New weapons in the fight against organic waste pollution

This article was originally published on afvalonline.nl.

A combination of smart cameras on garbage trucks, container chip registration and artificial intelligence can help collectors to reduce the pollution in organic waste. Suppliers of the technology are further along than many people think.

The application of artificial intelligence in the waste sector will undoubtedly take off in the coming years.

Want to map out exactly which addresses on a collection route are responsible for the pollution in the organic waste without having collection employees go through all the containers? It is no longer future music. During a meeting organized by the Vang Huishoudelijk Afval (Vang Huishoudelijk Afval) program (Vang-HHA) yesterday (February 28) it appeared that various suppliers now have smart solutions to map the pollution of organic waste at the level of a municipality, a district or even an address.

Municipalities can use this information to carry out very targeted interventions. Campaigns can be specifically tailored to the needs of a neighborhood and residents can be approached individually to help them separate their waste better. And there is a lot of demand for it in the municipality, as witnessed by the large turnout.

The pollution in organic waste has therefore been increasing in the Netherlands for years. This is a major concern for processors, who have to incur extra costs to properly process the waste stream, but also for municipalities that are asked by both the government and the processors to focus more on the quality of collected streams and less on quantity.

Driver in control

During the Vang-HHA meeting, the stories of five companies showed that the tide can be turned with new technology. For example, the Dutch development agency Jama, which specializes in RFID solutions, already equips side loaders with cameras that take a picture of the contents of a container when it is emptied. When emptying, the driver checks the screen of his on-board computer to see if any contamination is visible in the photo. If so, he indicates this. The photo is linked to the address of the disposer of the waste. The collection vehicle recognizes the container via a chip. The photo can then be shared with the responsible resident with a neat letter: "Can we help you separate your waste better?" If the resident makes a mistake again (several times), a stricter letter can follow. Because all containers with pollution are registered, a picture of the pollution in the VGF is also created at district and municipal level. Jama makes it all transparent via a dashboard. The municipality can anticipate this with campaigns at district level, for example, or, for example, by deploying waste coaches who visit 'problem neighborhoods'.

According to Jama, drivers now often see it as "a sport" to detect pollution, but from the third quarter of this year the company will add artificial intelligence to its solution. Pollution is then automatically recognized.

Smart data

The French smart data company Lixo already offers solutions in which the pollution in organic waste is determined with artificial intelligence. The company, which is only three years old, claims to have turnkey solutions for different types of collection vehicles and for different waste streams. It is not only possible to search for pollution in VGF containers, but also in, for example, PMD containers. Lixo also promises that municipalities can use their technology (again a combination of camera, chip and artificial intelligence technologies) to map out where waste separation is lagging behind. Another interesting thing about Lixo's solution is that municipalities can gain insight into the most common separation errors. Important information for drawing up information campaigns. Lixo further stated that its system can be implemented in three weeks.

The Belgian Viumore, which also works closely with Renewi when it comes to smart data solutions, sees many applications for the use of smart cameras, whether or not in combination with the use of artificial intelligence. For example, on a collection route for industrial waste, smart cameras can register how much hard plastic is among the residual waste offered. This information can be used to determine whether it is interesting to collect hard plastic separately. Viumore called on those present to experiment with new technology. Although at the same time it also warned that artificial intelligence is not always the solution. It remains only one of the means to achieve goals.

Integration

AMCS, known in the sector for its route optimization software (some 700,000 collection vehicles are now on the road with it worldwide), also advocates the use of artificial intelligence in determining pollution in organic waste containers. Drivers still perform differently when they have to judge this. However, AMCS works with cameras that make short films to record the pollution.

If the system detects contamination in the organic waste, it takes a still from the video on which this is clearly visible. A photo of the pollution can be shared with the disposer. The AMCS Vision AI solution also makes it possible to score different types of pollution. This also creates a picture of how bad the pollution is. An advantage of the AMCS solution is of course that it can be integrated into the broader AMCS package that many collectors already work with.

Interested in learning more about AMCS Vision AI? Download the brochure below.

Alternative processing

Klikotronics, part of the Kliko Group, is working with the German C-trace on systems for mapping pollution in organic waste with artificial intelligence. They offer a solution in which a camera first looks into the container from above before it is emptied. In this way, a container with a lot of pollution does not end up in the trough, but can, for example, be emptied later by another collection vehicle.

In Germany, just across the border near Kleve, a test is already underway in which two vehicles drive behind each other. The first vehicle collects the clean organic waste, the second vehicle the contaminated organic waste, which can later be processed in a different way. In this way, the bill for the processing of the rejected organic waste as residual waste could be sent to the disposer. An incentive to separate better next time. The solution can be combined with detection during emptying of the container. There should be a 'market ready' system in the summer of 2023.

In the meantime, Klikotronics and C-trace hope to be able to test their solution in the Netherlands for at least three months. No doubt someone in the audience will take the plunge and contact the companies about this. Because one thing was clear after yesterday's presentations: there is plenty of interest in solutions to reduce the pollution of organic waste in the market.