Schedule Coordination in a Multiple Hub Transit Network

Abstract

The timed transfer concept, which seeks to schedule vehicles from various routes to arrive at some transfer stations simultaneously (or nearly so), can significantly improve service quality in transit networks. It has been implemented in some cities but with insufficient efforts to optimize coordination among connecting routes. Our problem is to optimize the headways and slack times jointly for timed transfers to minimize the total costs of operating a multiple-hub transit network. In this paper, a heuristic algorithm is used to optimize the headways and slack times for all coordinated routes. Here, headways are integer multiples of a base cycle to ensure that vehicles on different routes can operate in phase and arrive nearly simultaneously at transfer stations. The results show that as demand decreases, optimized headways increase and the net benefits of coordinated operation also increase. For routes with significantly different demand or route length, coordination with integer-ratio headways is preferable to a single common headway. The sensitivity of the transit service characteristics to various demand and cost parameters is discussed. The results also show that the optimized slack times for routes vary with such variables as headways, vehicle arrival-time variance, transfer volumes, and passenger time values. For routes with high standard deviations of arrivals, it is not worth attempting schedule coordination.