Integration planning of freight deliveries into passenger bus networks: Exact and heuristic algorithms

With the increasing population living in cities, a growing number of small daily urban freight deliveries are performed, typically by private companies. Recently, more environmentally friendly urban logistics services have emerged to mitigate the negative effects of such activities. One example is the integration of freight deliveries into bus networks, traditionally dedicated to passenger transportation, to perform urban logistics activities within cities. In this paper, the integration of the freight delivery process into the urban bus passenger network is addressed where freight parcels are dropped by clients at bus hubs located outside the city center, transported by bus services from the hub to bus stops located in the city center, and delivered to the destination address by a last mile operator. Since bus vehicles supporting both passenger and freight flows need to be physically adapted, the aim is to support the decision-maker to select the minimum number of bus services that must be adapted for freight transportation. The optimization problem considers the freight demand uncertainty in terms of number of freight parcels, destination address, delivery time windows and last mile operator constraints which are modelled by a set of demand scenarios. An exact method based on an integer linear programming (ILP) and two heuristic algorithms based on a greedy randomized adaptive search procedure (GRASP) are proposed. The results show that the proposed optimization methods are efficient, giving valuable insights to stakeholders, in the fields of policy and practice, for the strategic decision of selecting the minimum number of buses to be physically adapted for freight transportation. In particular, the results show that all proposed optimization methods are of interest in practice since the type of problem instances for which each method is more efficient is clearly identified in the obtained computational results. Moreover, in the early stages of the integrated passenger and freight flows service, the impact on the required number of adapted bus services is mainly given by the last mile operator capacity of delivering freight from bus hubs to final parcel destinations, while the other factors (delivery time windows and distributions parcel destination addresses) do not have a significant impact on the required number of bus services.