Explore Our Pilots

Pilot 1 refers to the “Wiatrowy Sad” (Wind Orchard). It is a 17-hectare family-owned apple orchard located in Kałęczew, within the Lodz Heights Landscape Park (Lodz province) in Poland. The region is known for its rich agricultural history, with fruit cultivation gaining momentum after World War II, bolstered by the establishment of the Institute of Orchardery and Floriculture. This institute played a crucial role in spreading knowledge on modern apple-growing techniques, which the owner family utilized in developing their orchard. Wiatrowy Sad has been recognized with numerous awards for the quality of its products. It is known for its dedication to high standards, combining traditional methods with modern technology.

As part of the AgriDataValue project, Wiatrowy Sad integrates the SynField smart agriculture system, which includes a meteorological station and soil sensors to monitor real-time environmental data. This technology helps optimize orchard management by providing precise information on ambient temperature, humidity, wind speed, and soil conditions. Gathered data supports precision agriculture by allowing for more informed and sustainable decisions, optimizing fruit production and orchard management. The integration of this technology represents a step towards a digital orchard model, where data-driven solutions ensure efficiency and sustainability in response to market demands.

Pilot 2 is situated at an arable farm in Swifterbant, Flevoland, The Netherlands. The involved partner is Delphy. The pilot is being conducted in onion cultivation. The trial field contains 5 objects in 4 replications, i.e. 20 plots. These 5 objects all have an RMA-soil sensor (data via API) and a different irrigation strategy. We can provide the following data: the crop yield per object (average over 4 replications) and the mm of irrigation per object (average over 4 replications) per season.

Pilot 2 focuses on reduce wasted irrigation water while ensuring healthy crop growth and stable yields. For these purposes all the required data streams are established. Results of this Pilot can be verified through a combination of quantitative measurements, comparative analysis, and agronomic validation that jointly demonstrate reduced water use without negative impacts on crop performance.

Pilot 3 is situated at an arable farm in Zemgale, Latvia. Responsible partner for this pilot is ZSA, which is the biggest agricultural organisation of producers in Latvia. Wheat and hard wheat are grown on the field. The main subcase of this pilot is managing the winter wheat leaf diseases. Pilot 3 focuses on reduction of the usage of pesticides by identification of wheat leaf diseases. For these purposes all the required data streams are established.

Pilot 4 is located in Agrinio, Etoloakarnania (Greece), and covers forages, i.e. clovers and corn crops for livestock feeding. This pilot is owned by TBA and directly involves 15 farmers. Data collection for Pilot 4 is conducted via SynField devices which collect the measurements from the deployed sensors and forward them to SynField platform. From there, the measurements are extracted periodically and saved in the STORE component of AgriDataValue.

This pilot 5 takes place on a demonstration field of InAgro in Belgium that consists of multiple sections that have a crop rotation of both vegetables and arable crops. This demonstration field can be used to try out the newest innovations in agriculture in an environment that simulates an actual agricultural field without any risks to farmers. Pilot 5 delivers data of Inagro’s fields, mainly the Optifarm field containing multiple crops with a rotation of both arable crops and vegetables. The RGB data can be used to detect weeds, to create a model for spot-spraying which reduces pesticide usage. Yield data can be used for biomass estimation and nitrogen uptake modelling.

Pilot 6 focuses on the optimization and predictive management of greenhouse vegetable production (tomato and cucumber) under Mediterranean conditions in Almería (Spain). The responsible partner for this pilot is Tecnova, an Andalusian Technological Centre for the Agricultural Industry, comprising approximately 125 private companies. In this pilot, tomatoes and cucumbers are grown in a greenhouse environment. The main objective of this subcase is to optimize the management of crop development.

The pilot addresses three complementary use cases: precision irrigation and fertilization management, quality prediction through soluble solids content (ºBrix), and automated climate control via greenhouse window management. For these purposes, all the required data streams are established and integrated at pilot level.

Pilot 7 is situated in West-Flanders in Belgium with InAgro as responsible partner. The pilot is focused on vegetables such as Belgian Endives, potatoes, spinach and beans. Sufficient soil moisture is of great importance in these crops, for Belgian Endives especially around the time of seed emergence. These seeds are expensive, and a good emergence is essential to qualitive product. Soil moisture can be measured in field with sensors, but these sensors are often expensive and require proper installation and maintenance. Some commercial websites offer a soil moisture estimation based on satellite imagery. IOT GEOBAS weather stations were installed. These weather stations can measure soil moisture, soil temperature, water potential, precipitation, relative humidity and air temperature. However, the pilot mainly focusses on soil moisture estimations through satellite imagery.

Pilot 8 on leeks, a model to estimate nitrogen uptake will be made. It is the intention of optimizing the nitrogen fertilization by determining the nitrogen uptake over time by the crop. This will allow the farmers to reduce fertilization by using precision farming techniques. A correctly timed and dosed fertilization will also increase the leek yield and/or quality. Data was acquired by GEOBAS weather stations. This weather stations can measure soil moisture, soil temperature, water potential, precipitation, relative humidity and air temperature.

Pilot 9 focusses on post-harvest potato quality. A classification model will be created that will use hyperspectral images of potatoes to detect defects such as black spot.

Pilot 10 created a learning network of farmers willing to do experiments on reducing fertilizers with open field precision fertilization and reducing pesticides. Above all the goal is to increase farmers’ digital independence.

Pilot 11 is based in The Netherlands and will focus on the crop off apple. A model will be made for determining the crop load levels to achieve the highest yield without risking an off year for the next. Another model will predict the fruit size.

Pilot 12 is being implemented on stone and pome fruit farms in Aragon, northeastern Spain. It combines meteorological data from the SIAR network with field observations on phenology and pest prevalence from the Aragon Phytosanitary Network (Red FARA). Using Big Data and AI under the guidance of government agronomists, the project develops Machine Learning models to predict phenological stages and disease risk for each plot. Accurate forecasts help farmers optimize the timing and necessity of phytosanitary treatments, whose effectiveness depends on crop stage. The resulting dataset has already uploaded at zenodo repository, and it is available at: https://doi.org/10.5281/zenodo.17975075

Pilot 13 is located in Amfilochia, Etoloakarnania, Greece. The focus of the pilot is Pest Control on Mediterranean Fruit Fly. For pilot activities SynField smart agriculture devices has been installed, along with: a meteorological station for the collection of environmental measurements (e.g. wind speed and direction, temperature, relative humidity, rainfall), a soil sensor for the collection of soil-related measurements (temperature, volumetric water content, electrical conductivity), a leaf wetness sensor, and, a pyranometer for the measurement of solar radiation. Additionally, Synelixis SynTrap nodes have been deployed in the pilot, utilizing computer vision and AI for advanced pest monitoring. The system combines a traditional pheromone trap with a high-resolution micro-camera and a SynNano device (used for providing connectivity).

Pilot 14 is located in the Saint-Émilion vineyard in France. The objective of the pilot is the early detection and prediction of frost events, in order to support winegrowers in optimising their active frost protection systems. The Saint-Émilion vineyard covers approximately 7,500 hectares and is equipped with a large network of SynField smart agriculture devices and weather stations, enabling the exploitation of historical meteorological data (air temperature, air and soil humidity, soil temperature, wind, etc.), particularly during the past frost periods. Satellite data are also available and will be used to carry out a model for the prediction of frost events.

Pilot 15 is a vineyard, located in Tebano (RA), within the Emilia-Romagna region (northeast Italy). Covering 7 hectares, the vineyard is situated on flat terrain with a clayey loam soil. Cultivation follows an integrated management approach, combining sustainable practices to ensure a balanced ecosystem. The varieties grown are Sangiovese and Trebbiano, both grafted onto KOBER 5BB rootstocks. Pilot 15 is related to two use cases: 3.1 - Fruit trees disease forecast/detection; 3.2 - Anti-frost control. The data collect by the soil sensor is the soil moisture and soil temperature while the meteorological station records ambient temperature, relative humidity, wind intensity, wind direction, and rainfall.

Pilot 16 is located in Chora, Messinia in the Peloponnese Region of Greece and focuses on an olive grove involved in use cases UC 3.1, UC 3.2, and UC 3.4. The objective of the pilot is to support olive growers in managing tree diseases—most notably olive anthracnose, caused by Colletotrichum species—by improving early detection and outbreak forecasting to minimize losses in yield and quality. In addition, the pilot aims to assist growers handle climate-driven frost events, which threaten tree health, productivity, and olive oil quality, as well as control the olive fruit fly (Bactrocera oleae), a critical pest that requires effective monitoring. To power these efforts, the NILEAS olive orchard is equipped with smart IoT devices and sensors such as weather stations to monitor rainfall, air temperature, humidity, wind speed and direction. These streams feed weather forecasting models and track on-farm data like irrigation records. Additionally, the orchard features a dense sensor network powered by Synelixis’s SynField smart agriculture devices, which continuously monitor soil conditions, air moisture and temperature.

Leccino, Correggiolo and Ascolana. Responsible partner for this pilot is RI.NOVA. Pilot 17 is related to two use cases, namely: 3.1 - Fruit trees disease forecast/detection; 3.4 - Pest Control on Olive Fruit Fly. Regarding the olive fruit fly monitoring, data collection is based on observations of the traps. Data gathering is essential for detecting early signs of olive fruit fly infestations, enabling timely interventions that align with organic pest control practices.

Pilot 18 is located at BioRo in Romania. The use case focused on reducing wasted irrigation water enabling continuous monitoring and data-driven management of irrigation practices. At local level, the use case is managed through the active involvement of farmers and technical operators, who rely on the deployed digital tools to support daily irrigation decisions. The system enables the collection, integration, and analysis of heterogeneous data sources relevant to irrigation efficiency and crop water requirements. The pilot activities involve deploying IoT devices to collect data and monitor environmental and soil conditions directly from the field to users' smartphones or tablets. Key measurements include soil moisture and temperature, wind speed and direction, air humidity and temperature, leaf wetness, and solar radiation, alongside essential agronomic indicators such as evapotranspiration, dew point, precipitation sums, growing degree days, and chill hours. In addition, the system tracks operational data from irrigation systems—such as irrigation duration and water flow—to identify inefficiencies and potential water losses. Historical datasets and analytical models are then used to generate tailored irrigation recommendations based on local conditions. All data is aggregated and processed within a common digital platform, delivering visual dashboards, alerts, and decision-support tools to end users. The use case is validated by measuring reductions in unnecessary irrigation events and overall water consumption, as well as by evaluating how closely irrigation schedules align with actual crop needs.

Pilot 19 is situated in Melle, Flanders (Belgium) near Ghent, and covers a state-of-the-art Dairy barn at the ILVO Animal Science Unit research farm. In this dairy barn, various research trials are performed on topics such as feed components and feed efficiency, enteric emissions, colostrum quality, youngstock management etc. The gathered data includes general cow information, milk production and milk quality, etc. In pilot 19, methane emission reductions are aimed at by creating models able to predict the enteric methane emission from cows. Using such prediction models, methane reducing strategies can be tested and the effect of a strategy and its obtained reduction can be monitored at farm level. Furthermore, the cow data are combined with the urea content of the milk as labelled data to create a prediction model for milk urea content as an indicator for nitrogen emissions to create a prediction model that can be used to predict urea excretion via the milk, hence serving as an indicator for nitrogen emissions to test potential nitrogen emission strategies at animal level. In addition, general cow data are combined with data of cow activity and rumination gathered by collar sensors to define a prediction model for mastitis development. Documented mastitis cases linking mastitis cases to certain cows, dates and severity levels are used as labelled data for model creation. The model will hence be used to predict mastitis cases at an earlier timepoint, thus providing the possibility to take action sooner and potentially prevent the severity from worsening. Finally, general cow information on the gestation including gestation stage, expected calving date, inseminations dates are used in combination with activity data to predict the timing of the onset of the calving. The created model will be used to predict the onset of the calving more accurately, allowing the farmer to better monitor the process and potentially detect issues at an earlier timepoint, thus allowing taking action sooner to prevent potentially life threatening health issues for the cow and calf.

Pilot 20 is situated in Vecauce (Latvia) and covers a Dairy barn at the LBTU teaching and research farm with ZSA as a project partner. Vecauce is a multidisciplinary farm that combines student training, research, crop farming, dairy farming, biogas production, fruit growing, forestry and has created one of the largest and most productive herds of dairy cattle in Latvia. The barn houses more than 1000 cows. In this dairy barn, various research trials are performed on topics such as feed components and feed efficiency, enteric emissions, colostrum quality, youngstock management etc. The gathered data includes individual cow recognition, including general cow information, milk production and milk quality, etc. The production data are supplied by the feed data, milk data (production, fat, protein, lactose, urea), animal weight, etc.

Pilot 21 is located in Katouna, Etoloakarnania, Greece, near Agrinio, covers and Organic Cattle Farm and is owned by TBA (To Biologiko Agroktima - TBA), which actually stands for Organic Cattle Farm in Greek. This pilot pertains to a cattle fattening system. Approximately 320 of TBA’s animals are involved in Pilot 21. TBA has installed 281 smart collars from FarmLife which is collecting data related to rumination, heart bit (and multiple heart-bit in case of pregnancy), and comfort conditions. Data will be used for AI model training and federation to other pilots.

In the context of Pilot 21, IoT devices have been installed to collect data (SynField X5 device along with a SynAir device). The measurements that are collected by these devices include:

• Air temperature and relative humidity
• Barometric pressure
• Particulate Matter (PM)
• CO2 emissions
• NH3 emissions

Pilot 22 is situated in Melle, Flanders (Belgium) near Ghent and covers a Pig Campus with ILVO as partner. The pig campus comprises a pig barn fully equipped to perform research on pigs, including sows, piglets and fattening pigs. The production data are the commonly used pig farm indicators such as the number of piglets per litter per sow, the survival rate, temperature in the different barn compartments, and ventilation data. The barn has sensors for monitoring animal behaviours such as eating patterns, drinking patterns, water consumption, live weight, lameness detection, behaviour, and location in the compartments. Sensors monitor different types of nutrition with feeding stations, weather conditions (heat stress), management, stable environment, welfare and, behaviour, physiology and morphology of pigs. Within Pilot 22, methane emission reductions are aimed at by creating models able to predict the enteric methane emission from fattening pigs. Using such prediction models, methane reducing strategies can be tested and the effect of a strategy and its obtained reduction can be monitored at farm level. Firthermore, in this pilot pig and feed data are combined with air ammonia measurements as labelled data to create a prediction model for ammonia emissions. Similarly, the goal is to create a prediction model that can be used to predict ammonia emissions based on available farm data, and hence serve as an indicator for ammonia emissions to test the effect of potential emission reduction strategies. Finally, a prediction model for faecal consistency score is used as an indicator for diarrhoea in piglets. The model will hence be used to predict diarrhoea cases at an earlier timepoint, thus providing the possibility to take action sooner and potentially prevent the issue from worsening and allowing animals to recover sooner.

Pilot 23 is located in Agrinio, Etoloakarnania, Greece and covers electricity generation from biogas. The pilot is closely located and directly associated with Pilot 4: Forages and Pilot 21: Organic Cattle Farm, and is owned by TBA. In the context of the pilot, manure originating from the Organic Cattle Farm (Pilot 21) is transferred via underground pipes to anaerobe digesters where biogas is produced. The produced biogas is utilised by a generator of 5MW which directly provides electricity to the smart grid. The solid and liquid remainders of the procedures are used as fertilizers and irrigation at Pilot 4 which in turn produces forages (clover and corn) to feed the animals in Pilot 21 with, thus closing the loop and forming a fully circular ecosystem (UC 5.1: Fully Circular ecosystem).