Incorporate seagull optimization into ordinal optimization for solving the constrained binary simulation optimization problems

Constrained binary simulation optimization problems (CBSOP) are optimization problems with binary variables and stochastic objective function subject to given constraints. Solving the CBSOP by conventional optimization algorithms becomes highly time-consuming when the problem size is increased. Alth...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:The Journal of supercomputing Jg. 79; H. 5; S. 5730 - 5758
Hauptverfasser: Horng, Shih-Cheng, Lin, Shieh-Shing
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York Springer US 01.03.2023
Springer Nature B.V
Schlagworte:
Accuracy
Algorithms
Compilers
Computer Science
Constraints
Emulators
Heuristic methods
Interpreters
Linear programming
Manufacturing
Optimization
Optimization algorithms
Optimization techniques
Processor Architectures
Programming Languages
Real time
Simulation
Spline functions
Tensors
Ordinal optimization
Sorting conveyor system
Binary seagull optimization
Shortcuts layout
Optimal computing budget allocation
Tensor product B-splines
ISSN:0920-8542, 1573-0484
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Constrained binary simulation optimization problems (CBSOP) are optimization problems with binary variables and stochastic objective function subject to given constraints. Solving the CBSOP by conventional optimization algorithms becomes highly time-consuming when the problem size is increased. Although the ordinal optimization (OO) theory provides a reliable framework to solve CBSOP, the constraints still limit the efficiency and competitiveness of the OO theory. In this work, an approach incorporating binary seagull optimization into ordinal optimization, abbreviated as BSOO, is developed for solving the CBSOP in a reasonable time. The BSOO comprises three essential components: emulator, exploration, and exploitation. First of all, the regularized minimal-energy tensor product B-splines are regarded as an emulator to estimate the performance of a solution. Next, the binary seagull optimization algorithm is utilized to determine N exceptional solutions from the decision space. Finally, the reformed optimal computing budget allocation is employed to find an illustrious solution from the N exceptional solutions. To verify the proposed method, the BSOO is applied for finding the optimal layout of shortcuts for maximizing the capacity of the sorting conveyor system in a reasonable time. Experimental results of the BSOO are compared to five heuristic methods. The BSOO outperforms the five heuristic methods even after the latter took more than 30 times the CPU time that was consumed by BSOO upon completion. Test results reveal that the BSOO can be adopted in a real-time application of the sortation system.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0920-8542
1573-0484
DOI:10.1007/s11227-022-04880-y