UN-FAO reports show that about 30% of all food produced is not consumed in the end. To be able to feed future generations, innovations are needed to reduce this figure and to improve the efficiency of the food system.|
By nature of the product Food supply chains (FSCs) need to be managed and controlled carefully. Existing technologies should be better exploited and new technologies and innovation should be developed. Through sensors more and more data can be retrieved, which provide information that can be used to improve supply chain decisions, ranging from pre-harvest and post-harvest planning of production, processing, distribution and inventory management. Thus one can optimize the supply chain from Farm to Fork.
We consider three stages of a FSCs: pre-harvest decisions, food processing, and retailing.
These decision should be aligned to sustainability goals (including environmental, animal well-being, etc). during the course a number of decisions will be selected and solved using quantitative modelling and data driven approaches.
- pre-harvest decision relate to
- the choice of cultivar/species,
- the efficient use of resources (precision farming): water, fertilizers, (plant) pesticides, (animal) medicines, and energy,
- where to produce: worldwide, global vs local, greenhouses, land use vs vertical farming,
- processing decision relate to sourcing and processing (heterogeneous) living organisms into food product with often a short shelf life,
- retail decisions relate to distribution, inventory management (assortment planning and replenishment decisions), and sales forecasting.
Precision farming helps the farmer use data and information from the farm to improve efficiency of the production and adjust farm management to the varying conditions in different parts of the world. Data science can help in improving animal welfare by monitoring animal behaviour and animal health, stress can be reduced and animal welfare improved.
Processing industries can better match demand and supply by predicting the quality of raw materials using historical data. Food waste at retail outlets can be improved by better demand forecasting, e.g. including weather data to better predict demand and to optimize order quantities accordingly. Next to traditional forecasting techniques students are invited and challenged to apply other data driven approaches (e.g. machine learning) during computer practical sessions. The program is enriched with a field trip and a guest lecture.