Last-mile delivery with multiple deliverypersons: Formulation and exact solution methods for a two-echelon location-routing problem
Year: 2026
Type: article
Abstract: There is an increasing demand for cost- and time-efficient last-mile delivery due to the expansion of home-delivery systems. To respond to this need, many companies and academics have focused on inventive delivery schemes to reduce costs and improve the service level offered to customers. These efforts include integrating vehicles with walking carriers, which additionally dodges traffic and reduces the emission of greenhouse gases and other pollutants. The vehicle routing problem with time windows and multiple deliverypersons models an example of such delivery systems. In this problem, each vehicle may travel with more than one deliveryperson to serve more customers with each stop of the vehicle and reduce the overall time that the vehicle stays parked. As originally defined, it considers that the customers served from each parking location and the routes traveled by the deliverypersons are predefined. We propose a variant of this problem in which both decisions can be optimized, yielding more cost-effective solutions. The novel problem is formally defined and formulated. Theoretical properties and useful lower bounds are introduced and used to propose several valid inequalities. To take advantage of its structure, the proposed formulation is decomposed in a Benders scheme and solved exactly by a branch-and-Benders-cut algorithm. Extensive computational experiments show the suitability of the proposed methodology to solve the problem. Furthermore, managerial insights indicate that the inclusion of the customer clustering and deliveryperson routes in the optimization leads to an average cost reduction of over 10%, with this value being much higher for some instances.
Language: English
Subfield: Industrial and Manufacturing Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $4,040
OpenAlex ID: W7222190871
