Energy-Aware Flowshop Scheduling: A Case for AI-Driven Sustainable Manufacturing

Abstract

A fully verifiable and deployable framework for optimizing schedules in a batch-based production system is proposed. The scheduler is designed to control and optimize the flow of batches of material into a network of identical and non-identical parallel and series machines that produce a high variation of complex hard metal products. The proposed multi-objective batch-based flowshop scheduling optimization (MOBS-NET) deploys a fully connected deep neural network (FCDNN) with respect to three performance criteria of energy, cost and makespan. The problem is NP-hard and considers minimizing the energy consumed per unit of product, operations cost, and the makespan. The output of the method has been validated and verified as optimal operational planning and scheduling meeting the business operational objectives. Real-time and look ahead discrete event simulation of the production process provides the feedback and assurance of the robustness and practicality of the optimum schedules prior to implementation.

Green Hybrid Flow Shop Scheduling Problem Considering Sequence Dependent Setup Times and Transportation Times

“…Constraint (14) indicates that if two jobs are processed on the same machine and there is an immediate relationship, then the start time of the later job is not less than the sum of the completion time of the earlier one and the setup time between them. Constraint (15) indicates that for any jobs, the next operation does not begin to process until the current operation is completed and the job is transported to the next stage. Constraint (16) means that the transportation to the next stage starts immediately after the operation is completed.…”

“…For example, case J10s5 represents that 10 jobs are processed at 5 stages. The data in all cases are generated randomly, where the job size is [10,15,20,25,30,35,40], the number of stages is between [2][3][4][5][6], the processing time of each operation is between [3][4][5][6][7][8][9][10][11][12][13][14][15], the number of machines at each stage is between [1-3], the sequence dependent setup time is between [1][2][3][4][5], the transportation time is between [2][3][4][5][6][7][8], the energy consumption per unit in machining state is 1, the energy consumption per unit in setup state is 1.2, and the energy consumption per unit in transportation state is 1.5. For the convenience of calculation, the processing time, sequence dependent setup time and transportation time are taken as integers, and the minimum value is 1.…”

“…The authors in [14] developed an integrated model for machining parameter optimization and flow shop scheduling, and designed a teaching optimization algorithm to minimize the makespan and carbon emission. Danishvar et al [15] proposed deep neural network algorithm to slove the flow shop problem aiming at minimizing the energy, cost and makespan. Zheng [16] proposed a multi-objective fruit fly optimization algorithm to solve the resource-constrained unrelated parallel machine scheduling problem with the objective of minimizing the makespan and total carbon emissions.…”