Unrelated parallel batch processing machine scheduling for production systems under carbon reduction policies: NSGA-II and MOGWO metaheuristics.

In recent years, the problem of extensive energy consumption and carbon emission in manufacturing companies has been highlighted. Efficient scheduling strategies can help governments and production managers to control the raised environmental concerns resulting from this problem. The goal of this research is to address a green scheduling problem for unrelated parallel batch processes, as a group of highly energy-intensive machines, by considering the emission reduction regulation for the first time. To this aim, two new cases of unrelated parallel batch processing machine scheduling problems under the tax and cap-and-trade policies are defined. The problems are formulated using a mixed-integer linear mathematical model, where the goal is to minimize the makespan and total cost of the production system simultaneously. The total cost objective function includes the earliness and tardiness penalties, machine purchasing costs, and carbon emission costs. Also, several other features, such as setup time, release time, and the machines' purchasing decisions, are considered to provide a more realistic framework. As a solution approach, an interactive fuzzy programming method is presented to establish the single-objective counterpart of the models. Moreover, regarding the complexity of the problem, non-dominated sorting genetic algorithm-II (NSGA-II) and multiobjective gray wolf optimizer (MOGWO) algorithms are utilized as two multiobjective metaheuristics algorithms for large-size instances, for which the parameters are calibrated using the Taguchi design of experiments. Extensive analyses are established by solving several numerical instances. The outputs prove the superiority of the presented framework compared to the traditional models. The results also demonstrate that the newly developed cases can reduce the total cost of the production system by about 5.87 and 0.56% under the tax and cap-and-trade regulations, respectively. Besides, the statistical analysis of the results shows that NSGA-II outperforms the MOGWO algorithm for both cases. [ABSTRACT FROM AUTHOR]