A data-based opportunity identification engine for collaborative freight logistics based on a trailer capacity graph

Abstract

Logistics operators participating in horizontal collaboration can gain economic benefits and being better placed to meet environmental goals. Data-based approaches provide a viable, albeit suboptimal, solution that can enable real-time collaborative order sharing. Conventional data-based approaches for identifying collaboration (order sharing) opportunities are typically based on origin-destination (OD) matching between trips and shipments from different collaborating companies. This, however, prevents the exploitation of en-route collaboration opportunities. Hence, we propose a practical data-based engine for identifying collaboration opportunities during shipment planning stages that enables shipments to be matched according to both the OD and trailer trip routes. The engine is based on a multigraph approach, called the trailer capacity graph (TCG) approach. We further enhance the engine to improve its computational performance for real-time operations. Numerical experiments based on real-world data from two logistics companies show that the TCG approach identifies a significantly larger number of opportunities, and provides a higher total distance saving than conventional OD-based matching. The experiments also demonstrate that with trailer route approximation and route shape simplification, this engine allows trade-offs between the computational performance and the effectiveness of opportunity identification, which implies that the engine can be flexibly tailored according to user preferences.