Food is too valuable to waste.
But nearly $100 billion of it is thrown away in the hospitality sector every year.
When you’re catering for an unknown number of guests, you can’t afford to be underprepared. In many cases, this can lead kitchen staff to the other extreme — preparing too many meals. All of the extra, unused ingredients ultimately end up in the bin.
Winnow, a U.K.-based company, is using AI to take a bite out of food waste by empowering commercial kitchens to reduce the amount of food they dump.
AI for Reducing Food Waste
Around one-third of the food produced globally for human consumption is wasted every year. That amounts to a staggering 1.3 billion tonnes.
Winnow is helping professional chefs curb those numbers with its latest product, Winnow Vision, which automatically detects, identifies and measures food at the point it is thrown out.
The system involves a set of digital weighing scales on top of which sits a standard kitchen bin. Mounted above this is a camera and compute system containing an NVIDIA Jetson TX2 supercomputer on a module.
The module takes the images captured by the camera, as well as the weight recorded by the scales, and determines what is being thrown out and in what quantity. The neural networks used by the Jetson TX2 are trained using AWS instances with NVIDIA V100 GPUs on TensorFlow. To identify the wide variety of food the system may encounter, a huge amount of training data is needed — up to 1,000 images per food item.
The collected data is sent to the cloud for processing and regular reports are then created and shared with kitchen staff. The reports detail quantities and types of food being tossed, as well as recommendations as to how the kitchen can reduce waste.
Winnow co-founder and CEO Marc Zornes explains why the real-time deep learning results the Jetson TX2 delivers onsite — what’s known as “inference at the edge” — are key.
“It’s really important to us that the customer receives immediate results, in an environment that cannot guarantee a reliable and fast internet connection,” said Zornes. “Using the Jetson TX2 devices in the field enables us to provide, in real time, a ‘better than human’ understanding of what is being thrown into the bin on the edge, live, in the kitchen.”
The Jetson TX2 module can run multiple processes. Having a complete system on the edge means the Winnow team can reuse knowledge gained from working in the cloud and apply it to an edge paradigm. The Jetson platform is powerful enough to encompass current and future workloads, and flexible enough for Winnow to experiment and design new solutions.
Winnow Vision has already surpassed human levels with an accuracy rate of over 80 percent when identifying food that has ended up in the trash. This will increase with time as more and more data is collected.
The system is already installed in over 75 kitchens and Winnow plans to roll out the technology to thousands more in the coming years. IKEA and Emaar are among the companies that have implemented Winnow Vision in their kitchens.
Reducing the amount of food waste isn’t the only benefit for businesses. Automating the process increases efficiency in the kitchen, too. Staff require less training on food management and need to spend less time adjusting their menus.
Winnow has shown that by arming teams with analytics, food waste can be cut in half. The company estimates it has already helped commercial kitchens save more than $30 million in annualized food costs. That equates to preventing over 23 million meals from going in the trash.
With the advent of its new technology, Winnow has announced that it aims to save kitchens $1 billion by 2025.