Thanks to modern day technologies, the former process of sorting certain materials by hand can now be automated. In fact, thanks to optical sorting, different types of materials can be detected recognized, and then separated.
The article is intended for any professional in the waste and recycling industry interested in separating material. Optical sorting technology is no longer just for Material Recovery Facilities (MRFs). It is now applicable for sorting food waste, metals, fibers, construction and demolition waste, or even electronics. Of course, plastic reclamation also benefits tremendously from this technology. Why do we care?
As population grows, so does the number of and types of material recovery programs. Whether it is electronics or cardboard, the waste industry, which is generating an estimated $63+ billions a year, is largely focused around one central processing plan: effective sorting. Effective sorting drives up quality, reduces contamination, and pushes the economy of recycling and recovery forward. Meanwhile, as composition of goods seems to be consistently changing, sorting systems need to evolve at an equal speed.
Fast, quick, or cheap?
In almost every processing situation, the large volume of incoming material necessitates processing equipment to move and sort material at high speed. At the same time, the highest value is obtained from the purest, least contaminated streams. In the case of most MRFs, in order to accomplish these somewhat contradictory goals, they employ equipment that sorts the material automatically. Optical sort machines are both faster and more accurate than manual sorting.
How does it work? The operating principle of an optical sorter is quite simple: materials on a conveyor belt travel at high speed under a powerful light source. Part of the light’s wavelength is absorbed by the materials while the other is reflected and captured by lenses. These lenses transmit the signal to a spectrometer and/or camera, which associated each reading with a specific curve since each material has its own features or “signature”. This is what allows the optical sorter to detect different types of material.
Source: https://digital.hbs.edu/platform-digit/submission/tomra-potatoes-to-the-right-rocks-to-the-left/
Coupling with Artificial Intelligence As with other industries, artificial intelligence (AI) has tremendous potential in the waste and recycling industry. The ability to capture data from optical sorters, means that machinery also has the ability to learn from this data and “make” decisions that optimize sorting with the highest returns. This means that today’s connected machines can adapt as waste streams and recovery programs change. One Key player in this space:
TOMRA
TOMRA’s solutions enable advanced collection and sorting systems across food, recycling, and mining industries. The Norwegian company started out with Reverse Vending Machines (RVM) but has expanded quickly into other sensor-based solutions for sorting and processing a variety of products. In 2019, TOMRA developed its deep learning AI- technology, GAIN, to further enhance its current soring technology. GAIN aims to mimic human learning by training based on thousands of images to determine which items should be separated during the sorting task. Then, the technology continues to learn and adapt to new waste streams, which gives customers more flexibility to adapt to changing market conditions.
References:
https://www.wrap.org.uk/sites/files/wrap/NIR%20Good%20practice%20guidance%20for%20existing%20NIR%20users%20Final.pdf
https://plasticsrecycling.org/images/pdf/design-guide/Resources/NIR_Sorting_Resource.pdf
https://resource-recycling.com/recycling/2017/08/01/all-eyes-on-optical/
http://www.sherbrooke-oem.com/optical-sorter
https://digital.hbs.edu/platform-digit/submission/tomra-potatoes-to-the-right-rocks-to-the-left/
https://www.statista.com/statistics/192829/revenue-of-the-us-solid-waste-management-industry-since-2000/#:~:text=U.S.%20solid%20waste%20industry%20revenue%202000%2D2017&text=The%20solid%20waste%20industry%20in,dollars%20of%20revenue%20in%202017.