- AI helps a restaurant chain to create the perfect pizza
- Food companies turn to machines to invent new flavours
- AI-made whiskey will hit the market soon
- Tracking the taste preferences of various demographics
- Creating a smart food ecosystem by making recipes ‘shoppable’
- Algorithms cater to the needs of wine lovers
- AI could be vital to a healthy life and tasty menus
The future looks bright for the food and beverages industry. In fact, the global food and beverages market will reach nearly $9.5 trillion by 2022, and its growth can be mostly attributed to an “increasing population, strong economic growth and rising disposable income in many emerging countries”. In the US, for instance, food processing accounts for 16 per cent of the shipment value from all manufacturing plants, employing around 1.5 million workers. But despite the high production volume, this industry remains a low margin business. Competition is tough and a small increase or decrease in production and sales efficiencies can make all the difference between reporting profit or loss. That’s why companies are turning to advanced technologies such as artificial intelligence (AI) to gain a competitive edge.
And the various uses of AI show that the food industry is ripe for tech disruption. Some companies, for instance, use smart algorithms to create the perfect pizza, while others rely on a machine to suggest flavourful recipes. AI has also found use in the production of whiskey, and it can help people select wine that suits their taste. And once they collect enough data, algorithms could even help companies to develop highly addictive and delicious tastes suited for particular demographics. Clearly, AI has a lot to offer to the food and beverages industry, and we’re just at the beginning of that journey.
AI helps a restaurant chain to create the perfect pizza
The American pizza restaurant chain Domino's Pizza uses artificial intelligence in its quality control process. Its scanning device, called the DOM Pizza Checker, is placed above the cutting bench, and it analyses each piece of food before it’s given to the delivery person. For instance, the software flags pizzas with the wrong toppings or shape and alerts the staff to make a new one. Domino's Australia CEO, Nick Knight, explains that the program does this by “capturing an image of the pizza and using artificial intelligence to compare this data with a large dataset of correct pizzas, making a quick assessment”. DOM Pizza Checker is now in use in the company’s stores in Australia and New Zealand.
Food companies turn to machines to invent new flavours
Smart algorithms can also speed up and add value to the development of new spices and flavours. Businesses such as McCormick & Company, an American food company, take months and even years to develop new seasoning, and the process starts with first deciding on the basic criteria, such as whether the product is to be halal or kosher and who the target audience is. Once that’s clear, product developers test various combinations of ingredients. They carefully measure the amounts added, as even the smallest change can break a flavour, and it can take up to 150 iterations and consultations with dozens of experts such as chefs and chemists scattered around the world to decide whether the product is shelf-ready. But speed is of the essence in the highly competitive food industry, and that’s where technology can help.
IBM and McCormick & Company used data such as historical flavour formulas, raw material components, consumer test results, and sales figures to create a smart algorithm that helps food developers. The AI tool can, for instance, predict alternative materials for seasoning formulas, appropriate ratios of raw ingredients, human response, and the level of novelty of system-generated flavour formulas. And the initial tests look promising. IBM’s new AI helped experts to create new flavours in less than 50 iterations and reduce the amount of work by around 70 per cent.
The system also boosts the creativity of food developers. For instance, instead of teams using well-known formulas for certain flavours, the algorithm can now quickly explore more options and suggest alternatives to overused ingredients. Hamed Faridi, the chief science officer at McCormick & Company says that other food industry giants will turn to AI, too, and that “20 years down the road, this will be the system. There is no other way of making new products”.
AI-made whiskey will hit the market soon
Technology could also change the way companies make whiskey. Currently, the process involves liquor spending time in charred wooden casks, which results in a unique taste and a dark colour. An important consideration is also the type of drink that was held in the casks before, as traces of bourbon, wine, and sherry all lead to a different flavour of whiskey. Master distillers typically use all these variables to create a unique blend, but AI can make that process much faster.
The tech giant Microsoft, the Finnish software firm Fourkind, and the Sweden-based distillery Mackmyra Whisky have teamed up to develop a machine learning program that decides which whiskey flavour should be made next. The tool, which will produce the world’s first AI-based whiskey, makes that prediction by analysing existing recipes, sales figures, and customer preferences. It’s capable of generating more than 70 million recipes, and it suggests those with the highest chance of impressing customers based on currently available cask types. But the final decision on which recipe to use will still be made by humans. The first AI-generated whiskey is set to hit the market in autumn of 2019, while Fourkind plans to use its algorithm for other beverages, perfumes, sweets, and sneaker designs as well.
Tracking the taste preferences of various demographics
The tech firm Analytical Flavor Systems (AFS) went one step further and has built an AI platform called Gastrograph that helps food and drink producers optimise the flavour, aroma, and texture of their products for target consumers. The system relies on thousands of people from all over the world that report their impressions on how a new food or drink tastes. They can select specific descriptors, and the AI uses this data to develop taste profiles of products and learn more about the preferences of certain demographics.
The brewing company Yards Brewing, for instance, uses Gastrograph to gather feedback from people on how they perceive the taste of its beer. And since the software knows how the drink should taste, it alerts the company if there are anomalies in customer responses. Frank Winslow, Yards Brewing’s director of quality control, says that “Having those kinds of warnings early in the process is a huge step.'' AI can also help businesses that develop new products. They can use the data on the preferred tastes of young people, or those of other age groups, to improve the aroma and maximise the chances of success for new food items. And Asian companies are especially keen on using tools such as Gastrograph in product development.
Creating a smart food ecosystem by making recipes ‘shoppable’
Innovations in the food industry now extend to online recipes as well. The UK tech startup Whisk runs an AI platform that provides several benefits to enterprise users, including publishers, retailers, and IoT manufacturers. The first thing it does is make recipes ‘shoppable’ by matching ingredients with products at grocery stores. For instance, ‘50g of butter, cubed’ mentioned in the recipe could be matched with ‘250g Tesco Salted Butter’. The software then interacts with the store to fill the user’s basket, and retail industry giants such as Amazon, Walmart, Instacart, and Tesco are all connected with the platform.
The second benefit Whisk provides is that it analyses user preferences such as disliked ingredients, allergies, diet, and previous behaviour to create personalised meal plans and recipe feeds. And the third feature of this AI tool is that it connects IoT devices with the user’s journey. For instance, it provides the infrastructure needed to connect smart fridges with online recipes, so that users know which ingredients they already have and which ones they need to order. Samsung is using Whisk to develop a number of smart food apps that integrate personalised recipes with smart devices and retailers.
Algorithms cater to the needs of wine lovers
The software firm Tastry also uses AI to offer personalised services, but instead of recipes, it focuses on wine recommendation. The system works by first chemically analysing wine products and integrating the data into an AI engine. Then, consumers take a short survey on a mobile app, helping the system find the matching taste. And this concept has various applications, as it can, for instance, power in-store kiosks that help shoppers decide which wine to buy.
Retailers report a five per cent increase in gross sales after using Tastry. Manufacturers can also benefit, as this AI tool can help them to improve their wine before it hits the market and predict which of their current products have the most chance of succeeding. The company’s founder, Katerina Axelsson, has even more ambitious plans. She wants to improve the algorithm so that it can pair the wine to suitable recipes and to the food inventory in a store. Also, shoppers could get wine recommendation based on their shopping list, while manufacturers could use AI to develop wine tastes tailored to specific geographic regions.
AI could be vital to a healthy life and tasty menus
Artificial intelligence is changing the way we develop, shop, and prepare various food and drink products. It provides new flavours and tastes that companies couldn’t even think of, while users benefit from personalised recipes and healthy food. But this is just the tip of the iceberg of what advanced technologies could offer. For example, AI could one day pull in various types of data such as levels of stress, glucose, and medication, and use it to act as our virtual health coach and create customised dietary recommendations. The potential for innovation is impressive, and it remains to be seen how far companies can go in using smart algorithms to transform the food and beverages industry.