Public transport-based crowd-shipping with backup transfers

With the rising urbanization and booming e-commerce, traditional last-mile delivery systems fail to keep up with the exploding demand and to satisfy the need for faster, cheaper, and more environmentally friendly deliveries. Several new approaches are put forward as an alternative to classical delivery systems in this regard, yet none of them offers the same level of flexibility, capacity, reliability, and managerial control by itself. This thesis proposes a new last-mile delivery model that combines several new approaches and technologies to address this issue. More precisely, we suggest using public transit as a backbone network completed by automated service points, crowd-shipping, and backup transfers with zero-emission vehicles to provide a low-cost and environmentally friendly express delivery service. Offering cheap, reliable, and fast transfers between any two locations in an urban area, such a delivery network can provide unique opportunities for small and medium-sized businesses to participate in the e-commerce ecosystem and help large e-tailers to lower their costs and improve service quality. The design problem for the envisioned system is formulated as a two-stage stochastic program, and a branch-and-price (BP) algorithm is devised to solve it. Taking advantage of the nearly decomposable structure that would emerge in possible real-world applications, our study presents the first example of using decomposition branching in a BP framework to enhance computational efficiency. Extensive computational studies on realistic problem instances reveal valuable managerial insights for the proposed system and attest to the efficacy of the suggested methodology.