Guangyi Yang, Xiaohua Ma, Xinlong Chen, Qianxin Liu, Yongsheng Tao, Xuebing Bai
The integration of intelligent sensing, artificial intelligence (AI), and control technologies is reshaping traditional winemaking into a more data-driven and responsive process. This review proposes the concept of an Intelligent Oenological System (IOS) and critically reviews recent developments in smart winemaking technologies, with a focus on technologies involving Internet of Things (IoT) devices, sensor networks, and AI-based modeling. Key innovations include the deployment of multimodal sensors for real-time monitoring of temperature, pH, dissolved oxygen, spectral signals, and gas evolution across stages from grape maceration to wine aging. Advances in machine learning and digital twin models support predictive control of fermentation, whereas reinforcement learning and inverse design frameworks enable data-informed decision-making toward specific flavor targets. These technologies contribute to improved process precision, reduced resource consumption, and greater product consistency. Nevertheless, challenges, such as data interoperability, model interpretability, and system integration, continue to limit widespread adoption. Emerging solutions involving edge computing and standardized data ontologies are helping to address these issues. By aligning intelligent winemaking with sustainability, traceability, and personalized flavor goals, this review highlights its transformative potential for the future of the wine industry.