In the heart of South Africa, a groundbreaking study led by Ismail Adeleke from the University of Johannesburg’s Center for Cyber-Physical Food, Energy and Water Systems (CCP-FEWS) is revolutionizing the traditional art of amasi production. Amasi, a cherished fermented milk product, has long been a staple in many households, but its production has remained largely manual and labor-intensive. Adeleke’s research, published in the Journal of Agriculture and Food Research (translated to English as “Journal of Agriculture and Food Research”), is bridging the gap between tradition and technology, offering a glimpse into the future of artisanal dairy systems.
Adeleke and his team have developed an integrated Internet of Things (IoT) and machine learning (ML) framework that promises to transform the way amasi is produced. The system, built on a Raspberry Pi-based platform, continuously collects data on pH, temperature, and electrical conductivity (EC) during fermentation. This data is then transmitted to a cloud-hosted digital twin, enabling real-time monitoring and control of the fermentation process.
The study found that electrical conductivity (EC) could be calibrated against total titratable acidity (TTA) using various machine learning models. Convolutional neural networks (CNN) achieved the highest global prediction accuracy, followed closely by feedforward neural networks (FNN) and Random Forests. The team also developed a time-to-target acidity model, which takes into account fermentation conditions and desired acidity levels, achieving an impressive R2 value of 0.98.
“The potential of combining IoT, ML, and automated control for low-cost, real-time acidity management in artisanal dairy systems is immense,” Adeleke explains. “This technology not only enhances the quality and consistency of amasi but also empowers small-scale producers with tools to optimize their production processes.”
The system’s ability to maintain optimal fermentation through PID-controlled actuation of heating and stirring elements is a game-changer. It ensures that the amasi produced meets the desired quality standards consistently, reducing waste and improving efficiency. The end-to-end pipeline was validated across seven fermentation runs, demonstrating high sensor consistency, scalable architecture, and practical deployment feasibility in resource-limited settings.
The implications of this research extend beyond the dairy industry. The integration of IoT and ML in food production processes opens up new avenues for precision agriculture and small-holder food innovation. It paves the way for smarter, more efficient, and sustainable food systems that can adapt to the changing needs of consumers and the environment.
As Adeleke puts it, “This work highlights the potential of combining IoT, ML, and automated control for low-cost, real-time acidity management in artisanal dairy systems, with broader implications for precision agriculture and small-holder food innovation.”
The study, published in the Journal of Agriculture and Food Research, is a testament to the power of technology in transforming traditional practices. It offers a glimpse into a future where technology and tradition coexist, creating a sustainable and efficient food system that benefits producers and consumers alike. As the world grapples with the challenges of feeding a growing population, such innovations are not just welcome but necessary. They represent a step towards a future where technology plays a pivotal role in ensuring food security and sustainability.