An alternative last-mile delivery system: Leveraging microhubs and crowdshipping

Source abstract

Freight movements in urban areas have rapidly increased due to the rising demand for e-commerce and same-day delivery, leading to greater traffic congestion, air pollution, and greenhouse gas emissions. Microhubs with automated lockers or parcel stations have recently gained traction as a way to reduce the amount of delivery truck cruising and stopping on busy urban streets. Meanwhile, crowdshipping, a crowdsourced delivery method that utilizes trips people are already making for other purposes to deliver parcels, helps reduce congestion and emissions. In this study, we propose and evaluate an urban delivery service that combines microhubs with crowdshipping (M+C), where microhubs are visited by delivery trucks only and customers are visited by crowdshippers only. We mathematically formulate the truck routing among microhubs as a novel Many-to-Many Pickup-and-Delivery Problem with Split Load (M2MPDPSL) And crowdshipper dispatching and routing is a typical vehicle routing problem (VRP). To efficiently solve the M2MPDPSL, we develop an enhanced Adaptive Large Neighborhood Search (ALNS) with customized operators and validate the solution quality using benchmark instances from the literature. We compare the operational cost of this proposed system with a traditional Depot-Based (DB) delivery system, where all vehicles start their trips from a central depot and visit all customers. Our findings show that the M+C system achieves greater cost efficiency than the DB paradigm, particularly in medium- to high-density customer settings. Additionally, we demonstrate that M+C significantly reduces the number of trucks dispatched and the total vehicle miles traveled (VMT).