Integrated adaptive forecasting and energy aware scheduling model using genetic algorithm with dynamic stopping condition

Abstract

According to Annual Energy Outlook 2022, published by the US Energy Information and Administration, manufacturing energy intensity is predicted to be reduced from 2018 to 2050. Despite near term uncertainty due to COVID-19 outbreak, this change can be attributed to the paradigm shift towards environmental sustainability and energy efficiency resulting in efforts by manufacturers to reduce costs, achieve carbon neutral status, and adhere to local, state, and international regulations. Manufacturers implement various demand side management projects for this purpose, one of them being energy efficient scheduling. The job shop scheduling model takes into consideration energy and demand costs, earliness and tardiness costs, worker costs, machine depreciation cost, and resource leveling for worker. Artificial Neural Network model is used to provide highly accurate prediction of energy prices and genetic algorithm is used to obtain job sequence and assignment in reasonable time for the job shop scheduling problem.