Queue Management Project Model: Strategies for the Management of Queues at Upstream Junctions

This Working Paper is one of a series representing work under a SERC grant on queue management strategies for urban traffic control systems, whose objectives are:- (i)to generalise the strategies developed in an earlier study of queue management; (ii)to develop a computer graphics based representation of queue propagation and management; (iii)to test the strategies' applicability and performance in UK networks; (iv)to investigate their incorporation into standard signal optimisation programs. The study is based on earlier work in Bangkok (May et al, 1988) also funded by SERC, in which queue management measures were developed by trial and error to allow for the fact that queues from downstream junctions frequently blocked upstream junctions and could, as a result, unnecessarily obstruct crossing movements thereby reducing junction capacity and spreading congestion to other areas. The methods developed in that study involved:- (i)predicting the growth and decline of queues from downstream junctions; (ii)estimating the speeds of the starting and stopping waves which determine (i); (iii)identifying the most appropriate stage at the upstream junction in which queues should arrive; (iv)identifying the most appropriate time in that stage during which queues should be present; (v)adjusting the split between stages to allow for loss of throughput in blocked stages; (vi)adjusting the cycle time as necessary in the light of (v). It was steps (iii) and (iv) which involved the greatest element of trial and error, and it was accepted that further work was needed to identify alternative strategies for determining:- (i)during which stage to allow queues to block an upstream junction; (ii)how to adjust the stage timings at that junction to allow for the loss of capacity. This paper reviews these approaches and makes recommendations. It considers first of all, in Section 2, the options at an individual junction. It then considers in Section 3, the application of these options in a range of increasingly complex networks.