FOOD WASTE
Full plates, not full trash cans!
A lot of food ends up in the trash, not just in the home, but on its way to the consumer. AI prevents food waste during production and in the supply chain.
Imagine your husband surprises you in the morning by saying, “I’m cooking us something nice tonight!” You fear the worst! After all, your husband has shown so little talent in the kitchen that you can’t even trust him not to burn the pasta water! In the evening, however, you are surprised when your husband serves you an almost perfect vegetable risotto. Where does your husband’s sudden talent come from? He relies on technological support, a kind of AI for the home kitchen – the Thermomix. This appliance not only tells you exactly which ingredients to put in the pot, in what quantities, and in what order, but it also cooks everything at the right temperature for the right amount of time. This way, even the uninitiated can cook tasty meals, with a not-insignificant side effect: you don’t have to throw out food because it’s overcooked or burnt. What is now making its way into the home kitchen has long been common practice in the food industry. Manufacturers are using artificial intelligence both to create new products and to organize the supply chain. The main goal is to avoid food waste.
Sweet or salty? AI determines the ideal mix
Before creating a new food product, manufacturers determine the desired characteristics and composition of the product. What concentration of nutrients should it have? Should it taste sweet, sour, or salty? What consistency should it have? Often, many manufacturing processes with different compositions of individual ingredients are required to achieve the desired end result. This means mixing, tasting, discarding, and remixing. This process is repeated until the product has the desired characteristics. In this way, manufacturers produce a large amount of unusable excess that ends up in the trash.
With AI, this food waste can be eliminated. This is because the properties of an ingredient, such as taste, consistency, or sugar content, can be described by mathematical formulas. These formulas, in turn, form the basis of an optimization model. This model contains all the properties of all the ingredients needed. An algorithm then calculates the best mix ratio to give the final product the desired characteristics. Cereal manufacturers, for example, choose from up to 50 different ingredients when creating new mixes. Using AI, they determine which combination of ingredients will produce the right taste for a new product. This means that unsuitable combinations do not end up in the waste bin after production but can be eliminated from the outset.
VIDEO: AI IN EVERDAY LIFE
Food waste
In this video, our colleague Tyrone Castelanelli, AI Catalyst at INFORM, explains how AI is helping food manufacturers create new products. This includes the production of Cuvée, which has already begun so that you can make a proper toast on New Year’s Eve.
Ripe or not? AI determines harvest time in agriculture
The quality of the raw materials is crucial for the taste and success of a product. For this reason, agriculture is also using AI to improve its production processes and thus prevent low-quality goods from being produced and thrown in the trash. For example, a data-driven early warning system detects crop infection risks in a timely manner. To do this, wireless sensors measure many agricultural parameters such as temperature and humidity, which are crucial for plant growth. But the plants’ age, disease history, and current weather are also included in the fully automated measurement and analysis process. Predicting disease risk is thus becoming more accurate than ever before.
Farms are also increasingly using image-based AI technologies as part of their field management. To this end, they have drones flying over their cultivated areas. The areas and their plants are scanned from the air, a vast amount of data is analyzed directly, and cultivation is monitored extremely reliably in real-time. For example, it is possible to determine how ripe grains, fruits, and vegetables are in order to prepare for harvesting at the right time. The condition of the soil can also be precisely determined using image-based AI technology to alert for the need for water, fertilization, and pesticides.
Worldwide, 1.3 billion tons of edible food currently end up in the trash every year. Germany alone accounts for 1.8 million tons of this per year. The way food is handled along the entire food supply chain is responsible for this.
How much and how long? AI determines optimal conditions along the supply chain
However, preventing food waste is not only important in the manufacturing process. Worldwide, 1.3 billion tons of edible food currently end up in the trash every year. Germany alone accounts for 1.8 million tons of this per year, which corresponds to one-third of current German food consumption. Responsible for this is the way food is handled along the entire food supply chain. Here, too, AI can be applied in a variety of ways to prevent food waste.
For example, AI can be used to make precise predictions of fluctuations in demand. Using such a sales forecast, manufacturers then plan their manufacturing processes based on expected consumer behavior. This means that they can produce a product in the right quantity depending on the forecast sales. This not only avoids overproduction. Manufacturers also optimize all upstream procurement processes on the basis of the sales forecast because they purchase all ingredients only in the appropriate quantity. This means that oversized stocks of raw materials no longer have to spoil
Thanks to optimized sales forecasting based on AI, manufacturers plan all processes from procurement to production so that no raw materials spoil and end up in the trash.
In the supply chain, AI is also used to calculate the shelf life of fresh foods such as fruits and vegetables. Based on parameters such as water, sugar, or acid content, AI determines the shelf life of a food product and prevents already spoiled goods from reaching supermarkets, where they are then declared unsaleable and thrown away. The same applies to the perishability of fresh meat. Using data on hygiene, food safety, and quality, the AI calculates an estimated date by which a product will be spoiled. This way, it can be brought to market and sold quickly.
FUN FACT
AI develops chocolate
What do AI and chocolate have in common? You should enjoy both with common sense. Then both chocolate and AI will sweeten your life. AI is now also developing chocolate. A Finnish food company has analyzed over one million taste preferences using AI. The goal: a chocolate that tastes good to everyone, no matter which flavor they like best.
What am I cooking today? AI develops recipes
AI is helping at many crucial points to provide consumers with ideal products without unnecessarily wasting food. One question, however, remains: “What am I cooking today?” Your husband would probably ask the Thermomix after the success of the vegetable risotto. And surely he would find prepared recipes there. But who knows how they would taste? It would be better if you could specify your desired taste, and the device creates an individual recipe based on this information. The Thermomix is not yet capable of doing this. Until it is equipped with the AI required for this, you will either have to resign yourself to selecting your device or open a recipe book in the classic way. Or you can ask a tool like ChatGPT to create a recipe for you according to your taste preferences.
Bon appétit!