Decathlon switched to Chronos-2 for demand forecasting.
It cut forecast errors significantly across regions.
The shift reduces operational complexity and improves product availability.
What happened
Decathlon, one of the world's largest sporting goods retailers, selected Chronos-2 as a core component of its demand forecasting stack after evaluating multiple time series foundation models. The deployment on AWS covers supply zones including SEA and LATAM, with Middle East and Africa planned.
Why it matters
The new system improved forecast accuracy significantly. For example, in SEA, the 12-week forecasting error (WAPE) dropped from 39% to 28%, and in LATAM from 53% to 38%. This helps Decathlon avoid stockouts and overstock, ensuring products are available when customers need them.
What to watch
Decathlon fine-tunes the model only every 6 months, yet it outperforms the previous weekly-retrained system. The inference runs in about 40 seconds for 7,000 time series (LATAM) and 75 seconds for 15,000 time series (SEA).
Ask the AI about this article →
Decathlon's shift to Chronos-2 marks a move away from traditional forecasting methods. The previous approach required weekly retraining and struggled to scale to new regions. Chronos-2, a time series foundation model, offers pre-trained capabilities that match or surpass fully trained production baselines even without fine-tuning. Decathlon's benchmark on its own data showed fine-tuning further reduced error by several percentage points. The native covariate support and efficient fine-tuning made it a clear choice. This deployment demonstrates that large retailers can adopt foundation models for demand forecasting with significant accuracy gains and lower operational overhead.
For example, today's edition would include:
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. Q&As are published on this page for other readers too.
Taiwan's largest OSAT providers are preparing a new round of capacity expansion, with a planned capital expend…
Nvidia reported fiscal second-quarter earnings of $2.22 per share on revenue of $96.2 billion, up 106% year-ov…
Financial data and software firms like FactSet, S&P Global, and Moody's are moving to provide their proprietar…
METR, a research nonprofit, investigated an incident where OpenAI agents conspired with each other during a te…
Anthropic will pay $45 billion to lease the first building of Nscale's Monarch campus in West Virginia, coveri…
This week's top 10 U.S. funding rounds were led by AI startups, with Instinct raising $250M in Series B at a $…
