ADVANCEMENTS IN MILK QUALITY MONITORING: A REVIEW OF VIRTUAL SENSORS AND DIGITAL TWIN APPLICATIONS

BanuPriya G N, Harshitha K S, Manjula S, Niharika S, Priyanka C R, Sonam Singh R, Nalini Chitta.

With a focus on temperature and spoilage detection throughout the dairy chain, this review examines real-time milk quality monitoring. Microbial growth is significantly influenced by temperature, and IoT-based systems use microcontrollers in conjunction with temperature, pH, and gas or conductivity sensors to continuously monitor milk conditions and initiate early spoilage alarms, frequently 10–12 hours before obvious degradation. Virtual-sensor methods reduce dependency on laboratory testing by combining physical data with machine-learning models to forecast spoilage risk from temperature and spoilage-related parameters. By integrating such sensor inputs into cloud-based models of dairy tanks and refrigeration units, digital twin-oriented architectures enable dynamic monitoring and decision support for farm-to-factory operations. This work highlights scalable, inexpensive IoT and AI-driven frameworks appropriate for both small-scale and industrial dairy settings by consolidating existing architectures for real-time milk monitoring.The energy-efficiency and practical deployment issues of sensor nodes in distant dairy farms, highlights the necessity of reliable, low-power communication protocols for continuous real-time monitoring.

Keywords –IoT-based systems, Machine-Learning models, Microbial growth, Temperature and Spoilage-related parameters, Digital twin-oriented architectures.