873. Parallel Designs for Scatter Search
Invited abstract in session TB-15: Heuristic Search 2, stream Combinatorial Optimization.
Tuesday, 10:30-12:00Room: Esther Simpson 1.08
Authors (first author is the speaker)
| 1. | Sergio Pérez-Peló
|
| Informática y Estadística, Universidad Rey Juan Carlos | |
| 2. | Alejandra Casado
|
| Informática y Estadística, Universidad Rey Juan Carlos | |
| 3. | Jesus Sanchez-Oro
|
| Universidad Rey Juan Carlos | |
| 4. | Abraham Duarte
|
| Computer Sciences, Universidad Rey Juan Carlos | |
| 5. | Manuel Laguna
|
| Leeds School of Business, University of Colorado Boulder |
Abstract
Scatter Search (SS) is a well-established metaheuristic for tackling complex combinatorial optimization problems, known for its versatility and ease of adaptation. While several SS parallelization schemes have been proposed for specific problems, a general parallel framework remains unexplored. In this work, we introduce three parallel designs for SS, each addressing a distinct goal: reducing computational time, enhancing search exploration, and balancing intensification and diversification. Our experimental evaluation, based on benchmark problems where the state-of-the-art relies on sequential SS, provides insights into the impact of parallelization on solution speed and quality. Additionally, our publicly available code is designed for easy adaptation to new optimization problems. The results highlight promising directions for establishing a robust general framework for SS parallelization.
Keywords
- Metaheuristics
- Parallel Algorithms and Implementation
- Algorithms
Status: accepted
Back to the list of papers