TERRITORY
WinePredicter
The WinePredicter project consists of developing an agronomic information system for vine cultivation based on remote sensing and AI techniques.
SPAIN
AREA TO BE DIGITISED
Production Processes
SUBSECTOR #1
Agriculture
SUBSECTOR #2
Vineyard winemaking
CROP PRODUCTION SYSTEM 1, 2, 3
Rainfed
Conventional
Open-field
TECHNOLOGY
Big Data and Data Analytics, AI and quantum computing, Machine Learning
DIGITAL SOLUTION CATEGORY
Precision farming
STAKEHOLDERS
Agricultural cooperativas and associations
IMPACT
Productivity
The WinePredicter project (Monitoring and prediction tool for vineyard growing) consists of developing an agronomic information system for vine cultivation based on remote sensing and artificial intelligence techniques. This will enable substantial optimization of vine cultivation management, reduce inputs through maps that enable precision farming techniques, and increase grape production and quality.
Farm challenges
The main farm challenges were managing the high spatial and temporal variability of vineyards, anticipating yield and grape quality months before harvest, and integrating heterogeneous historical, climatic, and spectral data at parcel and sub-parcel level.
Assistance / Boost program
Ayudas a Agrupaciones Empresariales Innovadoras (AEI).
Innovative features of the initiative / solution
The solution innovatively combines AI-driven agronomic models with satellite and drone multispectral imagery, historical vineyard data, and an automated WebGIS platform to enable precision viticulture and early prediction of yield and grape quality.
Results obtanied
The project delivered validated predictive models for vineyard yield and grape quality, along with a fully operational, scalable monitoring platform that supports weekly vineyard tracking and harvest planning up to two months in advance.
Lessons learned
The lessons learned are that high-quality historical data and parcel-specific calibration are critical for reliable AI models, and that combining domain expertise with automated data processing significantly improves robustness under climatic variability.
