Data collection for sustainable machine learning in circular food industry/food production

Purpose and goal:

By 2050, global food production will need to increase by 70% to meet the needs of the growing world population (FAO, 2021). A circular food production is a prerequisite for meeting the increased needs for food in Sweden and the world. The project aims to create a common standard for data collection and subsequently ML (Machine Learning) models for streamlining circular food production.

Expected results and effects:

The project is expected to increase domestic self-sufficiency through the implementation of self-learning systems. Streamlined, circular food production has the potential to become a massive industry with enormous positive impact on welfare and the environment. Approach and implementation: Preliminary study/report

A structure for a quality-assured data topology is developed. The report will show how different data should be structured, handled and stored in order to easily digitize an industry. Pilot installations Pilot installation of a data warehouse with pre-processing that can identify deviations through ML.

Prototype A prototype based on ML that can optimize various processes, for example resource or cost optimization, is developed.