OMGPT: A Sequence Modeling Framework for Data-driven Operational Decision Making

2025-05-19

AI Summary

This work addresses sequential decision-making problems in operations research—including dynamic pricing, inventory management, resource allocation, and queue control—by introducing OMGPT, the first end-to-end generative pre-trained Transformer tailored for operations decisions. Unlike conventional approaches relying on analytical models or task-specific heuristics, OMGPT frames the mapping from historical system states to optimal actions as a unified sequence-to-sequence prediction task, enabling model-free, purely data-driven generalization. Methodologically, it integrates the Transformer architecture with Bayesian theoretical analysis and large-scale pretraining with transfer learning. Extensive experiments demonstrate that OMGPT significantly outperforms existing baselines across diverse operations tasks, exhibiting strong cross-task generalization and robustness to environmental variations. It establishes a scalable, transferable paradigm for operations decision-making, bridging deep learning and classical OR without requiring domain-specific modeling assumptions.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Planning, Routing, and Scheduling: Planning with Markov Models (MDPs, POMDPs)
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising

Abstract

We build a Generative Pre-trained Transformer (GPT) model from scratch to solve sequential decision making tasks arising in contexts of operations research and management science which we call OMGPT. We first propose a general sequence modeling framework to cover several operational decision making tasks as special cases, such as dynamic pricing, inventory management, resource allocation, and queueing control. Under the framework, all these tasks can be viewed as a sequential prediction problem where the goal is to predict the optimal future action given all the historical information. Then we train a transformer-based neural network model (OMGPT) as a natural and powerful architecture for sequential modeling. This marks a paradigm shift compared to the existing methods for these OR/OM tasks in that (i) the OMGPT model can take advantage of the huge amount of pre-trained data; (ii) when tackling these problems, OMGPT does not assume any analytical model structure and enables a direct and rich mapping from the history to the future actions. Either of these two aspects, to the best of our knowledge, is not achieved by any existing method. We establish a Bayesian perspective to theoretically understand the working mechanism of the OMGPT on these tasks, which relates its performance with the pre-training task diversity and the divergence between the testing task and pre-training tasks. Numerically, we observe a surprising performance of the proposed model across all the above tasks.

Problem

• Develops OMGPT for sequential operational decision-making tasks
• Proposes a unified framework for dynamic pricing, inventory, and resource tasks
• Uses transformer models without assuming analytical structures

Innovation

• Develops a GPT model for operational decision-making tasks
• Uses transformer-based neural network for sequential modeling
• Leverages pre-trained data without assuming analytical model structure