Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Artificial Intelligence for Operations Research: Revolutionizing the Operations Research Process

Zhenan Fan, Bissan Ghaddar, Xinglu Wang, Linzi Xing, Yong Zhang, Zirui Zhou

TL;DR

This survey investigates how artificial intelligence can revolutionize the operations research process by enhancing parameter generation, model formulation, and model optimization. It surveys four AI techniques (Graph Neural Networks, Recurrent Neural Networks, Reinforcement Learning, and Imitation Learning) and three parameter-generation paradigms (predict-then-optimize, smart predict-then-optimize, and integrated prediction and optimization). It then comprehensively catalogs AI-driven advances across continuous and discrete optimization, including AAC, algorithm design for gradient-based methods, ADMM, column generation, simplex, branching, node selection, cutting planes, and heuristics, with emphasis on MIP/MINLP, LP/CP, and large-scale problems. The paper also discusses NLP-assisted modeling via LLMs for textbook and real-world problems, highlighting both capabilities and limitations, and concludes with directions for integrating AI more deeply into the entire OR pipeline and software ecosystems.

Abstract

The rapid advancement of artificial intelligence (AI) techniques has opened up new opportunities to revolutionize various fields, including operations research (OR). This survey paper explores the integration of AI within the OR process (AI4OR) to enhance its effectiveness and efficiency across multiple stages, such as parameter generation, model formulation, and model optimization. By providing a comprehensive overview of the state-of-the-art and examining the potential of AI to transform OR, this paper aims to inspire further research and innovation in the development of AI-enhanced OR methods and tools. The synergy between AI and OR is poised to drive significant advancements and novel solutions in a multitude of domains, ultimately leading to more effective and efficient decision-making.

Artificial Intelligence for Operations Research: Revolutionizing the Operations Research Process

TL;DR

This survey investigates how artificial intelligence can revolutionize the operations research process by enhancing parameter generation, model formulation, and model optimization. It surveys four AI techniques (Graph Neural Networks, Recurrent Neural Networks, Reinforcement Learning, and Imitation Learning) and three parameter-generation paradigms (predict-then-optimize, smart predict-then-optimize, and integrated prediction and optimization). It then comprehensively catalogs AI-driven advances across continuous and discrete optimization, including AAC, algorithm design for gradient-based methods, ADMM, column generation, simplex, branching, node selection, cutting planes, and heuristics, with emphasis on MIP/MINLP, LP/CP, and large-scale problems. The paper also discusses NLP-assisted modeling via LLMs for textbook and real-world problems, highlighting both capabilities and limitations, and concludes with directions for integrating AI more deeply into the entire OR pipeline and software ecosystems.

Abstract

The rapid advancement of artificial intelligence (AI) techniques has opened up new opportunities to revolutionize various fields, including operations research (OR). This survey paper explores the integration of AI within the OR process (AI4OR) to enhance its effectiveness and efficiency across multiple stages, such as parameter generation, model formulation, and model optimization. By providing a comprehensive overview of the state-of-the-art and examining the potential of AI to transform OR, this paper aims to inspire further research and innovation in the development of AI-enhanced OR methods and tools. The synergy between AI and OR is poised to drive significant advancements and novel solutions in a multitude of domains, ultimately leading to more effective and efficient decision-making.
Paper Structure (30 sections, 67 equations, 9 figures, 6 tables)

This paper contains 30 sections, 67 equations, 9 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Preliminary: AI techniques
  3. Graph Neural Networks
  4. Recurrent Neural Networks
  5. Reinforcement Learning
  6. Imitation learning
  7. Model Paramater Generation
  8. Predict, then optimize
  9. Smart predict, then optimize
  10. Integrated prediction and optimization
  11. Model Formulation
  12. Textbook modelling problems
  13. Real-world problems
  14. Automatic Algorithm Configuration
  15. Algorithm Selection and Design for Continuous Optimization
  16. ...and 15 more sections

Figures (9)

  • Figure 1: AI techniques in the OR process
  • Figure 2: Graph representation for linear program.
  • Figure 3: Graph representation for quadratic programs.
  • Figure 4: Illustration of how AI models generate parameters and interact with the optimization model.
  • Figure 5: Example on a textbook modeling problem.
  • ...and 4 more figures