Solving the 2D Bin Packing Problem by Means of a Hybrid Evolutionary Algorithm

Abstract

Combinatorial optimization problems dealing with 2D bin packing find applications, for example, in the context of transporta- tion/warehousing and for the cutting of glass, wood, and metal. In this work we consider the oriented 2D bin packing problem under free guillotine cutting, a problem in which a set of oriented rectangular items is given which must be packed into a minimum number of bins of equal size. Our algorithm proposal to tackle this problem concerns an evolutionary algorithm that makes heavy use of a randomized one-pass heuristic for constructing solutions. The results of the proposed algorithm are compared to some of the best approaches from the literature. This comparison shows that our algorithms is very competitive to state-of-the-art approaches. In particular, the optimal solutions to four previously unsolved instances were found.