If you want to know more about AI and ML, then my last blog will inform you about them.
AMCS Vision AI leverages both Computer Vision and Artificial Intelligence to provide you with automated and granular insights into the composition of the material that you are collecting.
In the first use case for AMCS Vision AI, we have developed an application called Automated Material Identification Service, to identify contamination at the point of collection.
Contamination is a problem for many companies that collect material such as organic waste and recyclables. If for example, a customer is consistently putting plastics in a paper and cardboard bin, then it requires a further sort once delivered to the depot or materials recycling facility (MRF).
A report from The Recycling Partnership estimated that contamination represents on average 17% by weight in curb side containers in the United States.
A 2017 report from the Scottish Environmental Protection Agency also found that on average 17% of material sent to Scottish MRFs (Material Recovery Facilities) for sorting was contamination.
This adds cost to the bottom line because the contaminated material needs to be sorted and separated from valuable secondary commodities. In some cases, the recycled content may be so contaminated, that the entire load is consigned for landfill or incineration disposal. With many buyers, especially in the export market but increasingly domestic markets too, requiring material at a certain specification, recycling and waste management companies are having to respond with investment in contamination prevention technology.
More contamination tends to lead to:
- More rejected material that needs to go to landfill or energy-from-waste – also means additional gate fee cost
- Material that is contaminated tends to attract lower recyclate prices than purer loads
- Contamination means greater environmental impact, not just from material being diverted to landfill, but also the fuel and employee cost of moving material you may not want
- Greater sorting potentially means greater health and safety risks on MRF picking lines
- If the public see material, they thought was being recycled instead ending up in foreign landfill sites or illegal dumps, then it undermines the positive recycling message.
While sorting technology continues to improve rapidly, it will always be more cost-effective and easier to identify contamination at source during the collection process.
Replace Manual Inspections with Automation
Human beings are good at spotting contamination, but this can be a difficult and timely process, and employees cost money.
To make inspections work, and to identify where the contamination is occurring, inspections need to take place before the container is collected. This means a person needs to open every container just prior to collection, investigate it and see what is going on. Potentially, they also need to tag the with a sticker or leaflet that mentions the contamination and outlines ways to reduce it.
If this happens just prior to a collection, which is often the case as this is when the container is available, then it slows down the operation of collecting it.
Another way to inspect material is to do it once the vehicle load has been tipped at the depot or MRF. The problem with this approach is that it is likely to be too little, too late. While you might be able to pick up identifiers such as names and addresses sometimes on contamination, the reality is that most of the time there will be no way to identify who caused the problem.
AMCS Vision AI – Identifying Contamination at source
Here at AMCS, we realised that the solution lies on the vehicle itself. By putting a camera on a vehicle looking at the hopper, it is possible to constantly identify both accepted and contaminated material, while also ensuring collections happen efficiently.
The way it works is:
- A high-definition camera looks over the hopper
- A recording from the camera is triggered by the lift sensor
- The data from this recording is sent to AMCS Cloud. This data includes:
- The video streams GPS data
- Time/Date stamp
- Event ID
- Vehicle ID
- At the other end, the data is seamlessly available in the AMCS Platform
- Our analytics software, using AI and ML reviews each image, detects, measures and scores contamination
- Actionable insights are generated by the platform, helping you to identify the customer and enabling you to educate them to reduce contamination. You can also use the images from their collection to show them what is the accepted material, and what is not.
AMCS Vision AI (Automated Material Identification Service) has the potential to reduce your contamination levels, your disposal costs and increase revenue from material sales.
For those companies that are curious to explore AI or ML, it also offers a fantastic starting point to see the power of this technology.
Click here to learn more about how AMCS Vision AI can give you automated insights across your recycling operations to drive better environmental and financial performance.
Learn how the AMCS Vision AI Automated Contamination Management solution provides a fully automated and smarter way to detect and prevent contamination.