Food Pairing in Belgium launched the Digital Twin of a Consumer (DToC) technology.
This technology aims to improve food and beverage NPD by predicting “winning flavour combinations and product composition.
Business and information technology intelligence Foodpairing was founded 12 years ago. The original idea was to use science and data to help chefs create new and unique recipes based on previously “unheard of” combinations.
“During these 12 years, we have been able to build one of the largest flavour databases in the world,” Foodpairing founder and CEO Johan Langenbick told FoodNavigator.
Over time, the company found interest outside the restaurant world. Consumer goods companies, Langenbick noted, could also benefit from a better understanding of food preferences through AI.
“We realised that consumer data is critical for CPG companies to achieve the best possible results that’s why we started building this dataset. ”
Food pairing has always used digital twins, a technology that enables a virtual representation of a physical system or process in real-time. For food pairing, this meant developing digital twins of a food ingredient or product.
Now an AI developer is adding Digital Twin of the Consumer (DToC) technology to its toolbox. This means that companies can take advantage of the digital twins of food and consumers to develop new products.
“Adding DToC will help us improve the accuracy of our recommendations and provide clients with more solid information about winning product lines,” said Langenbick. “
Foodpairing’s vision is to effectively take 90% of new food products off the shelves and reduce development time from months to hours.
How does it do all this? To predict hedonic preference, or the second moment of truth (SMOT), food pairing creates a digital twin for each target consumer. Each digital twin contains like and dislike factors at a molecular level that can predict preferences for products or ingredients that the consumer has never tasted before, explained CEO Langenbick.
In the real world, you might do a consumer test or two, but with a digital double, we can search for the perfect tasting rooms for the products we like best. This search may turn up some very surprising flavour combinations. ” The results of the different digital twins are aggregated and ranked to present the “best” flavour concept for launch. The combination is further refined based on input from Rand and marketing from the client teams.
“Models not only have a taste concept, but they can also create a taste composition.”
Each concept is supported by: a tentative prediction (how interested the target group is in buying the product); an average preference (how much the target group likes the product); and the reasons for the success of the product.
Finding “Unlikely” Flavor Pairings depend on context and location, Langenbick explained. You can say that chocolate and chicken are surprising, but it’s a classic combination in Mexico. Dorade on a date? The Romans ate it.
Combinations previously developed by Foodpairing include strawberry chilli and chocolate drinks for convenience store chain 7-Eleven. Elsewhere, the company helped develop vanilla and coffee flavour pairings for the cheesemaker, as well as almond cookies with mozzarella, cured ham, and figs.
Other combinations include rum flavoured with tomato and plum, the CEO recalled.
Foodpairing’s DToC is currently for food manufacturers only, while its original digital twin technology is still available to chefs.
As for whether food manufacturers are hesitant to rely on artificial intelligence for flavour and product development, Langenbick argued that the opposite is increasingly true.
Over the years we’ve been in business, we’ve seen customers go from dismissal to respect, from confusion to acceptance, and now that we’ve embraced technology, we understand how AI can be used in taste. Experience is much more advanced today. “It was a few years ago,” he told the publication.
Source: Food navigator
