You are using software which is blocking our advertisements (adblocker).

As we provide the news for free, we are relying on revenues from our banners. So please disable your adblocker and reload the page to continue using this site.
Thanks!

Click here for a guide on disabling your adblocker.

App icon
FreshPublishers
Open in the app
OPEN
AI-Powered harvest forecast for pepper growers launched:

"We can better align the expected supply from our growers with the demand from our retail customers"

Trusted by tomato growers worldwide, Source.ag's AI Harvest Forecast solution now expands to pepper. "The model cuts forecast error nearly in half, giving growers, labor managers, and sales teams a reliable weekly picture of what's coming, up to eight weeks out", the team of the data and AI platform says upon the commercial launch of their pepper solution, with additional varieties expected in the coming weeks.

Harvest Forecast replaces the manual yield forecast calculator for peppers that growers have relied on within the Source Platform to estimate weekly harvest volumes. Instead of building forecasts by hand, pepper growers now get an automated, data-driven weekly prediction built from their cultivation data.

Results from pilot growers
Early results from pilot growers at Harvest House and Jansen Paprika's show the AI model cut average forecast error at the three-week horizon by more than 40%. Outlier weeks, where forecasts missed by a wide margin, dropped 28%. On certain varieties, accuracy improved even further.

© Source.ag

"With Source's Harvest Forecast AI, we can better align the expected supply from our growers with the demand from our retail customers," says Marco Hanemaaijer, Strategic Supply Chain Manager at Harvest House. "In the tomato segment, the tool is already proving its value: an accurate harvest forecast gives us control over the entire supply chain. We see strong potential for its application in bell pepper cultivation. Two of our bell pepper growers are now piloting it, actively sharing their experiences and contributing to its further development so that Source.ag can deliver a robust and practical solution."

"What used to be done manually in Excel — counting, entering data, making forecasts — now happens automatically with Harvest Forecast from Source," says Gerard Kremer, Cultivation Manager at Jansen Paprika. "In pepper growing, you always want to know what's coming. If you can predict that accurately, you can manage more effectively and efficiently, and align better with sales and planning. With labor costs continuing to rise, more efficient planning is no longer a luxury, it's a necessity. Source helps us do exactly that. Harvest Forecast for peppers has only just launched, but we already see a lot of potential in how it will develop."

Why the three-week horizon matters
For pepper sales and planning teams, the three- to four-week forecast window drives the most important decisions: supply commitments to retailers and cooperatives, labor scheduling, and logistics. Growers submit weekly harvest forecasts to their cooperatives and trading partners, a process that has historically been manual and error-prone.

Source's model updates the forecast automatically any time new information is available, combining plant data, greenhouse climate, and cultivation strategy. Head Growers enter rounds and percentage colored data weekly. The model handles the rest.

Availability and scope
Harvest Forecast for pepper launches with support for six varieties: Alzamora, Gina, Cadalora, Yedda, Frazier, and Silverstone, covering a significant share of the block pepper customer base. Additional varieties are available in the coming weeks.

Building AI forecasting across crops
The pepper launch follows Source's recent announcement of Harvest Forecast for tomato to a next-generation AI model, which reduced forecast error by 25% across cultivations.

"Pepper growers have been asking us for this for a while", says Lotte Adema VP of Product at Source.ag. "We had the forecasting foundation from tomato, but pepper is a fundamentally different crop. Getting the model to perform at this level took serious effort from our Data Science and Plant Science teams. Now pepper growers finally have a forecast they can share with their Sales teams and stand behind."

Demos are available upon request.

For more information:
Source.ag
[email protected]
www.source.ag

Publication date:

Related Articles → See More