This paper systematically reviews research trends in applying Large-Scale Language Models (LLMs) to operations research (OR) to support complex system decision-making in diverse fields such as transportation, supply chain management, and production planning. LLMs demonstrate the potential to transform natural language problem descriptions into mathematical models or executable code, generate heuristics, develop algorithms, and directly solve optimization problems. This paper categorizes three paths for applying LLMs to OR (automatic modeling, assisted optimization, and direct solution), reviews evaluation benchmarks and domain-specific application cases, highlights key challenges, and suggests future research directions.
