A network design problem formulation and solution procedure for intercity transit services

This study presents a network design problem formulation and solution procedure for an intercity transit service that has multiple routes and serves multiple terminals in the origin and destination cities. The proposed solution procedure consists of three steps: (1) a set of candidate terminals are selected; (2) candidate routes are generated between those terminals using a k-shortest-path algorithm; (3) a mixed integer linear programing model finds the optimal routes, terminals, frequencies, fleet size and depot locations, given a set of constraints and the objective of minimizing total passenger travel time and vehicle deadheading time. The solution procedure was implemented for a newly conceived transit service between Tucson and Phoenix in Arizona, USA. The final routes found by the model look reasonable, and a sensitivity analysis showed that the objective function is most sensitive to changes in the demand satisfaction ratio, and then to the maximum number of routes and fleet size.