Forecasting Supply Chain Disruptions with Foresight Learning
This addresses the challenge of anticipating infrequent, high-impact events for firms and policymakers, offering a domain-specific forecasting solution.
The paper tackles the problem of forecasting supply chain disruptions by training LLMs to produce calibrated probabilistic forecasts, resulting in a model that substantially outperforms strong baselines like GPT-5 on accuracy, calibration, and precision.
Anticipating supply chain disruptions before they materialize is a core challenge for firms and policymakers alike. A key difficulty is learning to reason reliably about infrequent, high-impact events from noisy and unstructured inputs - a setting where general-purpose models struggle without task-specific adaptation. We introduce an end-to-end framework that trains LLMs to produce calibrated probabilistic forecasts using realized disruption outcomes as supervision. The resulting model substantially outperforms strong baselines - including GPT-5 - on accuracy, calibration, and precision. We also show that training induces more structured and reliable probabilistic reasoning without explicit prompting. These results suggest a general pathway for training domain-specific forecasting models that produce decision-ready signals. To support transparency we open-source the evaluation dataset used in this study. Dataset: https://huggingface.co/datasets/LightningRodLabs/supply-chain-predictions