Perspectives from ISB

From Good Mornings to Great Evenings – AI in your Beverage Cup

A good morning usually begins with Tea or Coffee. Great evenings end (for some) with a beverage like, Wine, Whiskey or Beer. Artificial Intelligence may now help you get the beverage of your choice- just the way you like it! With customer engagement and intelligent automation, AI is making headway in Empathetic AI.  

In the beverage industry, Data is collected from sensors, warehouses and retailers. Smart systems and AI are deployed for meaningful use of this data. Machines are now able to make decisions using appropriate algorithms to identify data patterns and match them to consumer behaviour. 

The benefits of AI and big data are well known and understood in Crop Production and Protection, Production, Supply Chain Management & Logistics, and Retail and Consumer experience.

AI stands out in product development enabled by the digitization of flavor and taste in beverages. AI has helped the food and beverage industry to discover & develop new flavors. In 2018, Danish brewer, Carlsburg used AI to map and predict flavors from the yeast and other ingredients in beer. IBM developed an AI for McCormick to create better spices. And NotCo, which produces vegan NotMilks, uses AI to analyze molecular structures and find new combinations of plant-based ingredients.

A Colombian start-up, Demetria, is betting on its digital sensory fingerprint using AI. It has matched a coffee bean’s profile to the industry’s standard coffee flavor wheel, created in 1995. Roasters and producers of coffee will be able to assess quality and taste at any stage of the coffee production process with the minting of this new digital sensory fingerprint. Demetria uses handheld near-infra-red (NIR) sensors to read the spectral fingerprint of green coffee beans. The different colours and wavelengths of the light spectrum react differently to each organic compound present in the coffee, representing the whole chemical composition of the beans. AI then translates the NIR data into the sensory language the industry understands. Historically, the taste or sensory quality of coffee beans has been determined by cupping- a manual, exhaustive process carried out by certified tasting experts. Demetria superimposes data generated by NIR readings on to Cupping data. It then calibrates the AI to match a specific spectral fingerprint to an unmistakable taste profile.

Tastry, a sensory sciences company based in California, uses machine learning and advanced chemistry to teach artificial intelligence to “taste.” This technology seems to be game-changing in the wine industry. It provides valuable information to winemakers and retailers about the wines that customers enjoy. Tastry’s technology decodes aroma and flavor profiles by chemically analyzing thousands of wines in their lab. Further, the technology identifies the unique consumer palate of each individual that takes a quiz in a store or on their app. Consumer palates are just like fingerprints – no two are alike. The AI then uses these datasets (consumer palate data and wine profiles) to predict consumer product scores with over 92% accuracy. Tastry’s app, BottleBird, walks users through a quiz called a “palate survey” by asking about their taste and smell preferences – including questions about things like coffee, tobacco, and vinegar. Then BottleBird uses AI and the science of your palate to create personalized wine recommendations. This approach makes Tastry different from apps like Vivino, which rely on historical purchase and rating data.

Angela D’Orazio is the master blender and chief nose officer at Mackmyra – a multi-award-winning Swedish whisky distillery. The distillery is the producer of the world’s first AI-created whisky, aptly named- Intelligens. The distillery uses Machine Learning Models that are powered by Microsoft Azure Cloud platform and AI cognitive services developed by Fourkind. Mackmyra’s existing recipes, sales data, and customer preferences are run on ML models. The data set thus created can generate upto 70 million recipes that the AI predicts will be of good quality and popular with the patrons.

While the AI provided the ingredient suggestions for Intelligens, the all-important choice of the storage cask, in which the whisky would age and gain its flavor, was carefully selected by D’Orazio herself.  Using an AI-based algorithm to design the recipe is faster than a person carrying out the process manually. Innovative combinations have emerged due to the algorithm’s ability to sift through vast data.

Artificial Intelligence in Food & Beverage market was valued at USD 3.07 billion in 2020 and is expected to reach USD 29.94 billion by 2026 at a CAGR of over 45.77% during the forecast period, (2021-2026). 

The enormous expected growth will bring opportunities in a big way. AI can help the industry in logistics, predictive analytics, transparency and, above all, create beverages with desirable sensory qualities that consumers can enjoy. 


Leave a Message

Registration isn't required.

By commenting you accept the Privacy Policy