OptiMUS: Optimization Modeling Using MIP Solvers and large language models
This addresses the limited adoption of optimization tools in sectors like manufacturing and healthcare due to expertise barriers, representing a novel method for automating problem-solving.
The paper tackles the problem of formulating and solving mixed integer linear programming (MILP) problems from natural language descriptions by introducing OptiMUS, an LLM-based agent that develops models, writes code, and validates solutions, and it solves nearly twice as many problems as a basic LLM prompting strategy.
Optimization problems are pervasive across various sectors, from manufacturing and distribution to healthcare. However, most such problems are still solved heuristically by hand rather than optimally by state-of-the-art solvers, as the expertise required to formulate and solve these problems limits the widespread adoption of optimization tools and techniques. We introduce OptiMUS, a Large Language Model (LLM)-based agent designed to formulate and solve MILP problems from their natural language descriptions. OptiMUS is capable of developing mathematical models, writing and debugging solver code, developing tests, and checking the validity of generated solutions. To benchmark our agent, we present NLP4LP, a novel dataset of linear programming (LP) and mixed integer linear programming (MILP) problems. Our experiments demonstrate that OptiMUS solves nearly twice as many problems as a basic LLM prompting strategy. OptiMUS code and NLP4LP dataset are available at \href{https://github.com/teshnizi/OptiMUS}{https://github.com/teshnizi/OptiMUS}