Abstract
Fouling during the thermal processing of dairy products remains a significant challenge, reducing operational efficiency, increasing energy consumption, and complicating cleaning cycles. This review critically assesses current methods for measuring and predicting fouling during thermal processing in the dairy industry, emphasizing scientific principles, technical maturity, and industrial applicability. Unlike existing reviews, which are mostly focused on fouling quantification, this work highlights the shift toward prediction-driven approaches for fouling control and minimization. Traditional measurement techniques, such as monitoring thermal resistance and pressure drop, are evaluated alongside emerging methods, including acoustic, spectroscopic, and electrochemical sensors. Their respective limitations and strengths are discussed in terms of sensitivity, scalability, and industrial robustness. Advanced predictive tools, including deep learning, computational fluid dynamics, and dimensional analysis techniques, are explored for their ability to model the dynamic nature of fouling and support real-time decision-making. The integration of artificial intelligence with real-time process data acquisition is identified as a key innovation for improving fouling management and optimizing cleaning schedules. The review also considers the importance of small-scale experimental systems in linking laboratory-scale research with industrial applications. Development and utilization of tools for enhanced process efficiency through prediction, prevention, and control of in-process fouling are growing. Greater control in this regard offers substantial opportunity to meet future challenges in process optimization, shorten cleaning-in-place times, and advance sustainable dairy manufacturing through real-time monitoring, predictive analytics, and industrial-scale implementation. Addressing these challenges will require a multidisciplinary approach between researchers, engineers, and industry stakeholders to translate emerging technologies into practical, scalable solutions.