Quality improvement and workload reduction Gardener's Green Thumb
The production of ornamental plants and vegetables along the German-Dutch border is an important economic activity in the regional horticulture. However, the sector is increasingly under pressure due to rising labour costs, a shortage of skilled workers, international competition, and increasing quality demands from market players and consumers. Business leaders must increasingly work as horticulture "managers".
Data-driven decisions
The project "The Green Fingers of the Gardener" aims to contribute to relieving these business leaders in their daily tasks. Digitalisation, particularly artificial intelligence or machine learning techniques, can help smaller companies enhance their competitiveness.
The core of this project is the development of a self-learning "Decision Support System (DSS)". Activities during cultivation are recorded and automatically documented using sensors. With the help of machine learning methods, relationships are established from these generated observations to derive rules. These relationships are made between the current and historical environmental measurements (climate, light, plant, soil) and the decisions made in the cultivation (fertilisation, irrigation, climate change, etc.).
In the training phase, an interaction of the system with the grower may look as follows: ‘If in the past similar conditions prevailed in the greenhouse and on the plant, then you chose alternative A in 80% of cases, 15% for B and 5% for C. Do you want to choose A again? ‘(We know such proposals in similar systems, e.g. from the online store Amazon). As a result, the DSS continuously learns and can automatically initiate the necessary actions after a certain time and only inform the grower. The system can also derive rules from changed behaviour in production itself, for example when a new crop is set up.
The training phase
The system automatically seeks interaction with the grower in the form of proposals. The DSS continuously learns from all data and can automatically initiate necessary actions after a certain time and only inform the grower. The system can also derive rules from changed behaviour in the production itself, for example when a new crop is set up.
Made possible by:
Use this section to enhance your company's credibility.
