A variable neighborhood descent approach for the two-dimensional bin packing problem

Abstract In this paper, we propose a new variable neighborhood descent (VND) algorithm for the two-dimensional bin packing problem. A knapsack packing heuristic is performed to get the initial solution. The proposed VND involves three neighborhood structures in order to disrupt the actual solution and provide a better exploration of the search space. Computational experiments on a standard benchmark show that our algorithm is competitive with the existing heuristics and metaheuristics in literature in terms of both the quality of the solution and the computing time.