Machine learning-based intermittent equipment scheduling model for flexible production process

Abstract With the development of intelligent manufacturing, production scheduling is widely applied in industry to enhance production efficiency and reduce costs. Therefore, this study proposes an optimal scheduling model based on an intelligent algorithm that can scientifically and reasonably determine processing times for intermittent production equipment and overcome the shortcomings of the manpower scheduling used in the past, thus reducing companies’ electricity costs. Considering the tissue pulping-shop scheduling problem, a multi-objective optimization model is established for production scheduling. We apply the nondominated sorting genetic algorithm II to solve the optimal scheduling problem, and the experimental results reveal that the model can maintain a higher availability and reliability as compared to actual production. Consequently, the dispatch generated by this model can reduce the weighted electricity price by 0.04 yuan/kWh.