Minimizing total energy cost and tardiness penalty for a scheduling-layout problem in a flexible job shop system: A comparison of four metaheuristic algorithms
- Louisiana State University, Baton Rouge, LA (United States); Louisiana State University
- Louisiana State University, Baton Rouge, LA (United States)
- University of Miami, Coral Gables, FL (United States)
Job scheduling and machine layout are interrelated in improving energy consumption (EC) and productivity measures such as tardiness and represent two important decisions that must be made by manufacturers. This interdependency can be explained by transportation time, which connects scheduling and layout. Scheduling and layout, however, have not been thoroughly studied in conjunction using an integrated model in the context of sustainable manufacturing. Hence, we propose an energy-aware optimization model in which scheduling is integrated with layout in a single-level framework. More specifically, a single objective function is defined to minimize the facility energy cost and the job tardiness penalty, which control EC and tardiness respectively in a flexible job shop system. In order to model machine EC more accurately, we also consider three different machine states: a processing state and two idle states. Our case studies show that the integrated model exhibits better performance in controlling manufacturing EC and job tardiness than a non-integrated model in which machine locations are uncontrollable and transportation times between machines are unchangeable. To deal with large-sized problems, we also introduce four new metaheuristics. The performances of these new metaheuristics are compared in terms of objective function values and CPU times using various case studies. The results indicate that a hybrid ant colony optimization and simulated annealing (ACO-SA) algorithm provides better performance than the other algorithms. Specifically, our case studies show that the integrated model using ACO-SA can improve the objective function value by around 5% when compared to the non-integrated model.
- Research Organization:
- Louisiana State University, Baton Rouge, LA (United States)
- Sponsoring Organization:
- USDOE Office of Energy Efficiency and Renewable Energy (EERE)
- Grant/Contract Number:
- EE0007709
- OSTI ID:
- 1994459
- Journal Information:
- Computers and Industrial Engineering, Journal Name: Computers and Industrial Engineering Vol. 141; ISSN 0360-8352
- Publisher:
- ElsevierCopyright Statement
- Country of Publication:
- United States
- Language:
- English
Solving the Job-Shop Scheduling Problem with Reinforcement Learning
|
other | January 2020 |
Similar Records
A dynamic control approach for energy-efficient production scheduling on a single machine under time-varying electricity pricing
Exact and heuristic algorithms for minimizing Tardy/Lost penalties on a single-machine scheduling problem