Heuristics for the two-dimensional irregular bin packing problem with limited rotations

Abstract

This paper studies a two-dimensional irregular bin packing problem with some specific degrees of rotations. Despite a large number of studies on the two-dimensional irregular strip packing problem, much less attention has been paid to the two-dimensional irregular bin packing problem (2DIBPP), and most of the current research on 2DIBPP is about free rotations. However, limited by the texture of the material, sometimes only some specific degrees of rotations are allowed. To pack a set of irregular pieces with some specific degrees into multi-stock sheets and maximize the utilization, we propose a heuristic algorithm that combines some pieces into one block for packing and uses residual spaces with reference lines to help select the next piece to be packed. In order to further increase the utilization of bins, we adopt fine-tuning, movement, filling, replacement and swap operations, as well as partial and full repack strategies. The computational results show that the proposed algorithm can obtain superior results on the instances with small number of pieces, relatively large number of types of pieces, large bins and more specific degrees of rotations.