NEWS
According to the UN's Food Waste Index Report 2024, in 2022, the world wasted an estimated 1.05 billion tonnes of food. To put this into perspective, if this staggering amount were evenly distributed, each of the 8.1 billion people on Earth would receive nearly 130 kg of food.
These estimates include food wasted at retail, household, and foodservice levels. (In fact, even more food is lost in the part of the supply chain from field to manufacturing.) The foodservice sector alone is responsible for about 27% of it, or 290 million tonnes.
Food waste is a double-edged issue: for foodservice operators, it’s an additional monetary cost that erodes their profit margins. At the environmental level, its production generates greenhouse emissions without bringing any benefit. What’s more, when food waste is not properly disposed of and ends up in landfills without gas collection systems, it generates even more emissions, exacerbating the problem.
Classic food waste prevention measures and their limits
A 2021 study, published in the Journal of Foodservice Business Research, surveyed a small group of Swiss-German restaurant owners about their food waste prevention practices. The most common ones included reprocessing unused food, forecasting demand to reduce overproduction, and adjusting portion sizes.
However, surveyed operators also reported several challenges, such as accurately predicting the number of guests, lack of staff engagement, and difficulty finding chefs with the creativity to repurpose surplus food on the spot. Resistance from diners is also a barrier: many customers perceive unavailable menu items or spontaneous changes aimed at reducing waste, as poor service, rather than responsible waste-conscious practices.
Lower food costs with better forecasts
AI (Artificial Intelligence) is becoming a powerful tool in making food waste prevention measures more effective in foodservice. AI is an umbrella term that refers to the capability of computer systems to analyse data and solve problems in a way that resembles human intelligence. AI is divided into several subfields, including Machine Learning (ML), which uses models and algorithms to extract insights from large data sets. This is crucial for addressing a major challenge in food waste prevention: demand forecasting.
An accurate forecast for a restaurant would have to consider internal data, such as sale trends, performance of individual menu items, impact of price changes and promotions, and external factors, such as macroeconomic conditions, weather, local events, competitors' initiatives, public holidays, and strikes. The process would involve:
-
Tracking all relevant data regularly.
-
Understanding which factors have an impact on sales and to what degree.
-
Translating this analysis into actionable insights for decision-making.
With ML-powered Business Intelligence (BI) solutions for restaurants, it’s possible to automate this process saving time and increasing accuracy. Although these systems cannot factor in all possible variables, they can make baseline forecasts much more accurate, helping kitchen managers make data-based decisions. Additionally, better demand forecasts will optimise staffing levels, reducing the cost of overstaffing and preventing the negative impact of understaffing on service quality.
An example of an AI-based forecasting tool for foodservice business is Tenzo, a performance analytics platform for restaurants that integrates data from sales, labour, and inventory applications to improve operations.
At Nando's Singapore, for example, Tenzo improved the accuracy of sales forecast by 30%. Before Tenzo, Nando's Singapore used a static schedule that was simply divided into mid-week and weekend, with no real correlation to sales. With its ML-based tool, Tenzo provided a more accurate analysis of past sales data, which also led to a 15% increase in labour productivity.
Measuring food waste with image recognition
When you can measure a goal, you’re more likely to achieve it—and the same goes for reducing food waste. But the fast pace of commercial kitchens makes it nearly impossible to stop and log every item that ends up in the bin, along with its weight and cost. That’s where AI, and more specifically image recognition, can help.
AI-driven food waste management systems can track and weigh discarded food while calculating its value. These setups typically include a bin placed on a scale, with a camera equipped with image recognition technology positioned above it. After a period of training, the system learns to identify and weigh individual food items independently and multiply their quantity by the associated cost. All that is translated into reports that provide foodservice businesses with precise data on how food waste is affecting their bottom line and actionable insights to adjust forecasting and purchasing strategies.
Reducing food waste with AI: 3 success stories
Winnow, Leanpath, and Lumitics are examples of companies that have developed these AI-based solutions. Here are three examples of how they helped different types of commercial kitchens.
- Using Winnow, the Canary Wharf branch of Marriott International achieved a 67% reduction in food waste within just 6 months of installation. For example, based on waste patterns uncovered by the analytics platform, kitchen management started to use smaller containers after 10:00 am during their buffet breakfast, reducing food waste significantly.
- Google has deployed Leanpath's system in nearly 200 cafes in 21 countries, saving 4 million pounds of food waste in ten years. As Google's Global Procurement Manager Kristen Rainey says, “I think it's easy for every one of us in foodservice to think that we're on top of our food waste, but we actually need that data to help us really hone in on the areas where we can reduce it.”
- The largest-scale achievment so far, however, has been IKEA’s main retailer, Ingka Group. Using Winnow’s system, in 2021, Ingka Group reduced food waste by 54% across almost 400 IKEA store restaurants, compared to 2017. The amount of saved food is equal to 20 million meals, 36,000 tonnes of CO2 emissions, and 37 million Euros.
These success stories highlight three key points:
- The UN's Sustainable Development Goal 12.3, which aims to halve global food waste at the retail and consumer levels by 2030, is ambitious but attainable.
- While technology is essential, it needs people to be truly effective. Ingka Group’s Chief Sustainability Officer emphasised how the engagement of food coworkers was instrumental in this achievement.
- To secure employee engagement, leadership must take the first step, setting goals and holding themselves accountable, creating a clear escalation path and setting the tone for success.
PUBBLICAZIONE
08/10/2024