Improving transit service connectivity : the application of operations planning and operations control strategies

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Civil and Environmental Engineering, 2004.Includes bibliographical references (p. 131-133).(cont.) showed that the benefits accrued from coordinating schedules on Route 53 were not significant mainly due to the headway compatibility requirement which reduces the number of transfers amenable for improvement. Greater benefits were encountered when the schedules on the connecting routes were allowed to change as well. For Route 63, schedule coordination is not worth attempting due to the combination of the short six-minute headway on that route and the high variability in vehicle arrival times. On the control side, the practice currently adopted at CTA is to hold a "ready" vehicle at a transfer stop if the connecting vehicle has already arrived and this is likely to be an effective as well as easy-to-implement control policy.Providing direct public transportation service for every origin-destination pair is very expensive if not infeasible for a transit agency, so it relies on the willingness of passengers to transfer. However, transfers usually reduce the attractiveness of transit because they add uncertainty, discomfort, waiting time and cost to most trips. This research focuses on examining different transfer coordination strategies that can reduce the disutility of transfers by minimizing the expected waiting times of transferring passengers. Two models are developed to assess scheduling and real-time holding decision rules for vehicles on a transit corridor. The operations planning model involves the simultaneous application of two planning strategies: changing the terminal departure time and inserting slack time. The operations control model has the capability to utilize any available current network information and to determine optimal dispatch times for vehicles at transfer stops. The two models were tested on a hypothetical corridor to illustrate their applicability in coordinating transfers. On the planning side, results showed that there is a high threshold for introducing slack time to the schedule and that the greatest benefits from schedule coordination are attained when the variance of vehicle arrival times is small and the headway on the analysis corridor is long. On the control side, it was shown that transfer demand is a major driving factor behind any holding recommendation. Moreover, the greatest benefits from real-time coordination occur when the required holding time and the preceding headway of the vehicle on the destination line are short and its following headway is long. The application of the models to two CTA bus routes: 53 and 63,by Bassel J. Younan.S.M