A tabu search heuristic for the truck and trailer routing problem

Two new construction heuristics and a tabu search heuristic are presented for the truck and trailer routing problem, a variant of the vehicle routing problem. Computational results indicate that the heuristics are competitive to the existing approaches. The tabu search algorithm obtained better solutions for each of 21 benchmark problems. Many real-life vehicle routing applications include the use of trailers. Whenever a truck and a trailer can be treated as a single vehicle, that means the trailer is never uncoupled, a normal vehicle routing problem can be solved. However, as is the case in the truck and trailer routing problem, customers may exist that are not reachable by trailer. For this reason, the trailer has to be uncoupled and left behind at a parking place and has to be again picked up at a later point in time on the route. Besides routing, the problem therefore includes also the decision of determining the best parking place and the number of times a trailer should be uncoupled. The purpose of this article is to introduce two simple, but efficient, construction heuristics for this problem and to present a tabu search heuristic with a variable number of sub-tours for further improvement.