Application of time-series analysis to predict vehicle queue length at signalized intersections with heterogeneous traffic conditions

Queue length estimation is imperative as a tool of designing traffic characteristics of an intersection while queue length prediction enables to identify the traffic congestion in advance. This study was aimed to estimate the queue length at signalized intersections having heterogeneous traffic conditions. Furthermore, the study predicts the vehicle queue length of a different intersection with using the estimation results. Therefore, two case studies were conducted in two locations for the queue estimation and queue prediction respectively. The queue estimation study was conducted in a signalized pedestrian crossing in Malabe, Sri Lanka and the queue prediction in a signalized intersection in Armour Street, Sri Lanka where the vehicle density reflected the intense heterogeneity. Heterogeneous traffic conditions were defined as the diversified vehicle composition. The heterogeneity was assimilated with the consideration of Passenger Car Units (PCU) in the measurements of the traffic flow. The influential factors of the queue length were contemplated with the arrival flow, discharge flow and the signal configuration. The time-series analysis integrated the unrestricted Vector Auto Regression (VAR) models as the queue length and the other considered variables which depends on the time. Moreover, the results have indicated a higher accuracy in the queue estimation, practically applied for the prediction constituting to the traffic characteristics of the formed vehicle queue.