LLM for Large-Scale Optimization Model Auto-Formulation: A Lightweight Few-Shot Learning Approach

Abstract: Large-scale optimization is a key backbone in modern business decision-making. However, the process of building these models is often labor-intensive and time-consuming. We address this by proposing LEAN-LLM-OPT, a LightwEight few-shot leArNing framework for LLM-assisted large-scale OPTimization auto-formulation, which takes a query (a problem description and associated datasets) as input and orchestrates a team of LLM agents to output the optimization formulation. LEAN-LLM-OPT innovatively applies few-shot learning to teach LLM agents how they could effectively apply reasoning and customized tools to build optimization models. Specifically, upon receiving a query, a problem classification agent first determines the type of the problem. Then, a few-shot example generation agent consolidates a set of examples that demonstrate how optimization models are built for problems of the same type step-by-step. Finally, a model classification agent follows these examples to extract relevant information from the input datasets and generate the final output (together with the executable programming code). Extensive simulations validate that LEAN-LLM-OPT attains state-of-the-art accuracy compared to existing methods, especially on large-scale optimization problems. Additionally, in a Singapore Airlines choice-based revenue management use case, LEAN-LLM-OPT demonstrates its value by achieving leading performance across a variety of scenarios. Along the way, we introduce Large-Scale-OR and Air-NRM, the first comprehensive benchmarks for large-scale optimization problems.

Bio: Hanzhang Qin is an Assistant Professor at the Department of Industrial Systems Engineering and Management at NUS. He is also an affiliated faculty member at the NUS Institute for Operations Research and Analytics and the NUS AI Institute. His research was recognized by several awards, including INFORMS TSL Intelligent Transportation Systems Best Paper Award and MIT MathWorks Prize for Outstanding CSE Doctoral Research. Before joining NUS, Hanzhang spent one year as a postdoctoral scientist in the Supply Chain Optimization Technologies Group of Amazon NYC. He earned his PhD in Computational Science and Engineering under supervision of Professor David Simchi-Levi, and his research interests span stochastic control, applied probability and statistical learning, with applications in supply chain analytics and transportation systems. He holds two master's, one in EECS and one in Transportation both from MIT. Prior to attending MIT, Hanzhang received two bachelor degrees in Industrial Engineering and Mathematics from Tsinghua University.