Big data applications in food supply chains

Food supply chains are characterized by innovation, not only in products but also in processes. This paper aims to identify big data applications in the food and drink sector and present its findings as a state-of-the-art literature review. Academic databases were searched using ‘food’ or ‘drink’ and ‘big data’ keywords. Scholarly publications from 2015 onward are identified and presented in broad categories of demand prediction and retail operations optimization. The review recognized big data applications as a great opportunity for food supply chains. The applications aimed 1) to understand the customer base and inform marketing communications strategy, 2) to predict demand and organize retail operations to meet this demand, and 3) to optimize prices, assortment, and inventories based on demand patterns. Applications in this review focused more on descriptive and predictive analytics than prescriptive analytics, possibly due to the emergent nature of these applications. Descriptive analytics applications focused on capturing data, summarizing the status quo, and developing customer segments which can then be managed using varying marketing strategies. Predictive analytics applications focused on demand prediction with novel approaches proposed by the machine learning community. Prescriptive analytics applications aimed at promotion optimization and pricing for profit maximization. Cognitive analytics applications extracted customer reviews from online stores to inform which products should be marketed in what way. The review offers managerial insights on circumstances where big data analytics could prove beneficial. Managerial implications suggest that data integrators enable big data applications by ensuring the data collected are accurate, timely, and complete to inform descriptive, predictive, and prescriptive analytical models.