systems, traffic signal plan transition is the process of switching from one timing plan to another to accommodate changes in traffic demand. In traditional coordinated system operation, whether plans are set by time of day or by traffic responsive plan selection, traffic flow can be severely disrupted by the sudden changes in traffic signal timing plans. Even in case of advanced traffic control systems such as Sydney Coordinated Adaptive Traffic System (SCATS), the flow disruption during the transition period can cause a temporary, but sharp decline in system performance. Therefore, the quest for a signal plan transition method that minimizes flow disruption during the transition period is well motivated.;Although various transition methods have been developed and refined, existing transition methods are not based on optimizing operational measures of effectiveness during plan transition periods. As the first step in remedying this situation, this research developed a nonlinear mathematical model that provides constrained delay minimization through incremental and simultaneous adjustments in offset, cycle length, and phase splits during plan transitions. A simulation study was performed to comparatively evaluate the delay performance of the proposed versus existing transition methods. Also, statistical analyses using ANOVA and multi-comparison tests were conducted to validate the simulation results. According to the analysis results, currently used transition methods tend to assign an excessive amount of green time to the main street, resulting in additional side street delay without performance improvement for the total transportation network. By contrast, the proposed transition method showed measurable improvement in delay performance under a broad range of traffic and geometric conditions.