Our laboratory. Empa’s Laboratory of Biomimetic Membranes and Textilesaims to develop materials and systems for the protection of the human body and its health. The products developed in collaboration with industry are used in the fields of occupational safety, sport, medical applications, and health-tech.Background.Optimizing postharvest cold chains of fruits and vegetables across all unit operations is of key importance to maintain fresh food quality and to reduce food losses. Temperature and the gas composition in the air are affecting decay and food quality, so they need to be controlled during precooling, refrigerated transport, and cold storage. By optimizing these environmental parameters, shelf life can be maximized. Currently, extensive monitoring is performed on the environmental conditions in food supply chains (air temperature and humidity). However, often, such sensing only covers a part of the cold chain and not the entire journey from farm to fork. Taking into account the entire journey is essential to quantify how the fresh-food quality evolves, and what the final quality and shelf life are that the retailers, and thus the consumer, receive. In addition, only simplified analyses are performed on these huge datasets. This makes that there is most probably a lot of unexplored information in the data. Objective. The key objective is to better use this sensor data to identify when and where the quality loss occurs, and how commercial cold chains can be improved, to reduce food loss.To this end, data-driven and physics-based modeling are used, among others by integration into digital twins.This project is performed in collaboration with a Swiss retailer. In this project, you will be able to improve future supply chains of fruits and vegetables in order to reduce the environmental impact of the food we consume.Your tasks:Organize experiments in commercial cold chains, process the measured sensor data statistically, reformat data in databases, and analyze data for variability.Analyze the data with statistical techniques (e.g. PCA, Monte Carlo) to identify critical problem locations in the cold chain. As a next step, more advanced data-driven techniques, such as machine learning, can be explored.Use this data to build up and use physics-based and/or data-driven digital twinsPropose solutions to improve the shelf life and reduce food losses and test these solutions in the field by full-scale trails.Support other physics-based and data-driven projects in the lab and set up a framework for quality assurance in modeling & simulationSupport in the organization of modeling and simulation events.The work will be performed at the facilities of Empa (St. Gallen). Regular visits to the facilities of the Swiss retailer company are likely to be required.