<![CDATA[In a flexographic printing machine, color changeover actually takes up excessive time and serves as a bottleneck in the corrugated cardboard packaging manufacturer. This causes inefficiency and reduces the output. The objective of this study is to optimize production scheduling using a Genetic Algorithm (GA) to minimize sequence-dependent setup times (SDST). The current manual scheduling method used in the studied flexographic printing environment does not explicitly account for the sequence-dependent color changeover structure considered in this study, creating opportunities for improved scheduling performance through optimized production sequencing. To address this problem, the SDST problem was first formulated as a Mixed-Integer Linear Programming (MILP) model based on systematic observation and historical production data. A customized GA was then developed to generate high-quality scheduling solutions, and a systematic parameter tuning process was conducted, identifying an effective configuration of population size 500, 100 generations, and mutation rate 0.7 to ensure stable convergence. Results show that the proposed GA framework reduced setups times by over 70.78% (equivalent to 5.04 hours per shift) compared to the facility’s existing manual scheduling baseline and also outperformed the greedy heuristic benchmark, consistently achieving an average setup time of 8,466.63 seconds across multiple runs with low variability, demonstrating reliable performance. This study shows that GA can serve as a practical approach for optimizing scheduling in flexographic printing and closely related sequence-dependent color changeover production contexts, although the current model is based on deterministic conditions with fixed job sequences, which may limit responsiveness to dynamic production uncertainties such as machine breakdowns or rush orders, suggesting the need for future enhancements using simulation and multi-objective optimization approaches.]]>
Copyrights © 2026
