scispace - formally typeset
Search or ask a question

Showing papers by "Margaret Bell published in 1983"


01 Apr 1983
TL;DR: In this paper, the MODIFIED ABRUPT algorithm (commonly called CRASH) is found to generate no statistically significant additional stops or vehicular delay on main roads, irrespective of the level of flow at which the plan-change is made and whether the change is from off-peak to peak or vice versa.
Abstract: Several algorithms exist to implement a change between fixed-time signal plans. These are reviewed and the constraints influencing the process of change are examined. A comparison of actual networks reveals the daily and cyclical variations in flow which prevail in practice and within which the plan-changing process must be seen. Studies using TRANSYT are reported and finally 5 plan-change algorithms are assessed against an "ideal" plan-change, using Monte Carlo simulation techniques. The results suggest that significantly less delay (at the 5% level) and fewer stops are incurred when plan-changes are implemented at off-peak rather than at peak flows. The MODIFIED ABRUPT algorithm (commonly called CRASH), which is in common use, is found to generate no statistically significant (at the 5% level) additional stops or vehicular delay on main roads, irrespective of the level of flow at which the plan-change is made and whether the change is from off-peak to peak or vice versa. The variation in delay caused by implementing a plan-change with the MODIFIED ABRUPT method is found to be less than 2%. It is concluded that plan-changes should always be implemented during off-peak flow conditions and that the commonly used MODIFIED ABRUPT algorithm is the most satisfactory method under prevailing constraints. (Author)

3 citations