Genetic algorithms for quality improvement in fabric roll cutting problem: a real-world industry case study

In the textile industry, the challenge of cutting multiple pieces of equal length from n rolls of varying lengths frequently arises after quality inspection, often resulting in significant material waste. This task is further complicated by the need to optimize fabric quality, posing a major challenge for manufacturers. This paper presents two genetic algorithm (GA)-based approaches that integrate defect distribution into the fabric cutting process. The first method, Genetic Algorithm for Roll Cutting Length (GARCL), employs adapted chromosome structures and customized crossover and mutation operators to evenly disperse defects. The second approach, Defect Locator Genetic Algorithm (DLGA), focuses on identifying defect clusters and strategically removing them to enhance overall fabric quality. Experiments on real-world industrial cases demonstrate that both methods perform effectively across diverse scenarios, achieving up to a 47% improvement in fabric quality and a 5% reduction in fabric utilization compared to traditional cutting practices. These findings highlight the potential of the proposed methods to minimize trim loss while improving product quality in textile manufacturing.