Artificial Intelligence in Beer Brewing: A Comprehensive Review

Artificial intelligence is increasingly used in beer brewing, but the literature is spread across brewing technology, food quality, process engineering, and data science. This review synthesizes 188 publications on AI applications in beer brewing (176 primary studies and 12 reviews), identified through searches in seven bibliographic databases and manually screened for relevance against an explicitly stated criterion for what qualifies as an AI application. The studies were characterized by brewing phase, data modality, computational approach, and task and are mapped here onto brewing phase and the main computational approach used. The review shows that AI applications are not evenly distributed across the brewing process. Quality assurance and fermentation are the most studied areas, while bottling, production planning, and raw material assessment are less represented. Neural networks are especially common in fermentation modeling, where they are used for state estimation, prediction, control, and optimization. Computer vision is mainly applied to raw materials, bottling, foam, color, and product inspection. Electronic-nose and spectroscopic data, interpreted using chemometric and machine-learning models, support beer authentication, defect detection, sensory prediction, and quality classification. Industry 4.0 systems and digital twins provide infrastructure for process-wide data collection, planning, and sustainability assessment. Overall, AI in brewing is best understood as a set of phase-specific tools rather than one uniform technology. Wider industrial adoption will require validated datasets, robust sensors, interpretable models, and integration with routine brewery operations.