Loading...
Menu
Loading...
Artificial intelligence for agriculture and food processing. Tune the irrigation, forecast the harvest, and automate quality control in the country's breadbasket.
The Saiss plain around Meknes is some of the most productive farmland in Morocco: olives, vines, cereals and citrus, feeding a food processing sector that is still organising itself. The pressures are equally real, though. Water is tightening, prices move, and export buyers now want traceability on paper.
Soil probes paired with machine learning let you run drip irrigation on evidence, forecast yield per plot, and catch a disease while it is still in one corner of the field. The National School of Agriculture and Moulay Ismail University are on the doorstep, which makes the region a natural place to do this work.
We work with the farms, cooperatives and food businesses of Meknes on AI that pays for itself and survives a Moroccan growing season.
AI connected to field sensors, managing water, inputs and the crop cycle on what the ground reports rather than on the calendar.
Computer vision for olive sorting, fruit ripeness checks and defect detection on the packing line.
Water use down 20 to 40 percent, fewer post-harvest losses, and production costs you can compare against last year rather than against a projection.
Our solutions in Meknes
Moisture probes and weather data driving drip irrigation in real time. On the olive and vine estates of the Saiss plain the drop in water use is measurable, not theoretical.
Machine learning over satellite, climate and historical data to forecast yield plot by plot, so seasonal labour and post-harvest logistics get booked against a real number.
Automated inspection by computer vision: olives sorted by size and ripeness, foreign bodies caught on the packing line, and label checks that hold up at the European border.
Digital platforms for member management, traceability on what each member delivers, revenue distribution, and the organic and fair-trade certification paperwork the region's cooperatives have to keep current.
Meknes sits in the middle of the Moroccan agricultural value chain, and we work along the whole of it, from the field to the packing house.
FAQ
That is where the gap is widest, because a cooperative's difficulty is organisational rather than technical. Centralising deliveries, calculating distributions and issuing each member's paperwork takes weeks of manual work every season. A member can check their own deliveries from a phone instead of coming to the cooperative office.
Over one season the gain shows up mainly in water and in losses avoided at sorting, rarely in yield itself, which depends first on the weather. So we scope the commitment to pay for itself on those two lines, because promising a yield increase in year one would amount to selling you a favourable season.
No. On most Saiss farms we see, the existing network can be controlled as it stands. We add soil moisture probes, a weather station and control on the valves already installed. A very old or very mixed network will need some localised work, which we price after a site survey rather than over the phone.
Weather stays the dominant variable, so the model works on the uncertainty rather than on a firm figure. It combines your plot history, in-season readings and regional climate data to give a range that tightens as the season progresses. That is enough to negotiate a supply contract, not enough to promise a tonnage in January.
Back to the plot and the delivery date, which covers what export buyers usually ask for. The limit is what you record today: a cooperative keeping delivery notes on paper cannot reconstruct past seasons. So we start by making the capture reliable, and traceability then builds itself season after season.
Ask for a free audit of your farm or your processing unit. We identify where AI actually helps and send an action plan within 48 hours.