Described in this paper is the queen streetcar model, a large fortran program currently under development that simulates light-rail vehicle operations on the queen route in toronto. This service operates in mixed traffic and on a reserved right-of-way over a 21-mi route. There are 140 passenger stops and 38 traffic signals along the route, and three separately scheduled streetcar services operate overportions of the route. the model is designed to analyze the impact of a range of opearting policies on the regularity of streetcar service. Specifically, the model is intended to allow examination of operating procedures and then permit comparison with alternate means ofregulating service, such as alternative short-turn strategies; use of a centralized automatic vehicle monitoring system; and introduction of traffic signal priorities for transit vehicles, reserved rights-of-way, or larger capacity vehicles, or both. the model proceeds by computing the amount of time that each streetcar will spend in a logical set of "states" within each link within the network, where streetcar states include moving in a link, loading and unloading passengers at a stop, and so forth. The model simulates operations duringa 2:00 to 7:00 p.m. Weekday period using 5-sec time increments. At each 5-sec interval, each streetcar currently in the system is examined to determine if it will remain in its current state for at leastanother 5 sec or if it is about to go to its next state. If the latter is the case, the appropriate next state is determined, the amount of time that the car will spend in this state is computed, and thesystem records are updated accordingly. Major sources of randomnesswithin the model include passenger arrival rates and boarding times, and delays as a result of interactions with other traffic within the shared right-of-way. this paper appeared in transportation research record no. 1152, rail rapid transit and planning tools. For covering abstract see irrd no 818412.
Abstract