Can technology reboot the recycling market?
Recycling has been common practice for over 50 years. Unfortunately, practice does not always lead to perfection. Garbage is often tossed into recycling bins, contaminating all of the contents. Another common practice, wish-cycling, mixes unusable materials with otherwise-valuable materials.
That’s just two issues that recur in recycling. The problems have weakened commodity values, and diminished values can make recycling impractical. After considering the financials, several large municipalities no longer collect and process recycling.
Fully aware of the declines in recycling quality as well as lower commodity values, Solid Waste Management (SWM) companies are finding solutions. Among the technical solutions, robotics and Artificial Intelligence have been implemented at both curbside and Materials Recovery Facility (MRF) locations.
CURBSIDE SOURCE SEPARATION
Self-sorting waste bins assist hurried humans at the food courts in airports, stadiums, and other public locations. These systems use Artificial Intelligence (AI), constantly refining the ways that they separate food court waste into garbage, recyclables. and organics. Prototypes now in development will sort the waste even faster and more accurately. As systems and the market mature, better equipment will be available in more places.
AI also works along city streets. Vehicle and bin sensors, connected through smart systems, assess materials and avoid contamination issues before the truck reaches the MRF. Even before a truck deploys, customer service representatives can access a bin’s time and location-stamped records to focus on severe or recurring concerns.
Along many MRF conveyor belts, the mechanical arms and camera eyes of robotic sorters supplement the manual sorting work done by employees. A training specialist from the systems manufacturer may also be at an employee’s side, providing orientation to a technologically advanced workplace.
Robots receive training as well. Most of that is done as machine learning during system setup. That brings about Artificial Intelligence (AI) which continues to develop after a system goes live. In daily operations, AI boosts the robotic brain that controls system eyes, arms, and hands. All of those may be deployed within enclosed equipment or at varied locations along a sorting line.
The electronic eyes of a Visual Information System (VIS) — spectroscopic, 3D laser, or conventional video cameras — scan garbage that passes along a conveyor belt. In some cases, individual pixels within an image are separately analyzed. The AI-boosted brain then assigns digital descriptions to food cartons, glass, cans, plastic containers, and more.
As VIS capture images and AI classifies individual pieces of material, MRF operators receive real-time graphics and reports with which they can refine the system and monitor outbound residue.
Beyond VIS imagery, other specialized sensors detect specific properties of material passing along the conveyor. Complex Autonomous Quality Control (AQC) systems recognize and sort several types of plastics in varied sizes and shapes. Data from metal sensors triggers robot arms to extract ferrous, aluminum, other metals. As a single plastic item is classified by type, it is also weighed. Objects which weigh more than others of similar size and composition are pulled from the line. They may be bottles which should be drained before further processing.
Civic leaders and MRF management have seen their technology investments return value many ways. Some of the benefits have been quick wins, others will take more time. Some benefits are easily quantified, others are not:
Robotic sorting machines can safely and efficiently tackle hazardous and dirty jobs. Robots won’t get stuck by a needle or develop back injuries. They can reach into a waste stream and lift out items that don’t belong there.
On single stream container lines, robots typically achieve 70 to 80 picks per minute. Manual sorters may pick about half that rate.
Even as a greater variety of materials can be handled, higher-quality sorted bales are produced.
- Human Resource Management
As the routine sorting is done by machine, employees can be trained in the technical skills relevant to a 21st-century facility.
With the data provided by AI systems, MRF operators can optimize system performance and better respond to new sources and types of materials.
 Waste360.com — NWRA Recognizes Best Recycling
 Waste360.com — Big 3 Solid Waste Companies Talk Recycling
 CBC.ca — CBC Radio (Canada)
 RecyclingToday.com — Single-stream recyclers share robotics AI insights
 RecyclingToday.com — Suppliers of robotics AI discuss technology applications
 RecyclingToday.com — Video robotic sorting systems
 RecyclingToday.com — MRF adds robotics