Research
Communication | Open Access
Volume 2022 | Communication ID 537
Volume 2022 | Communication ID 537
Solving Job-Shop Scheduling Problem by recurrent neural networks and a new hybrid matrix metaheuristic: a comparative study |
Lotfi Nohair, Abderrahim Eladraoui, Abdelwahed Namir
Received |
Accepted |
Published |
May 14, 21:31 |
July 01, 2022 |
July 15, 2022 |
Abstract: Based on the research of Zhang [1] and Willems [2], this communication proposes a recurrent neural network to solve Job-shop scheduling problems [3]. Firstly, the problem was translated in an integer linear programming model which the objective is to minimize the makespan, subject to three types of constraints: 1) Starting time constraints (ST units): the starting time of each operation must be a positive integer number. 2) Sequence constraints (SC units): An operation can only be scheduled after the preceding ones have ended. 3) Resource constraints (RC units): Machine can process ...