A Two-Stage Algorithm for a Dynamic Multi-Trip Vehicle Scheduling Problem

Based on the analysis on the dynamic factors of vehicles in distribution services, a new dynamic multi-trip vehicle scheduling problem which considers vehicles´ dynamic conditions and multi-trip distribution is presented. A two-stage solving strategy for the problem, i.e., the first stage for making distribution plan and the second stage for real time scheduling is addressed. Then a two-stage algorithm for the problem, i.e., the first stage using tabu search algorithm and the second stage using local search algorithm is developed. Thus both the powerful global searching capability of tabu search algorithm and the high converging speed of local search algorithm are made full use of. The effectiveness of this algorithm is demonstrated by experimental computations.