This sessions explores current research related to bus system operations, tools, and technology.
Multicriteria Mixed Transit Fleet Resource Allocation
Huan Ngo, No OrganizationShow Abstract
Rohan Shah, CDM Smith
Sabyasachee Mishra, University of Memphis
Agencies and policymakers are often challenged with equitable and optimal allocation of funds among transit agencies for not just regular operations and maintenance, but also for active fleet management including purchase of new buses and rehabilitation of aging fleet. The paper models a hierarchical structure of resource allocation where federal support for such fleet management is routed through the state, and ultimately to local transit agencies. The resource allocation framework encompasses multiple dimensions such as selection of different improvement program options (rehabilitation, remanufacturing, and replacement) of a mixed transit fleet spread over a temporally continuous planning period, and applied across a set of agencies in a state. The framework is built on an optimization model for capital allocation among transit agencies with an objective function to maximize passenger miles traveled under several agency-specific budget, capacity and policy constraints, and planning objectives. It is numerically demonstrated on real-world data using a set of transit agencies spread across the state of Tennessee, containing a combined fleet size of 254 buses at various degrees of aging. Results indicate that an average, 40 percent additional mileage is generated through the planning period with the same levels of fleet size, with nearly 30 percent of the fleet receiving some form of improvement treatment per year.
Planning of Fast-Charging Stations for a Battery Electric Bus System Under Energy Consumption Uncertainty
Zhaocai Liu, Utah State UniversityShow Abstract
Ziqi Song, Utah State University
Yi He, Utah State University
This study addresses the planning problem of fast-charging stations for a battery electric bus system considering the energy consumption uncertainty of buses. A robust optimization model that represents a mixed integer linear program was developed with the objective of minimizing the total implementation cost. The model was then demonstrated using a real-world bus system. The performances of deterministic solutions and robust solutions are compared under a worst-case scenario. The results demonstrate that the proposed robust model may provide an optimal plan for a fast-charging battery electric bus system that is robust against the energy consumption uncertainty of buses. The trade-off between system cost and system robustness is also discussed.
Exploring Satisfaction with Arterial BRT and Local Bus in the Twin Cities: A Machine Learning Approach
Xinyi Wu, University of Minnesota, Twin CitiesShow Abstract
Xinyu Cao, University of Minnesota, Twin Cities
This study explores how the service features of an arterial BRT service without dedicated right of ways affect riders’ satisfaction differently from local bus, using the 2016 Rider Survey data in the Twin Cities. Based upon the three factor theory, we employ the gradient boosting algorithm to construct factor structures of the BRT and local bus. Using satisfaction with individual service attributes and overall satisfaction with transit, this study finds that influential attributes and improvement priorities differ between the BRT and local bus. The riders of the arterial BRT respond less sensitively to service attributes than local bus riders, partly because of the superiror performance of the BRT. The results also indicate that the BRT does not require immediate service attention, whereas local bus needs to prioritize the improvements of reliability, travel time, hours of operation, behaviors of other passengers and atmosphere, shelter/station conditions, and personal safety while waiting. This study makes an incremental contribution to the literature on arterial BRT. It also makes a methodological improvement by proposing a new machine learning algorithm to improve the salience of the results and to overcome the limitations of regression.
Large-Scale Transit Signal Priority Implementation: District of Columbia's Path to Success
Bailey Lozner, Kittelson & Associates, Inc. (KAI)Show Abstract
Kevin Lee, Kittelson & Associates, Inc. (KAI)
Ahmed Raja, District Department of Transportation
Mohammad Habib, District Department of Transportation
Burak Cesme, Kittelson & Associates, Inc. (KAI)
In 2016, the District Department of Transportation (DDOT) deployed Transit Signal Priority (TSP) at 195 intersections in highly urbanized areas of Washington, DC. Years prior to deployment, DDOT outlined a multi-year process that would aid in the realization of their ambitious objectives. In collaboration with a broader regional implementation and in partnership with the Washington Metropolitan Area Transit Authority (WMATA), DDOT set out to apply a Systems Engineering driven process to identify, design, test, and accept a large-scale TSP System. Application of this Systems Engineering process led to the successful, widespread deployment of TSP in an urbanized operating environment. While numerous research efforts to date focus on smaller scale efforts, this paper seeks to close gaps in the literature with a process-focused case study, highlighting intersection selection, system verification testing, and validation for acceptance as applied by DDOT for establishment of an operational, large-scale System. The paper concludes with a discussion of lessons learned.