인문학
사회과학
자연과학
공학
의약학
농수해양학
예술체육학
복합학
DBpia AI
AI 에이전트
AI 아이디어
AI 리더
AI 뷰어
Citeasy
크롬 서지관리 다운로드
지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학술저널
- 저자정보
- 발행연도
- 2026.6
- 수록면
- 107 - 122 (16page)
- DOI
- 10.7315/CDE.2026.107
이용수
초록· 키워드
While the application of reinforcement learning in manufacturing process optimization is expanding, its large-scale iterative training requires simulation environments that accurately reflect complex process structures and resource constraints. However, current simulation models must be manually constructed for each process and reconfigured whenever the process changes, which limits the scalability of simulation-based optimization. To address this, this study proposes an ontology-based framework that automatically generates a discrete-event simulation. The framework takes manufacturing process information as input through a unified JSON schema, maps it as instances onto an OWL ontology, and then infers process connections, assembly points, and other relationships implicit in the input through SWRL-based semantic reasoning. The inferred knowledge is used to automatically generate SimPy-based simulation code, reducing repetitive manual modeling. The accuracy of the auto-generated models was validated through comparative analysis with the commercial tool Visual Components. Furthermore, the auto-generated simulation was integrated into a reinforcement learning environment and applied to manufacturing operation optimization. Productivity improvements over heuristic policies demonstrated the practical effectiveness of the proposed framework.
상세정보 수정요청해당 페이지 내 제목·저자·목차·페이지정보가 잘못된 경우 알려주세요!
목차
- ABSTRACT
- 1. 서론
- 2. 관련 연구
- 3. 온톨로지 기반 시뮬레이션 자동 생성 및 강화학습 통합 환경 프레임워크
- 4. 시나리오 기반 사례 연구
- 5. 결론
- References
참고문헌
참고문헌 신청최근 본 자료
