Demo: Rapid Optimization of Fixed-Line Transit Holding Policies

Abstract

When multiple buses service the same route, they have the potential to become bunched, i.e., the inter-arrival time (headway) is small between some buses and large between others. As the variance in headway times increases, the expected waiting time for all passengers arriving at stops increases. Transit operators can hold buses according to a policy at a set of predefined stops to even headway at the expense of increased travel time. Recent work has shown how to compute the worstcase headway bounds under various classes of holding policies typically employed by transit operators. In this demonstration, we show how to utilize these bounds to optimize the holding policies and locations for a fixed-line transit route. We formulate this as a multiobjective optimization problem: the goal is to minimize the variance in headway times while simultaneously minimizing the maximum additional holding time added across the route. We demonstrate this optimization using data from a real bus route in Nashville, TN to show the benefits of headway time bound calculation in the public transit design and optimization process.