Abstract:
The problem of excessive empty backhaul distances is a major challenge in the planning and operation of truckload transportation, in which goods are picked up from an origin and delivered to a destination without mid-route pickups or deliveries. The empty backhaul distances can be reduced by combining two or more truckload trips together to form a sequence of continous move truckload trips. The objectives of this research are to develop a truckload continuous move optimization model and solution algorithms of this problem for large-scale transportain network, incorporating major operational complexities, namely, fleet-commodity compatility, trip-based cost function, and time windows. We present a Continuous Move Optimization Model (CMO), which is based on the set partitioning formulation. We develop two solution approaches -- an exact column-generation-based branch-and-bound algorithm and a heuristic algorithm -- which yield significant empty haul distance reduction under relatively short runtimes, and provide a comparison study measuring the effectiveness and applicability of the two methods. The results show substantial reduction in empty haul distances, ranging from 30% to 42%. Our findings indicate that as problems become larger, empty haul reduction increases but at a decreasing rate. The comparison between the branch-and-bound with column generation approach and the heuristic shows better cost savings with the former and markedly better runtimes with the latter.