ORPilot: A Production-Oriented Agentic LLM-for-OR Tool for Optimization Modeling

2026-05-04

AI Summary

This work proposes a multi-agent optimization modeling system tailored for industrial settings to address challenges such as ambiguous problem definitions, messy real-world data, and poor solver compatibility. The system leverages conversational requirement clarification and coordinated agents for independent data ingestion and parameter computation to automatically generate a solver-agnostic intermediate representation (IR). It further incorporates a self-repair mechanism driven by solver feedback. Innovatively combining large language models with an IR recompilation technique that requires zero LLM invocations during execution, the framework supports mainstream solvers including Gurobi, CPLEX, PuLP, Pyomo, and OR-Tools. Evaluated on the IndustryOR benchmark, it outperforms existing state-of-the-art methods in accuracy, while achieving competitive performance on NL4OPT and NLP4LP, demonstrating its effectiveness and generalization capability in real-world business optimization scenarios.

Abstract

This paper presents ORPilot, an open-source agentic AI system that translates real-world business problems into solver-ready optimization models. Unlike academic LLM-for-OR tools that assume clean problem specifications with preformatted inline data, ORPilot is designed for production conditions: ambiguous descriptions, large-scale raw operational data, and the need for portability across solver backends. The system introduces four novel components: (1) a conversational interview agent to elicit complete problem specifications, (2) a data collection agent that retrieves data independently of prompts, (3) a parameter computation agent to bridge raw tabular data and model-ready parameters, and (4) a solver-agnostic Intermediate Representation (IR) for deterministic, zero-LLM-call recompilation to Gurobi, CPLEX, PuLP, Pyomo, or OR-Tools solvers. Additionally, self-correcting retry loops utilize solver tracebacks for targeted repairs. ORPilot represents the first attempt to target production-level business problems rather than textbook operations research (OR) cases. Evaluation on real-world problems demonstrates promising results. When tested against traditional academic benchmarks: IndustryOR, NL4OPT and NLP4LP, ORPilot outperformed state-of-the-art tools in accuracy on the IndustryOR benchmark and delivered comparable performance on NL4OPT and NLP4LP.

Problem

• optimization modeling
• production-oriented
• real-world business problems
• solver-agnostic
• large-scale raw data

Innovation

• Agentic LLM
• Optimization Modeling
• Solver-Agnostic IR
• Production-Oriented OR
• Self-Correcting Retry