[2020] Genetic Programming with Adaptive Search Based on the Frequency of Features for Dynamic Flexible Job Shop Scheduling.pdf (748.71 kB)

Genetic Programming with Adaptive Search Based on the Frequency of Features for Dynamic Flexible Job Shop Scheduling

Download (748.71 kB)
conference contribution
posted on 29.10.2020, 01:02 by Fangfang Zhang, Yi Mei, S Nguyen, Mengjie Zhang
© 2020, Springer Nature Switzerland AG. Dynamic flexible job shop scheduling (DFJSS) is a very valuable practical application problem that can be applied in many fields such as cloud computing and manufacturing. In DFJSS, machine assignment and operation sequencing decisions need to be made simultaneously in dynamic environments with unpredicted events such as new job arrivals. Scheduling heuristic is an ideal candidate for solving the DFJSS problem due to its efficiency and simplicity. Genetic programming (GP) has been successfully applied to evolve scheduling heuristics for job shop scheduling automatically. However, GP has a huge search space, and the traditional search algorithms do not utilise effectively the information obtained from the evolutionary process. This paper proposes a new method to make better use of the information during the evolutionary process of GP to further enhance the ability of GP. To be specific, this paper proposes two adaptive search strategies based on the frequency of features in promising individuals to guide GP to evolve effective rules. This paper examines the proposed algorithm on six different DFJSS scenarios. The results show that the proposed GP with adaptive search can converge faster and achieve significantly better performance than the GP without adaptive search in most scenarios while no worse in all other scenarios without increasing the computational cost.

History

Preferred citation

Zhang, F., Mei, Y., Nguyen, S. & Zhang, M. (2020, January). Genetic Programming with Adaptive Search Based on the Frequency of Features for Dynamic Flexible Job Shop Scheduling. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (12102 LNCS pp. 214-230). Springer International Publishing. https://doi.org/10.1007/978-3-030-43680-3_14

Title of proceedings

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Volume

12102 LNCS

Publication or Presentation Year

01/01/2020

Pagination

214-230

Publisher

Springer International Publishing

Publication status

Published

ISSN

0302-9743

eISSN

1611-3349

Exports

Logo branding

Categories

Exports