WG Packaging and storage
The objective of this working group is to develop an integrated system that combines sensors and mathematical models to optimize packaging and storage conditions for fresh produce. By leveraging sensor technology, mathematical modeling, and data-driven control strategies, the goal is to improve the quality and extend the shelf life of fresh produce during storage (see also 'Rethinking packaging').
Respiration sensor for real-time monitoring of gases in fresh produce storage. Equipped with an optical O2 sensor (0-25% range) and two CO2 sensors (0.5% and 20% range), it provides crucial data on O2 and CO2 concentrations. Successfully used for fruit respiration monitoring in commercial apple storage, the modular system shows potential for in-situ monitoring and integration into RQ-based DCA storage. However, further development is needed to enhance efficiency, effectiveness, data analytics, and hardware integration.
Airflow sensor for measuring air speed between fruits stored in bulk bins. To optimize airflow and reduce costs in cooling agricultural produce, understanding the spatial distribution of airflow is crucial. Hence, an omnidirectional airflow sensor was developed to measure air speed between horticultural produce stored in bulk bins. This sensor enhances comprehension of airflow conditions near the stored produce, ultimately improving the quality of fruits and vegetables in cold storage.
Condensation sensor for fruit cooling and cold storage:
To maintain the quality of stored fruits, managing humidity and condensation is vital. To address this challenge, a condensation sensor was developed to directly detect surface condensation on fruits. The sensor consists of a 0.1 mm thick electrically conductive film integrated onto a standard flexible printed circuit board (PCB), incorporating interdigitated finger electrodes on a single 0.1 mm thick polyimide foil that is attached directly to the fruit surface. The sensor is designed to monitor the formation, retention, and evaporation of condensed water on apple surfaces under different postharvest scenarios, including temperature fluctuations and re-warming. Future work will focus on integrating real-time sensor measurements with shelf-life prediction models to enable optimized cooling strategies and proactive condensation management in fruit storage and other cold-chain environments.
Heat flux sensor for measuring heat transfer during fruit cooling. It captures the initial heat flow from the food to the surroundings and monitors temperature fluctuations during long-term storage caused by compressor cycling and metabolic heat from the fruit. This real-time sensor, based on a Peltier element (PE), enables understanding of the heat exchange between the fruit and the cooling system. The data provided by the sensor aids in modeling and designing new refrigeration systems, with the goal of extending food shelf life and achieving energy efficiency.
Mathematical modeling drives our packaging design process. It considers factors like gas and water vapor permeability, respiration rate, and transpiration of specific fruits. We also account for storage temperature, package size, product quantity, and optimal O2, CO2, and humidity conditions to minimize moisture condensation. Leveraging our engineering expertise, we've developed predictive mathematical models for fresh produce packaging design and shelf life. These models are applied to select packaging materials and undergo simulations and validation through lab-scale and commercial-scale trials across a range of products.
Ethylene modelling for fruit packaging. An integrative model describes ethylene production by the fruit, its transmission through the packaging and perforations, and its removal by ethylene scavengers. The model predicts ethylene accumulation in the headspace of the package under different temperatures, package configurations (size and number of perforations), product mass and scavenger kinetics. A web-based simulation interface transforms the model into an accessible design tool for evaluating packaging options and supporting decisions aimed at optimising shelf life.
Integrating mathematical models, sensors, and electronic modes is an effective approach for real-time control of O2 and CO2 levels in storage or transport containers for fresh produce. A mathematical model that balances respiration rate and gas exchange through a perforated tube-blower assembly enables indirect gas control based on temperature. Thermocouples monitor‘s container temperature, triggering the air-blower to adjust its operation if deviations occur. The model estimates blower ON time to regulate O2 concentration by exchanging air between the container's inner and outer atmosphere. Currently, the concept is suitable for CO2-tolerant commodities with higher CO2 control (>15%). Further research is needed to implement this approach in real-world supply chains with variable temperature fluctuations.
Interview with Dr Tuany Hoffmann on the development of and work on digital twins:
Digital twins for fruit storage. Our recent work focuses on digital twins for commercial fruit storage. Distributed sensors measure the local conditions experienced by the fruit, including air temperature and relative humidity, fruit-surface temperature, airflow, heat flux and surface wetness. These measurements are transmitted wirelessly and processed in real time. Physics-based models use the live sensor data to calculate dew-point undershoot, predict condensation and evaporation, estimate surface-wetness duration and determine fruit-mass change caused by transpiration, respiration and condensation. The platform has been tested during commercial apple storage with multiple wireless and wired sensor units. It combines LoRa-based communication, real-time data processing, model services and dashboard visualisation. The system currently supports monitoring and interpretation; automatic risk alerts and closed-loop storage control are future development steps.
Real-time ethylene prediction for fruit storage. We are developing a model-based digital platform that combines live measurements of temperature, O₂, CO₂ and airflow with dynamic models of ethylene production, removal and transport. Sensor data continuously update the model to estimate ethylene concentration without relying on dense networks of dedicated ethylene sensors. State-observer methods reduce measurement noise, spatial modelling estimates ethylene distribution, and uncertainty analysis quantifies prediction reliability. The approach is validated in controlled fruit-storage experiments and is intended to support scalable digital monitoring and decision support for postharvest storage.
