Decision Information Meets Large Language Models: The Future of Explainable Operations Research

2025-02-14

AI Summary

This work addresses the critical lack of interpretability—and consequent deficits in transparency and trustworthiness—arising from integrating large language models (LLMs) into operations research (OR). To this end, we propose Explainable Operations Research (EOR), a novel framework grounded in rigorous formalism. Methodologically, we introduce the first theory of “decision information,” quantifying how constraints and parameter perturbations affect optimal solutions; design a bipartite graph modeling mechanism that tightly couples LLM-driven what-if analysis with OR decision structures to enhance operationality and comprehensibility; and establish the first industrial-scale OR interpretability benchmark, defining new standards for transparent, verifiable explanations. Empirical evaluation demonstrates substantial improvements in both explanation accuracy and practical utility. Our framework thus provides both a theoretical foundation and an engineering paradigm for robust OR–AI integration.

Technology Category

• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Planning, Routing, and Scheduling: Model-Based Reasoning
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Operations Research (OR) is vital for decision-making in many industries. While recent OR methods have seen significant improvements in automation and efficiency through integrating Large Language Models (LLMs), they still struggle to produce meaningful explanations. This lack of clarity raises concerns about transparency and trustworthiness in OR applications. To address these challenges, we propose a comprehensive framework, Explainable Operations Research (EOR), emphasizing actionable and understandable explanations accompanying optimization. The core of EOR is the concept of Decision Information, which emerges from what-if analysis and focuses on evaluating the impact of complex constraints (or parameters) changes on decision-making. Specifically, we utilize bipartite graphs to quantify the changes in the OR model and adopt LLMs to improve the explanation capabilities. Additionally, we introduce the first industrial benchmark to rigorously evaluate the effectiveness of explanations and analyses in OR, establishing a new standard for transparency and clarity in the field.

Problem

• Enhancing transparency in Operations Research using LLMs
• Developing explainable optimization with Decision Information
• Establishing benchmarks for explainable OR effectiveness

Innovation

• Integrates Large Language Models
• Uses bipartite graphs
• Introduces industrial benchmark