Performance Measurement and Determination of Introduction Criteria for Peak Demand Responsive Transit Service

Abstract

Peak demand responsive transit (peak DRT) services are being proposed recently to supplement urban public transportation systems’ limitations. The peak DRT aggregates passengers with similar origins, destinations, and departure times into groups and provides a demand-responsive and express service for each group. This study evaluated the performance of introducing the peak DRT, then established the criteria to determine O-D pairs to introduce the DRT service. To achieve this, we assumed that the DRT should be introduced for O-D pairs with great demands and inferior public transportation services and identified O-D pairs to introduce the DRT under varying criteria: the minimum demand and the minimum difference of generalized travel time between private car and public transportation. We clustered individual public transportation trips on smartcard data by origins, destinations, and departure times, then generated a DRT route for each cluster and evaluated its cost and benefit. This procedure was repeated under varying criteria, and a criterion that maximizes the benefit-to-cost (B/C) ratio or maximizes the benefit while maintaining B/C at least one was selected. As a result, the number of DRT passengers as well as the benefit of DRT introduction were higher under a criterion maximizing the benefit than a criterion maximizing the B/C. This result implies that the criterion for peak DRT introduction should be determined to maximize the benefit than the B/C to provide the DRT service for more people.