Last-mile delivery of e-commerce is a complex mix of simultaneous delivery of goods and pickup of returned goods. With return rates being substantial in the e-commerce sector, this problem is highly relevant. To address this challenge, this study formulates a rolling horizon simultaneous pickup and delivery problem with time windows in a two-way logistics system. The proposed approach integrates a multi-objective mixed-integer linear programming model with a rolling-horizon framework, allowing forward deliveries and reverse pickups to be postponed based on different priorities and penalties, while capturing the interdependence between outbound and return flows. Economic efficiency, environmental impact, and social factors are explicitly incorporated into the objective function. For large-scale instances, an adaptive large neighborhood search with simulated annealing metaheuristic is designed and compared against a weighted-sum mixed integer linear program approach. Numerical experiments on Solomon datasets demonstrate that the proposed method efficiently achieves high-quality solutions, reveals trade-offs among sustainability objectives, and shows how coordinating forward and reverse flows can enhance operational adaptability. The results also highlight an unintended outcome of the proposed rolling-horizon planning: generation and propagation of backlog of demand. This highlights the need for adaptive mechanisms that can manage such issues in multi-period planning.

One of the limitations of ride-sharing is that matched drivers and riders need to have similar itineraries and desired arrival times for ride-sharing to be competitive against other transport modes. By allowing a single transfer at a designated transfer hub, their itineraries need to be only partially similar, and therefore more matching options are created. In this paper, we develop an optimal matching approach that matches riders to drivers, taking into account multi-modal routing options to model competition and collaboration between multiple modes of transport. We allow for transfers between modes and between multiple drivers. We model this as a path-based integer programming problem and we develop a simulated annealing algorithm to efficiently solve realistic large-scale instances of the problem. Our analysis indicates that a single transfer hub can reduce significantly the average generalized cost of riders and the total vehicle hours traveled by creating efficient matches. As opposed to previous studies, our work shows that ride-sharing not only attracts former public transport users but also former private car users. By allowing for intermodal transfers and by choosing the cost parameters such that transfers are favorable, itineraries where commuters use their car first, before sharing a ride on the second part of their journey, becomes an appealing alternative. Multi-modal ride-matching with transfers has the potential to increase ride-sharing, reduce the number of vehicle hours traveled in private cars, and reduce the number of cars that are present in urban areas during peak hours of congestion.

Last-mile delivery is one of the most polluting parts of the supply chain. This is partially caused by increased congestion in urban areas and repetitive stop-and-go traffic. One possible alternative to this is to use micro-mobility to replace large motorized vehicles. However, these vehicles are usually slower and have lower capacity. In this work, we propose a multi-modal logistics system for last-mile delivery that combines the use of trucks, metro and micro-mobility. This innovative type of system uses the metro to distribute the parcels to micro-hubs across the network and uses micro-mobility only for the final part of the parcel's itinerary from the micro-hub to the front door of the customer. We focus on finding the optimal micro-hub locations in such a system. We use a continuum approximation of the operational and tactical decisions which includes routing of the micro-mobility vehicles. The whole problem is then modeled as a Mixed Integer Linear Programming (MILP) model for the strategic decisions regarding the micro-hubs, which include location, capacity, and fleet-assignment decisions. We evaluate the results of a case study of the city of Madrid, which illustrates that a multi-modal last-mile delivery system can significantly improve a traditional last-mile delivery system in terms of operational costs and pollution.

The performance of ridesharing systems is intricately entwined with user participation. To characterize such interplay, we adopt a repeated multi-player, non-cooperative game approach to model a ridesharing platform and its users’ decision-making. Users reveal to the platform their participation preferences over being only riders, only drivers, flexible users, and opt-out based on the expected utilities of each mode. The platform optimally matches users with different itineraries and participation preferences to maximize social welfare. We analytically establish the existence and uniqueness of equilibria and design an iterative algorithm for the solution, for which convergence is guaranteed under mild conditions. A case study is conducted with real travel demand data in Chicago. The results highlight the effect of users’ flexibility regarding mode preferences on system performance (i.e., the average utility of users and the percentage of successful matches). A sensitivity analysis on the level of subsidy and the distribution of utility between matched riders and drivers shows that uneven distributions of utility may lead to a higher percentage of successful matches. Additional insights are provided on the effect of a user's origin and destination locations on their role choice and likelihood to be matched.

Crowd-shipping is a last-mile delivery concept in which commuters pick up and deliver parcels on their pre-existing paths. In urban areas, crowd-shipping circumvents problems that traditional last-mile delivery systems suffer from, such as road congestion and lack of parking spaces, especially if more sustainable modes of transport are utilized, like bikes or e-bikes. Using transfers between crowd-shippers allows for expanding the service area and improving the overall performance. However, as this requires synchronization over space and time, it makes the problem more complex. In this work, we develop a model that can encompass fully heterogeneous crowd-shippers and parcels. Thereby, it allows for both direct time-synchronized transfers as well as intermediate storage at designated parcel lockers. We design a column generation algorithm to solve large-scale realistic instances to optimality. We extend the problem to allow crowd-shippers to carry multiple parcels at the same time and for this, we extend the algorithm to simultaneous column and row generation. We evaluate the performance of our algorithm as well as the potential of crowd-shipping with transfers on a realistic case study of a bike-based crowd-shipping system in Washington DC. Our methods solve realistic instances with 1000 crowd-shippers and 1000 parcels within minutes. The results show that a gain in revenue and service level of 30% can be obtained by allowing transfers. By letting part of the population of crowd-shippers carry two or three parcels at the same time, the revenue and service level can be further increased by 30 to 50%. Maximum locker capacities are shown to be reasonable and are the highest in areas where there is a large gap between the moment when parcels are dropped off and when they are picked up from parcel points, which are mainly in the city center.