EPSO for solving non-oriented two-dimensional bin packing problem

The non-oriented two-dimensional bin packing problem is dealing with a set of rectangular pieces that need to be packed into identical rectangular bins. Moreover and in order to minimize the number of bins, the pieces are allowed to rotate by 90° without overlapping. There are many real life applications for this operations research problem. Among these applications: loading of boxes to pallets, trucks and containers, packing of box bases on shelves and other applications in the wood and metal industry. In this paper, we propose evolutionary particle swarm optimization algorithm (EPSO) for solving the non-oriented two-dimensional bin packing problem. Extensive numerical investigations are performed to determine the solution quality of the proposed algorithm. Moreover, the performance of our proposed algorithm is compared with a best known greedy algorithm published in the literature.