452 test examples for evaluating models on calibrated probabilistic forecasting of infrequent, high-impact supply chain disruptions. The dataset was created by LightningRodLabs for the paper 'Forecasting Supply Chain Disruptions with Foresight Learning' and last updated on 2026-04 03.
Use Cases
- Benchmarking probabilistic forecasting models based on noisy and unstructured inputs.
- Evaluating model calibration for infrequent, high-impact events.
- Researching foresight learning methods for supply chain disruption prediction.
Strengths
- 452 examples in the test split provide a basis for model evaluation.
- Designed specifically for evaluating calibrated probabilistic forecasts.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count for the full dataset is unknown, which may limit suitability assessment.
- Description metadata is limited; actual data quality requires manual inspection after download.
Provenance
- Source
- LightningRodLabs
- Collection Method
- Created as an evaluation dataset for the associated research paper.
- Time Range
- null
- Freshness
- Last updated 2026-04-03 13:56:59; freshness should be verified.
- Geography
- null