Role and Relevance of Data Analytics in Beverage and Brewing Industry

ABSTRACT The digitalization of alcoholic‐beverage production has generated data‐rich environments, but the indexed literature does not clearly distinguish prediction‐oriented analytics from methods designed for continuous statistical process surveillance. This review examines that distinction in brewing while avoiding inferences about undocumented industrial adoption. The primary objective of this study is to map the academic literature on data analytics in alcoholic‐beverage production and, from a quality‐engineering perspective, to distinguish product‐characterization, prediction/optimization, and process‐surveillance roles within brewing research. To address this objective, we conducted a bibliometric analysis of 5691 journal articles and reviews indexed in the Web of Science for 1936–2025, obtained after screening 5722 records. The Web of Science was selected for its standardized citation metadata and export structure, and the bibliometric analysis was complemented by descriptive analyses in R. The results reveal a markedly uneven research landscape: wine‐related studies account for 52% of the corpus, whereas brewing represents a smaller but still substantial subset of 947 publications. Within brewing, five main thematic areas are identified: fermentation monitoring and process dynamics, product characterization and quality differentiation, microbiological stability and spoilage detection, predictive modelling and process optimization, and ingredient innovation and formulation development. Although chemometrics, spectroscopy, and machine‐learning approaches have expanded rapidly, explicit SPC/MSPC‐oriented studies remain comparatively less visible in the retrieved brewing literature. This finding concerns publication visibility and does not establish the level of industrial adoption, which may be obscured by proprietary practice or alternative terminology. The bibliometric evidence is interpreted as evidence of academic publication visibility, not as a direct measure of industrial adoption. A fully synthetic MSPC–ML example is included solely as a reproducible proof of concept illustrating the complementary roles of process monitoring and quality prediction; it does not constitute experimental or industrial validation. The contribution is a critical mapping of the academic brewing‐analytics literature that separates characterization and prediction from continuous process surveillance, identifies where SPC/MSPC terminology is and is not visible, and defines methodological priorities for interpretable monitoring research.