Pondering Artificial Intelligence
Over the last several years, robotics technology and artificially intelligent solutions have increasingly found their way into the recycling and waste management industry. Used in sorting applications, this technology is becoming increasingly mainstream and affordable. It can be easily integrated into current systems for processing materials at MRFs, C&D and other recycling facilities, for managing both single and mixed streams.
According to ZenRobotics, a current leader in the field that recently announced the installation of the first wind-powered robotic sorting system in Sweden, their ZRR2 unit, with two arms, “makes up to 4,000 picks per hour,” can work 24/7, be configured as multiple units, and provides “low energy consumption and up to 98 percent purity.”
It is tough to imagine how manual sorters can compete with this, even when equipped with the various other types of sophisticated sorting technology that is currently available, such as optical and magnetic sorting, air-ejection and eddy current separators. Much of this technology is also now being designed from the start with integrated artificial intelligence.
The latest sorting system available from BHS/NRT, for example, incorporates Max-AI, which is described by its manufacturer as utilizing “deep learning technology” and employing “both multi-layered neural networks and a vision system to see and identify objects, similar to the way a person does.” (Details at www.recyclingproductnews.ca and more to follow in our May/June 2017 issue.)
Indeed, the complete transformation of our industry at this point seems inevitable. Emerging robotics technology is not only very fast and efficient, and increasingly so every year, the technology is literally becoming “intelligent,” meaning it is being programmed to learn on the job, as it goes. This results in efficiencies that get significantly better, the longer it works.
This issue, our story “Advancing the sort” by Associate Editor Lee Toop features a pilot project underway through the Carton Council, in partnership with Alpine Waste & Recycling out of Colorado and using AMP Robotics technology. The testing being done in Colorado is an example of how the industry is working to “take advantage of advances in machine learning – helping computerized systems educate themselves on the job.”
Of course, this all raises the question, as the use of robotics and artificial intelligence in all industries does, what impact on labour does the adoption of this kind of technology bring with it? The loss of sorting line jobs in the industry is the immediate downside, to which there is no solution being put forth as of yet. For now, to some degree we may have to simply accept this as the price that must be paid to keep up with advancing technology. Definitely, the recycling industry is one in which profit margins can be relatively low and where increased efficiencies in sorting materials and extracting maximum value from our waste streams is essential for any recycler’s long-term sustainability.
We certainly welcome comments with respect to this issue anytime, and look forward to hearing more about AI as well as seeing the latest robotics technologies available, at WasteExpo 2017, coming up in New Orleans, May 8 through 11.
This article was originally published in Recycling Product News, April 2017, Volume 25, Number 3.