Modeling and Managing the Urban Co-modal Freight Transport System

Abstract：The rapid growth of parcel volume has brought unprecedented pressure to the existing logistic network. Instead of expanding the scale of the transport fleet, an alternative is to exploit the existing public transport systems to carry some freight, which are typically underutilized during off-peak hours. The concept of urban co-modality arises in this context, which proposes to integrated the public transport system into the logistic network. This research is intended to investigate the service network design for the bus-integrated freight transport system, and further the impacts of the adoption of the system on different stakeholders involved. In the currently completed work, we investigated the situation where passenger flows and freight flows are given beforehand. With the objective of minimizing the overall operating cost, the problem was modeled as mixed integer problems using both a node-arc formulation and path-arc formulation. To tackle the problem with practical size, a two-stage algorithm was proposed that solves the relaxed problem by column generation and cutting plane in the first stage, and then obtains feasible solutions in the second stage. The computational experiments proved the effectiveness of the algorithm on large scale problems. In the under-going work, we are intended to investigate the situations where passenger flows are stochastic and where passenger flows and freight flows are stochastic. In such cases, new algorithms for practical scale problems should be designed. Moreover, the demand uncertainty, along with their effects on the service network design will be investigated.