Since 2013, its San Francisco-based start-up, Compology, has been using cameras and artificial intelligence to control what is thrown in trash cans and trash cans at companies like McDonald’s restaurants and department stores. Nordstrom. The issue is to make sure that the rubbish bins are full before emptying them and to prevent recyclable materials such as cardboard from being contaminated with other waste, so that they do not become waste either.
To help them do it right, Compology puts garbage control cameras and sensors in industrial waste containers. The cameras take pictures several times a day and when the container is lifted to dump it. An accelerometer helps to activate the camera on garbage day.
Artificial intelligence software analyzes images to find out how full the container is and can also let the customer know when there is something where it should not be, such as a garbage bag thrown in a container full of cardboard boxes for recycling. Gates said the company’s cameras can reduce the amount of non-recyclable materials thrown in garbage containers up to 80%.
With McDonald’s container containers in Las Vegas, for example, Compology cameras and sensors showed that the company did a good job recycling cardboard packaging. But from time to time garbage bags were also thrown in there, Gates said.
“Once we saw that the garbage bags were inside the cardboard containers, we sent a notification to the local people via a text message, letting them know that they should take it out before the truck comes. the next morning and telling them that putting rubbish in when recycling bins is a form of pollution, something they shouldn’t do in the future, ”he said.
Brent Bohn, owner and manager of dozens of McDonald’s restaurants in the Las Vegas and Phoenix areas, uses Compology in Las Vegas restaurants to help make sure the restaurant workers recycle properly.
“The cameras have really streamlined it and provided responsibilities for us, but also for our suppliers and the carriers we work with,” he said.
Compology trained its system to sort garbage with tens of millions of images, Gates said, and uses images taken from garbage containers that are now used to better determine fullness and what’s inside. To date, it has processed more than 80 million images of the 162,000 cameras it has installed.
“The more images we get from inside the containers, the more accurate we can be,” he said.
The Compology service costs companies between $ 10 and $ 20 a month per container. It usually saves them thousands of dollars per container of annual waste transportation costs, Gates said, as it can also use AI to predict what the service schedule for each container should be, so it will only be collected when it is likely to be full.
Over time, Gates hopes that Compology can help standardize the way waste is measured and reported, which is currently inconsistent in the U.S.
“You’ve been able to measure how much electricity, water and gas you’ve used for decades,” Gates said. “What we’re doing is being able to measure how much waste you produce.”