A machine learning-based decision-making framework for the green flexible flowshop scheduling problem by considering preventive maintenance

This study focuses on one of the important issues in the manufacturing systems named the flexible flowshop scheduling problem (FFSP) by considering the environmental aspects and preventive maintenance. To this end, the current article developed a novel machine learning-based decision-making framework. Therefore, a mathematical model is proposed to minimize the tardiness, Greenhouse gases (GHG) emissions, and energy consumptions so that the preventive maintenance is considered. Then, to tackle uncertainty, a data-driven approach based on the robust fuzzy optimization and Gradient Boosting Algorithm is developed. In the next stage, a novel solution approach by combining a heuristic algorithm and the fuzzy goal programming method is proposed to handle the multi-objective nature of the suggested model and its complexity. Then, several sensitivity analyses have conducted. The achieved outputs indicate the efficiency and effectiveness of the developed solution approach because it can obtain high-quality solutions in a reasonable time. The developed machine learning algorithm, with an accuracy of over 92%, has estimated the key parameters of the model. According to the obtained outputs, increasing the processing time has led to a significant increase in the total tardiness, GHG emissions, and energy consumption. Moreover, the results demonstrate that all objective functions ( i.e. , total tardiness, GHG emissions, and energy consumption) have increased by increasing the setup time parameter.