Development and Applied Research of Evolutionary Computation Methods for Dynamic Systems Changing Over Time

2023 Fiscal Year Final Research Report

• Project/Area Number: 20K11972
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 61040:Soft computing-related
• Research Institution: Hiroshima University
• Principal Investigator: Hayashida Tomohiro 広島大学, 先進理工系科学研究科(工), 准教授 (20432685)
• Co-Investigator(Kenkyū-buntansha): 広谷 大助 県立広島大学, 地域創生学部, 准教授 (30432686)
• Project Period (FY): 2020-04-01 – 2024-03-31
• Keywords: 動的システム / 進化計算 / 長期メモリ

Outline of Final Research Achievements

In 2020, we began developing parallel distributed optimization methods based on GA (Genetic Algorithms) and PSO (Particle Swarm Optimization), and optimization methods using long-term memory for dynamic systems. We also formulated and conducted numerical experiments on the dynamic production planning problem. In 2021, we developed parallel distributed genetic programming using long-term memory and began extracting optimization rules, confirming their utility through numerical experiments. The research was extended in 2022 due to the impact of COVID-19, and in 2023, we focused on presenting our findings. Our work has been presented at multiple domestic and international conferences, and one academic paper is currently under review. Notably, the proposed method's effectiveness in dynamic scheduling problems was confirmed, demonstrating practical applicability.

Free Research Field

最適化

Academic Significance and Societal Importance of the Research Achievements

本研究では、遺伝的アルゴリズムおよび粒子群最適化手法を基礎とする並列分散最適化手法と、動的システムのための長期メモリを用いた最適化手法を開発した。これにより、動的生産計画問題の定式化と数値実験を通じて、動的環境における解探索の効率を大幅に向上させた。長期メモリを用いた並列分散型遺伝的プログラミングの開発と、最適化ルール抽出手法の確立を行い、その有用性を数値実験で確認した。特に、動的スケジューリング問題において提案手法の有用性が確認され、実践的な適用可能性が示されたことは、本研究の学術的意義を強く示している。