A tabu-based variable neighborhood local search for n-vehicles exploration problem

In this paper, a tabu-based variable neighborhood local search (TBVLS) is proposed to solve the n-vehicle exploration problem (NVEP), in which a fleet of n vehicles' trip sequence is determined so as to ensure one of the vehicles visits the farthest distance. In TBVLS, tabu search is employed to conduct iteratively local search around the solution space, which is incorporated with three variable neighborhood local search operators, including swap, insert, and inverse. It is expected via this hybrid algorithm that the ability of searching promising region could be enhanced by the variable neighborhood local search, and the global search around the neighborhood of current solution could be diversified by the three different operators. Numerical results about 14 benchmark instances are provided, and the comparisons suggest that TBVLS could achieve better performances in the 14 instances, which demonstrates the effectiveness of the proposed TBVLS in sloving NVEP.