Bridging the Sim-to-Real Gap in Reinforcement Learning-Based Industrial Dispatching through Execution Semantics

Jonathan Hoss, Noah Klarmann

arXiv:2605.29078 · 2026-05-29 공개 · arXiv · PDF

reinforcement-learning observability sim-to-real-gap industrial-dispatching execution-semantics event-driven-scheduling discrete-event-simulation policy-neutral-layer

Abstract

Event-driven scheduling policies are increasingly deployed in industrial environments, where decisions are made under asynchronous and partially observed system states. As a result, decision states are not temporally consistent, action admissibility is not explicitly defined, and the origin of execution errors remains ambiguous. These issues limit both reliability and interpretability. To address this gap, a policy-neutral execution and measurement layer is proposed to mediate between scheduling policies and the industrial execution environment. The layer constructs decision-valid snapshots from asynchronous event streams, defines a standardized execution contract with explicit action admissibility, and records outcomes as divergences between policy intent, transactional outcomes, physical execution, and human intervention. This enables a separation between decision semantics and execution behavior and makes deployment mismatch observable and structurally attributable. The proposed framework is evaluated using a discrete-event simulation. The results show analytical benefits across all observation lag regimes, as undifferentiated execution failures are transformed into structured, typed outcomes with full attribution coverage. Operational benefits are strongest under low observation lag, where avoidable execution errors can be prevented before commitment. Overall, the layer turns execution uncertainty into supervisory data for evaluation and policy refinement.

한국어 요약

📋 한 줄 요약

**[산업 스케줄링 RL / Sim-to-Real]** Policy-neutral 실행·측정 layer가 비동기 event stream을 decision-valid snapshot으로 구성, 표준화된 실행 contract·action admissibility 정의·divergence(intent·transactional·physical·human) 기록으로 deployment mismatch를 구조적 attribution으로 변환.

🎯 핵심 기여도

💡 핵심 아이디어

RL 산업 스케줄링의 sim-to-real 격차는 정책 알고리즘 자체보다 비동기·partial 관측 환경에서의 실행 의미 부재가 더 핵심 원인이며, 결정 의미와 실행 행동을 분리하는 policy-neutral layer로 unstructured 실패를 structured supervisory data로 변환해야 한다.

🔬 기술적 접근법

📊 주요 결과

💭 의의 및 한계

**의의**: Sim-to-real 격차의 원인을 알고리즘 외부의 실행 의미·관측 일관성에서 찾는 관점 전환, policy-neutral 설계로 algorithm-agnostic 적용, 구조적 attribution으로 디버깅·정책 개선 가능. **한계**: Discrete-event simulation 평가로 실 산업 검증은 후속, 표준화된 contract 정의의 도메인별 맞춤 부담, 비동기성·partial 관측 모델링의 충실성 의존.

🚀 실용적 활용