A GIS-based decision support tool for route deviation transit scheduling and service design

Urban transit providers are faced with declining ridership on traditional fixed route services in low density suburban areas. Ridership on such routes could be increased by adding more flexibility to the fixed route structure. In a route deviation system, service is provided at a limited number of fixed time stops, and service is also provided on-demand to customers off the standard route. Furthermore, route deviation is being used to meet the requirements of the Americans with Disabilities Act (ADA), which has mandated increased responsibilities for transit agencies. The objective of this dissertation is to investigate the technical and operational feasibility of providing route deviation service in a real-time manner. A geographic information systems (GIS) based decision support system was developed for scheduling real-time route deviation service. The DSS, wholly embedded in ArcView 3.0a, consists of three main components: the input module, the routing and scheduling module, and the output module. The routing and scheduling module constitutes an algorithm to process each trip request and determine if the desired pickup and delivery are feasible given the schedule constraints at the fixed stops. The research also examined two key route deviation design parameters: service area size and schedule slack time distribution. A methodology for sensitivity analysis was developed to elucidate the interaction of these design parameters. The two performance measures used to evaluate the design parameters were the number of feasible deviations and the additional slack time. The methodologies for developing the GIS based DSS and service design were applied to a transit agency in the Hampton Roads Area as a case study. The results indicate that as the area of the service zone is increased the number of feasible deviations that can be accommodated decreases. However, the additional dwell time, or the excess slack time, increases with the reduction in service area. Since these two objectives conflict, a multiobjective 0–1 integer nonlinear optimization model was developed for determining the optimal choice of the design parameters. The methodologies outlined for scheduling and optimally designing a route deviation service will serve as a guide for transit agencies aspiring to implement route deviation service. The research demonstrates the applicability of GIS software for real-time transit operations.