Streamlining Real-Time Sensor Data Collection and Querying Processes From Environmental Parameters in a Food Cold Store

Logistical inefficiencies in fresh produce supply chains lead to significant food loss, which can be mitigated through real-time monitoring of cold storage environments. This project streamlines data collection and analysis using a Digital Twin methodology based on the Actor Model. Each sensor is modelled virtually, developing protocols for decoding manufacturer-specific payloads and creating an open-source data pipeline for automated data requests, decoding, database updates, and real-time analytics. The implemented system provides immediate insights into temperature, humidity, and gas levels, enhancing food preservation, extending shelf life, and reducing waste. This framework lays the foundation for scalable, costeffective cold chain monitoring, with potential extensions to refrigerated transportation, predictive analysis and machine learning applications, positioning it as a valuable tool for optimizing cold chain logistics and minimizing food loss.