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

TL;DR

The paper addresses the need for transparent explainability in Operations Research by introducing Explainable Operations Research (EOR), a framework that combines what-if analysis with the novel concept of Decision Information to quantify and explain constraint- and parameter-driven changes. It leverages bipartite graphs and LLMs to generate two types of explanations—attribution and, more importantly, justification—within an end-to-end workflow that uses Commander, Writer, and Safeguard agents. An industrial benchmark (IndustryOR) with 30 problems and 10 queries each is established to rigorously evaluate both modeling accuracy and explanation quality, with comparisons to Standard and OptiGuide baselines across multiple GPT-4 variants. The results show that EOR improves both the correctness of updated models and the depth of explanations, demonstrating practical potential for transparency and trust in OR applications. The work lays a foundation for future enhancements in automated evaluation of explanations and more precise modeling techniques that integrate user-driven what-if queries into explainable optimization pipelines.

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.

Figures (5)

• Figure 1: The framework of EOR.
• Figure 2: The overall workflow of EOR.
• Figure 3: An example illustrating the code and explanations generated by EOR.
• Figure 4: An example illustrating codes generated by different models.
• Figure 5: An example illustrating explanations generated by different models.

Theorems & Definitions (1)

• Definition 1