Simulation-Based Comparison of Holding Strategies for a Multibranch Light Rail Service

Abstract

Light rail transit services face many operational challenges, such as capacity constraints, mixed-traffic interference, and branch junctions. The service plans developed for these lines typically specify a precise schedule for each vehicle. Running time and demand variability, special events, and incidents make it challenging to adhere to schedules. Operators can enact real-time control actions to mitigate delays. This research compared the effectiveness of schedule- and headway-based holding strategies applied en route and at terminals (i.e., dispatching) on a simulation model of the Massachusetts Bay Transportation Authority Green Line, a four-branch light rail line. The effects of control point placement at terminals, along branches, along a central trunk, and in combinations of these three were studied, as were the effects of limiting holding at midroute stations. Holding strategies were compared on the basis of service and passenger-oriented performance. Headway-based holding was found to be a more effective method for ensuring that passengers experienced reasonable wait times within scheduled headways. Holding at terminals appeared to be the most beneficial to passengers; additional holding along the branches and limited holding along the trunk were shown to enhance these benefits. Holding only within the trunk of a multibranch service worsened service because of blockages from held trains.

Improving High-Frequency Transit Performance through Headway-Based Dispatching: Development and Implementation of a Real-Time Decision-Support System on a Multi-Branch Light Rail Line

“…Rolling horizon optimization models have been used by Delgado et al ( 7 ) and Sánchez-Martínez et al ( 8 ) to simulate holding and dispatching times considering capacity constraints. Fabian and Sánchez-Martínez, using a simulation of the Green Line, found that headway-based dispatching at terminals resulted in shorter and more reliable passenger wait times than en route holding alone and holding by schedule ( 3 ).…”

“…Real-time control actions, including holding, expressing, and short-turning, can be implemented in attempts of correcting headway variability as it occurs. Considering holding-type actions, previous simulation-based research shows that dispatching on headways at terminals yields the shortest and most reliable passenger wait times of any single control strategy ( 3 ).…”

Multiline holding based control for lines merging to a shared transit corridor

“…Only a few recent examples have examined real-time holding on a segment of overlapping lines. These studies focused on evaluating different operation schemes and on comparing regularity-based and schedule-based strategies (Hernández et al 2015;Fabian and Sánchez-Martínez 2017). The most relevant example in multiline control is the work of (Argote-Cabanero 2014; Argote-Cabanero, Daganzo, and Lynn 2015) who successfully extended the applicability of an isolated line holding criterion to multiline networks.…”

Evaluation of Subway Bottleneck Mitigation Strategies using Microscopic, Agent-Based Simulation

“…Simulation has been used extensively for evaluation of rail operations. Fabian et al ( 15 ), for example, utilized a data-driven mesoscopic simulation model to evaluate real-time holding strategies on a multi-branch rail system. The simulation results show that regularity and on-time performance could be improved using appropriate holding strategies.…”