Home icon

How Deutsche Bahn redefines forecasting using Chronos models – Now available on Amazon Bedrock Marketplace

Machine Learning Blog



Deutsche Bahn has revolutionized its forecasting process by adopting Chronos-Bolt, a time series foundation model available on Amazon Bedrock Marketplace. The solution addresses key challenges in enterprise-level time series forecasting.

  • Previously, Deutsche Bahn had heterogeneous forecasting processes requiring significant manual effort
  • Chronos-Bolt enables zero-shot forecasting with up to 250 times faster inference
  • The model treats time series data like a language, similar to how LLMs generate text
  • In test cases, Chronos models achieved superior accuracy compared to traditional statistical methods
  • Deutsche Bahn is developing a company-wide forecasting service that reduces forecast preparation time from weeks to hours

The solution demonstrates how foundation models can democratize forecasting across organizations by making sophisticated predictions more accessible and efficient.



Go to article

The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.

Related articles

May 5
2025
Amazon Bedrock Data Automation now supports extraction of custom insights from audio
May 13
2025
Improving weather forecasting accuracy using Amazon AppStream 2.0 graphic-intensive instances
May 14
2025
Amazon Bedrock Guardrails now supports cross-region inference
May 30
2025
Bridging the gap between development and production: Seamless model lifecycle management with Amazon Bedrock

The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.