2023 Fiscal Year Final Research Report
Development of high-performance scheduling methods by integrating domain-specific knowledge and generalized solutions methods
| Project/Area Number |
19K04105
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 18020:Manufacturing and production engineering-related
|
|---|
| Research Institution | Hiroshima University |
|---|
Principal Investigator |
Eguchi Toru 広島大学, 先進理工系科学研究科(工), 准教授 (80253566)
|
|---|
| Project Period (FY) |
2019-04-01 – 2024-03-31
|
| Keywords | スケジューリング / ジョブショップ / 遺伝的アルゴリズム / 優先規則 |
|---|
| Outline of Final Research Achievements |
This research developed a scheduling method to enhance the efficiency of systems engaged in high-mix, low-volume manufacturing. As the number of parts and processing machines increases, the complexity of production scheduling problems grows exponentially, making it impractical to find optimal solutions on a practical scale. In this study, neural networks are employed to automatically learn effective scheduling strategies for efficient optimization. By integrating this learned knowledge with an optimization method, we developed a high-performance scheduling approach.
|
| Free Research Field |
生産システム
|
| Academic Significance and Societal Importance of the Research Achievements |
製造業においては顧客ニーズの多様化により効率的な多品種少量生産が求められている．そのような生産を行う工場では生産工程は複雑になり，適切な生産順序指示が重要であるが，この問題はNP困難と言われ，最適化が非常に難しい問題である．本研究はこの問題に取り組み，効率的に高性能なスケジュールを作成する方法を提案した．特に，近年の労働者不足により，作業者スケジューリングも含めたより効率的な生産スケジューリングが求められており，その実現の社会的意義は大きい．
|
