Hybrid Ant Colony System to Solve a 2 Di

In this paper we present a study of an Ant Colony System (ACS) for the two-dimensional strip packing problem. The objective of this research is to evaluate the effectiveness of incorporating a simple optimization method at each cycle of the ACS. We propose a hybrid approach where local optimization is applied to a subset of newly generated solutions, steering them towards local optimums. Through computational experiments, our ACS algorithm, when integrated with an efficient local search procedure, showcases its ability to compete against an existing genetic algorithm, achieving high-quality solutions in a timely manner. This study highlights the relevance of the two-dimensional Strip Packing Problem (2SPP), which is NP-hard and has significant applications in various industries such as glass, paper, and textiles. The additional constraints imposed in our analysis, including non-rotation of pieces and employing 3-stage level packing patterns, further reflect real-world constraints faced in practical applications. Our findings indicate that the proposed hybrid ACS approach efficiently addresses the complexities of the 2SPP while maintaining low execution times.