Project PRAGMATIC (Prediction of Agriculture Manufacturing Cost) – R&D work in the area of yield and raw material cost prediction of agricultural products sourced in the supply chain from crop to production lineGrant Agreement No. POIR.01.01.01-00-2298/20-00 dated 30.11.2021. Purpose and planned end result of the project: The project aims to develop a prototype of an innovative IT platform containing algorithms for the prediction of production costs of agricultural raw materials for three reference crops, i.e.: blueberries, apples and potatoes in the supply chain from the field to the production line. The system will use predictive models based on machine learning and artificial intelligence methods, using ground data (soil, plant, climate, crop technology) and satellite imaging data (crop and soil condition). The system will consist of the following modules: – Knowledge Base (BW) module in the form of a data repository, synchronized on an ongoing basis with available datasets, – Violation and Anomaly Detection (WNA) module, which will allow detection of data records that do not match the analyzed datasets,- Yield Prediction Module (PPL) which will allow prediction of yield for a selected crop type taking into account various factors (attributes), – Cost Prediction Module (PPK) will serve users as an aid to production optimization and budget planning, – Results and Configuration Visualization Library (WWK). This library, will be a set of various components, used by analytical modules to visualize results and display them in the form of diagrams and charts with appropriate marking of prediction confidence levels (PPL and PPK modules) and signaling of possible problems in case of detection of anomalies in the data (WNA module). The main recipients of the project’s results will be medium and large farms, cultivating selected varieties of apple, potato or blueberry, which are looking for intelligent, precise, algorithm-based tools, allowing them to obtain yields adequate to market demand and carry out agrotechnical treatments in an environmentally, climate and people-friendly manner. The goal of the project will be achieved through:

  • Develop and produce laboratory hybrid predictive models based on the most end-user accessible data, without defining varietal diversity within species.
  • To produce tools to carry out the optimization and testing process in a near real-world environment,
  • Preparing the prototype for commercialization by involving end users in the evaluation and validation of the developed system.

Value of the Project The total amount of eligible expenditure is PLN 3,753,823.14 European Funds contribution The maximum amount of funding – European Funds contribution – is PLN 2,851,927.94, which is 75.97% of the total eligible expenditure.