A scheduling decision-making framework using machine learning algorithm for energy efficient integrated factory

Sustainability enhancement is key factor in industrial decision-making. Maximizing use of energy-efficient factory to fulfil customer order deadlines is a significant problem for manufacturers. This paper focus on the study of integrated factory with four different route configurations. Two dis-patching rules (First come first serve and end due date) are considered for investigations. First, six different machine learning algorithms applied for performance characterization including, Mean Absolute Lateness (MAL), Root Mean Squared Lateness (RMSL), and Fill Rate (FR) to predict effective utilization of the framework. Then, to combine the energy utilization data of different configurations with predictive framework to focus decision strategy for sustain-able production. The robustness of the proposed strategy is examined by metrics of tests. Our results illustrate the application of machine learning approach-es in decision-making to execute efficient scheduling decisions and sustainable manufacturing.