IISE Transactions

Advances in Reinforcement Learning and Large Language Models for Intelligent Systems

In today’s rapidly evolving industrial and technological landscape, ensuring high quality and reliability in complex, intelligent systems is becoming increasingly challenging. Traditional methods for quality prediction, control, reliability assessment, and maintenance optimization often fall short in complex and uncertain environments, particularly when the underlying physical models are unknown. Reinforcement Learning (RL), a key area of artificial intelligence, offers a powerful framework for tackling uncertainty and complexity in such environments. More recently, the emergence of Large Language Models (LLMs) and foundation models has introduced new capability in reasoning, generalization, and multi‐modal decision support—further expanding the boundaries of automation, prediction, control, fault detection, planning, and system design. RL and LLMs hold enormous potential for enhancing resilience and intelligence across domains, including manufacturing, transportation, healthcare, and logistics.
This special issue aims to promote and showcase cutting‐edge research on RL‐ or LLM‐based methods for improving the quality and reliability of complex intelligent systems. We welcome submissions that bridge theoretical development and practical implementation. Topics of interest include, but are not limited to:

1. Core Methods and Algorithms

• Learning Foundations of RL
  • Safe, interpretable and robust RL for mission‐critical and safety‐sensitive applications
  • Causal RL to improve quality and reliability in complex systems
  • Physics‐informed RL for intelligent systems
  • Multi‐agent RL with various coordination mechanisms
  • Generalist agents for dynamic and diverse environments
  • Sim‐to‐real transfer for real‐world deployment of RL
• LLM‐Centric Modeling, Reasoning and Generation Capabilities
  • Enhancing LLMs’ relevance and accuracy for quality and reliability applications
  • Knowledge representation and reasoning using LLM for intelligent systems
  • LLM‐empowered decision‐making in complex systems
  • Combining LLM with RL for intelligent systems

2. Applications in Complex Intelligent Systems

• Quality Control and Maintenance Planning
  • RL or LLM for process monitoring, quality control, fault detection, and root cause analysis
  • RL or LLM for predictive maintenance and failure prognosis
• System Optimization and Virtualization
  • RL or LLM‐based resource allocation, scheduling, and control for system efficiency and resiliency
  • RL or LLM for digital twins
• Cybersecurity
  • RL or LLM for intrusion detection and risk assessment in cyber‐physical systems
  • RL and LLM for attack‐resilient systems by design
• Personalized Production and Services
  • RL or LLM for high‐quality personalized product design and manufacturing
  • RL or LLM for personalized recommendation and decision‐making in healthcare

This Special Issues is sponsored by Focus Issue of Data Science, Quality, and Reliability.