Queue Length Studies

Queue Length Studies

A& Q Traffic Data Provides Queue lengths Surveys at signalized intersections is used in performance evaluation and for feedback signal control. Evaluation of performance measures such as intersection delay, travel time and spillback usually requires the queue length probability distribution, which can be derived from the statistics of demand and the signal control laws. For isolated intersections the distribution may be revealed through probabilistic analysis or through simulation. For a network of intersections one must resort to simulation to estimate the joint distribution of queue lengths. But the number of simulations needed to estimate a multi-variate queue length distribution is so large that such procedures have not been reported in the literature. Instead simulations are used to estimate measures such as average delay and travel time. Queue-based feedback control methods are proposed . These methods require knowledge of queue lengths in real time. Since they cannot be measured directly by current detection technology, one must estimate the queue lengths based on other measurements. A simple approach is to use detector vehicle counts at the entrance to the queue (e.g. from an advance detector) and at the exit of the queue (from a stop bar detector) to construct a naive queue estimate as the cumulative difference between the flows of vehicles at the entrance and the exit. But unknown biases in detector counts and random errors make this naive estimate useless, so estimation algorithms propose alternatives.