← Back
AWS ML Blog

How Decathlon runs demand forecasting at scale with Chronos-2

•12 min read•
#inference
Level:Advanced
For:ML Engineers, Forecasting Engineers
✦TL;DR

Decathlon deployed Meta’s Chronos‑2 model on AWS to forecast weekly demand for tens of thousands of SKUs across multiple continents, boosting forecast accuracy by 11–15 percentage points. The rollout leveraged AWS’s scalable inference infrastructure to run the model on a weekly cadence, dramatically reducing operational complexity compared to their legacy pipeline. The integration enabled real‑time inventory planning and supply‑chain optimization, though it required migrating the entire forecasting stack to AWS.

⚡ Key Takeaways

  • Forecast accuracy improved by 11–15 points using Chronos‑2.
  • Deployment used AWS’s scalable inference services (e.g., SageMaker or ECS) to run weekly predictions.
  • Operational complexity was cut by automating the entire pipeline, eliminating manual model‑tuning steps.
  • Forecasts are generated on a weekly schedule, feeding directly into inventory and replenishment systems.
  • The solution depends on AWS infrastructure; migrating to another cloud would require re‑architecting the inference pipeline.
  • WhyItMatters: Engineers building production forecasting systems can adopt Chronos‑2 on AWS to achieve significant accuracy gains while simplifying operations, enabling more responsive supply‑chain decisions.
  • TechnicalLevel: Advanced
  • TargetAudience: ML Engineers, Forecasting Engineers
  • PracticalSteps:
💡 Why It Matters

Engineers building production forecasting systems can adopt Chronos‑2 on AWS to achieve significant accuracy gains while simplifying operations, enabling more responsive supply‑chain decisions.

Want the full story? Read the original article.

Read on AWS ML Blog ↗

More like this

The Pulse: What can we learn from the Railway hack?

Pragmatic Engineer•#deployment

Model Compression Beyond Quantization

Machine Learning Mastery•#inference

Introducing Amazon SageMaker HyperPod Global Inference Gateway: One endpoint across Regions and clusters

AWS ML Blog•#deployment

Helping AI text detectors keep up with changing conditions

MIT News AI•#llm

EXPLORE AI NEWS

Daily hand-picked stories on LLMs, RAG, agents and production AI — curated for engineers who ship.

BROWSE NEWS

GET THE WEEKLY DIGEST

Join engineers getting the Monday signal-over-noise AI breakdown. No spam, unsubscribe anytime.

LEARN AI ENGINEERING

Curated courses, research papers, repos and tutorials built for engineers leveling up in AI.

START LEARNING