This paper presents a digital freight platform based on a spatio-temporal graph to foster collaboration between shippers and carriers. It offers a data-driven approach to address the complexities of freight transport planning by considering geographical and temporal constraints. Using graph modeling, the platform provides practical solutions that enhance operational efficiency and sustainability in freight transportation within a highly competitive global market. The platform addresses the selective routing problem of multi-depot vehicles with time windows (SFTMDVRPTW), involving confirmed trips-orders already assigned. To solve this problem, a recommendation engine is developed, offering a two-phase route planner for trucker: first, the system identifies transport tasks based on the temporal and geographical criteria of a truck route using an interactive map. Then, graph traversal algorithms are applied to explore the constructed graph and generate truck schedules aimed at maximizing profitability while considering both mono and multi-objective models. The performance of the platform is evaluated using numerical experiments, demonstrating its relevance in route planning and improving logistics efficiency. • Data-driven digital freight platform built on a graph model. • Integrated spatio-temporal graph modeling of freight transportation operations. • Two-phase recommendation engine generates feasible and flexible truck routes. • Bi-objective optimization balances economic and operational performance. • Supports selective routing and collaborative transportation planning.
Saoud et al. (Wed,) studied this question.