Reducing Waste in the Foodservice Industry
Food waste is a major global issue. Food in landfill is both catastrophic for the environment and, given high levels of world hunger, a major moral issue. Now, the development of an exciting smart fridge prototype shows how AI could drive innovative solutions to sustainability issues.
When we think about food in the context of sustainability, it’s important to distinguish between food loss and food waste. Food loss refers to food that is lost along the supply chain – from harvest to processing – before it becomes available for sale. This can be due to a number of reasons, including extreme weather events and crops lost to pests. By contrast, wasted food is generally defined as that which makes it to the shelf, but is thrown away before it is eaten. Two main culprits here are ‘plate waste’ and food that spoils before being eaten.
Around 1.3 billion tons of food are lost to food loss and waste (FLW) each year, at a cost of about $1 trillion. FLW has a negative environmental impact, too, responsible for roughly 8% of global greenhouse gas emissions. And it contributes to hunger, which the EPA estimates affects almost a billion people worldwide.
According to the World Food Program US, it’s estimated that close to a third of food in high-income countries is wasted, while roughly 40% of food in low-income countries is lost.
Winning Team 'Copenhagen Collective' at the SICK Solution Hackathon | Image ©Andreas Kusy Photography
Now, the new smart fridge prototype serves as an exciting use case for AI driving greater sustainability – and we were privileged enough to see it first-hand when it was conceived at last year’s SICK Solution Hackathon in Germany. We were proud to act as a partner of the event, sending seven coaches to assist five different teams.
One of the teams we supported, ‘Copenhagen Collective’, came up with an award-winning concept that could help to reduce food waste in restaurant and foodservice businesses.
The Copenhagen Collective identified the problem that 10% of restaurant food is wasted without ever having been served on a plate. To combat this, the team focused on two main categories, spoiled food and misplaced food, and proposed a system that would tackle both.
Input for the AI Nose comes from two different sensors. A gas sensor combined with machine learning detects changes in food when it starts to spoil, while the optical sensor watches out for misplaced food in order to avoid problems like cross-contamination. During the Hackathon, the team designed the model to detect changes in chicken meat as it spoiled.
In the next step, data from the sensors is collected and transferred to the cloud in real time using SICK LiveConnect and Amazon Web Services or Microsoft Azure solutions (like Azure IoT Hub).
The training data and machine learning algorithms that train the AI are located in the cloud to form a prediction-making model that can interpret the sensor data.
The predictions generated by the model – for example, that an item of food will spoil soon – are distributed to the user via the application. This occurs either via a dashboard or, in urgent cases, as a push or SMS notification.
This and similar concepts could have significant real-world benefits for restaurants, cafeterias, and the foodservice trade more generally: Anywhere that big kitchens play a major role. They have the potential to bring less waste, improved safety and better legislative compliance.
AI gathers and applies incredible amounts of information to increase productivity and solve profound challenges facing the world.
Along the way, it’s crucial that AI is itself sustainable, given how much energy it utilizes. Global data centers make up more than 2% of all carbon emissions. This figure could be reduced with improved AI processes, including potentially lowering the amount of data involved in AI training.
Notwithstanding this challenge, the power of AI can be harnessed in a variety of ways to play a fundamental role in sustainable, innovative development.
AI is expected to drive growth, with PwC research indicating it could contribute $15.7tr to the global economy and boost GDP in local economies by up to 26% by 2030. At an individual consumer level, AI could drive a range of product enhancements resulting in greater variety and personalization.
As this progress occurs, it will be vital to ensure that these economic benefits are equally distributed. If not, AI also has the potential to increase the gap between the haves and the have-nots of the world.
PwC estimates that AI for environmental applications could contribute up to $5.2tr to the global economy by 2030 thanks to AI-infused clean distributed energy grids, more sustainable supply chains, and better weather and environmental monitoring.
We are already seeing the utility value of AI in major environmental projects. EarthScan provides analysis on the likelihood of climate-based shocks, while ClimateView created its ClimateOS platform to help build emissions reductions into city planning. Agriculture is another significant area of potential, with machine learning and farming digitalization having the capability to streamline processes.
An analysis of goals from the 2030 Agenda for Sustainable Development indicates that AI is likely to have both positive and negative impacts. AI could support circular economies and smart cities while increasing productivity and fighting against climate change. At the same time, caution is needed, as AI could easily be used to over-exploit ecosystems through resource extraction.
As AI is increasingly utilized on a range of ambitious projects for people and the environment, it seems clear that the technology will be one of the keys to unlocking a sustainable future. One initiative we’re proud to be part of is smartFoodTechnologyOWL, whose goal is to reduce waste and increase the supply of high quality food by bringing digital transformation to the food industry.
The OWL initiative and smart fridge prototype give us some of the first exciting clues as to how this technology could move things forward.
NTT DATA Business Solutions
Mar 25, 2024
NTT DATA Business Solutions
Feb 12, 2024
NTT DATA Business Solutions
Sep 26, 2023
Tina Dyhr Dupont
Jul 06, 2023
NTT DATA Business Solutions
Jun 15, 2023
NTT DATA Business Solutions
Jun 01, 2023
