Hybrid heuristics for the cut ordering planning problem in apparel industry

Cut order planning (COP) is important role in managing the costs of fabric and setup for starting a new section in the manufacturing process. An efficient COP solution can significantly reduce the total cost of manufacturing in the apparel industry. With the development of artificial intelligence (AI) techniques for optimization problems in industrial applications, two heuristics are combined in this study: simulated annealing-based genetic algorithm (SA-GA) and tabu-search-based genetic algorithm (TS-GA) algorithms to solve the COP problem. Subsequently, we compare three metaheuristics: simulated annealing, genetic algorithm, and tabu search. The COP problem is formulated as a mixed-integer programming model. The objective is to determine the length of used fabrics and the number of opened sections while minimizing the total cost. The optimal parameters of each technique are determined using the design of experiment approach. An ANOVA test was performed to compare the performances of all presented algorithms. Results show that these AI techniques are applicable to the apparel industry and can reduce the total cost. The results obtained by our proposed methods are close to the optimal solution, in which the standard deviation of the related cost is less than 10%. The results of these algorithms are applicable to large-size problems.