Tarapata Justyna, Murphy Tara R, Finnegan Eoin W, O'Callaghan Tom F, O'Mahony James A
School of Food and Nutritional Sciences, University College Cork, Cork, Ireland.
Dairy Processing Technology Centre, University College Cork, Cork, Ireland.
Compr Rev Food Sci Food Saf. 2025 Jul;24(4):e70209. doi: 10.1111/1541-4337.70209.
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.
乳制品热处理过程中的结垢仍然是一个重大挑战，它会降低运营效率、增加能源消耗并使清洗周期复杂化。本综述批判性地评估了乳制品行业热处理过程中测量和预测结垢的当前方法，重点强调科学原理、技术成熟度和工业适用性。与现有大多专注于结垢量化的综述不同，这项工作突出了向基于预测的结垢控制和最小化方法的转变。传统测量技术，如监测热阻和压降，与新兴方法，包括声学、光谱和电化学传感器一起进行了评估。从灵敏度、可扩展性和工业稳健性方面讨论了它们各自的局限性和优势。探索了先进的预测工具，包括深度学习、计算流体动力学和量纲分析技术，以评估它们对结垢动态特性进行建模并支持实时决策的能力。人工智能与实时过程数据采集的集成被视为改善结垢管理和优化清洗计划的关键创新。该综述还考虑了小规模实验系统在将实验室规模研究与工业应用联系起来方面的重要性。通过预测、预防和控制过程中的结垢来提高过程效率的工具的开发和利用正在不断发展。在这方面实现更大的控制为应对未来过程优化挑战、缩短就地清洗时间以及通过实时监测、预测分析和工业规模实施推进可持续乳制品制造提供了大量机会。应对这些挑战需要研究人员、工程师和行业利益相关者之间采取多学科方法，将新兴技术转化为实用、可扩展的解决方案。
