Information System and DSS tool for Cereals Cultivation – Digi-PILOTE

Delivering strategic advice to wheat farmers through a mobile application that processes information from the cloud and data from IoT solutions.


Concept

Farmers are dealing with increasingly unpredictable climatic conditions. Therefore, this Flagship Innovation Experiment (FIE) supports wheat producers in maintaining a high level of production in terms of quantity as well as quality. As this is becoming a problem not only for Mediterranean regions but for the world at large, it is crucial to put this FIE’s mobile application onto a larger territorial scale. The main goal is to optimize the irrigation and nitrogen fertilisation of wheat. To achieve this, the tool developed uses parcel data collected from sensors, satellites and crop models to generate technical and strategic advice for its end-users. Once the application reaches an operational scale, the incorporated information system will also ensure a seamless connection with other farm software, enabling producers to control critical parameters throughout the year.

To continuously upgrade the tool’s calibration and extend its functionalities, this FIE relies on a network of digital farms as well as farmer associations, which currently rely on different Decision Support Systems (DSSs) and parcel management software, each with their own data standards. In order to successfully manage the centralised information processing, the seamless data transmission to the cloud combined with assimilated data from various Internet of Things (IoT) solutions is at the heart of this mobile application.

Implementation 

 

The FIE was deployed in 28 farms, in wheat and durum wheat farm fields.

There were 3 phases of development for the FIE: the first focused on coupling the agronomic model and the sensors; the second the development of the IT platforms with an accent on interoperability and finally the dissemination of the tools to stakeholders at the regional and national levels.

The offer is divided into 3 services é-PILOTE, Prédict-IS and CHN Inside. 

The main objective was the agronomic validation, through dedicated trials, of the outputs of the CHN crop model coupled with leaf index and chlorophyll data from satellite data, an essential prerequisite before marketing. The objective of the agronomic model CHN is to pilot in real-time the irrigation and fertilisation activities. This model was developed for wheat, durum wheat and corn. With the data collected from sensors, the  CHN model could be improved and provide better predictions. The coupling is attractive, as data from sensors are not always reliable (issue with availability, damaged sensors, weather conditions). 

é-PILOTE integrates data from the fields t prepare modelisation with CHN to propose dynamic advice. Testing was performed on 21 different fields. 

Prédict-IS is in development and proposes to translate the CHN model and train farm advisors

The first step was the development of the information system to communicate the different data sources together: sensor’s data (IoT and weather stations), farmer’s data (é-PILOTE application) and the CHN crop model. During the implementation period, the é-PILOTE HMI was improved by adding new functionalities (ex: chatbot, plots, yield and protein grain forecast); the API which makes it possible to run the CHN model has been developed and also makes it possible to inject data from sensors (Airbus, Hiphen, connected weather stations). The expected benefits were a yield increase of about +0 to 10 quintals more per hectare.

Lessons Learnt 

Even if we were worried about our capacity to manage field trials during the two cropping seasons 2019-2020 and 2020-2021, at least all the trials were conducted, as well as IT developments. However, the dissemination of the results has been hugely impacted due to the impossibility of organising meetings with farmers.