by Chad Bebber
Our prediction: this will be a year of continued investment in equipment upgrades, robotics, and safety initiatives at MRFs across the country. That prediction is based on what we’re seeing at the dozens of sites, large and small, where we’ve embedded our high-performance work teams.
What’s driving these investments is a common goal: reduce cost and improve quality.
That’s Leadpoint’s goal, too.
Base Decisions on Facts & Data
The demands placed on MRF operators and managers get bigger and more complex every year. These individuals are expected to deliver continuous improvements to the bottom line, whether that means cutting costs, reducing contamination levels, lowering workers’ comp claims, or improving up-time on the sort line.
Today’s recycling leaders can no longer make decisions on the fly. They are expected to speak with facts and data – specifics that will help them achieve their site’s goals and add value to their customers, communities, and stakeholders.
The Leadpoint Data Tools mobile-friendly dashboard puts these data in one place, all accessible in real time, and provides four actionable productivity benefits to our customers.
1 – Employee Scorecard
Any site’s cost of labor is wages plus incidentals. But there’s much more to calculating the cost and value of each hour worked. Consider these data points:
- What is the target and actual performance of each individual sorter?
- What are the performance achievements and problems on each line and why?
- How does individual and team productivity vary by shift, time of day, or day of the week?
Our dashboard provides data points like these that can identify where coaching, training, or a change is needed to keep the MRF operating efficiently.
2 – Customized Data & Reporting
The individual KPIs tracked at any MRF can vary, with nuances or unique measurements. That is why Leadpoint’s customers appreciate that our dashboard is fully customizable, and can add value to the other services we provide and track.
For example, one of our customers was keen to get reporting on bale break audits. The customer gave us some information on their costs, we built in the measurement and created a feature in that customer’s dashboard.
3 – Predict the Future
No, we don’t have a crystal ball! What we do have is the ability to track and measure downtime by issue type, cause, and area inside the plant. That data is used to identify the most frequent problems, calculate their effect on operational speed and quality, and report trends.
With that data in hand – literally, on your smartphone – MRF operators can predict and plan maintenance work in off-hours, instead of having to shut down the plant. We think that’s a value-add, and so do our customers.
4 – Cost Savings
We created the Leadpoint Data Tools dashboard when we saw how many customers were spending labor dollars tracking and collecting data about performance, productivity, and efficiency. The data has to be collected somehow; our customers appreciate that we do the data collection and analysis for them, as part of our ongoing service.
People + Robots
We’ve watched our customers make tremendous investments in robotics, optical scanners, and other technology that replaces or supplements their human capital. Every time there is a bot put into a MRF, the expectation is that headcount and labor cost will be reduced enough to create a positive return on investment.
In our experience, operators may want to stay flexible in thinking about outcomes. An expected reduction of, say, three people may turn out to be just one. Why?
- Robots are new to the industry; we’re all learning about how to maximize their use in a MRF.
- Bots work best alongside people and process adaptation.
- Bot technology needs people to program, operate, maintain, and adjust them regularly.
- Humans are adaptable; AI isn’t there yet.
Because the need for human labor inside the MRF will shift in the future, we believe it’s imperative for operators and managers to know as much as they can about every individual hour worked, action taken, and issue created or resolved inside their plant.
With Leadpoint Data Tools in hand, that know-how is available, in real-time, customized to your site. Let us show you how!
Chad Bebber is Regional Sales Director at Leadpoint. Chad joined Leadpoint as a member of the Operations Support team in 2018. He was promoted to East Region Director of Operations in 2019 and to Regional Sales Director in 2020. His experience in the recycling industry, combined with his knowledge of manufacturing, allow him to consult with customers and prospects about enhancing the safety, operations and people challenges faced at their plants. Prior to joining Leadpoint, Chad held general manager positions with Green Life Waste Solutions in Burlington, NC, Recycling Management Resources in High Point, NC, and Weyerhaeuser/ International Paper in both Charlotte, NC and Itasca, IL. Contact Chad at chad.bebber@leadpointusa.com
One could possibly go into a very pertinent discussion about the upper limitations of artificial intelligence within a recycling environment.
This is because human labor supply is central to the service that this organization provides outside MRF suppliers.
Of course, this would be incredibly technical to decipher and hard to make understandable to a lay audience.
Here’s an article that’s generalist but relatively technical about the limitations of deep learning, a subset of AI, which is directly used with the vision systems of a robot pickers on the sort line.
https://lupinepublishers.com/material-science-journal/pdf/MAMS.MS.ID.000138.pdf
“In general, anything that requires reasoning -like programming, or applying the scientific method -long-term planning, and algorithmic-like data manipulation, is out of reach for deep learning models, no matter how much data you throw at them”
As it stands, it looks like AI can classify most objects in a material stream with high accuracy, but developing a fine mechanical sortation arm to pull both paper, plastic bags, any kind of non-regular ‘metal’ encountered in the stream in any 3 dimensional space arrangement with or without the presence of desirable or undesirable commodities may be an insurmountable mechanical and mathematical coding problem.
For instance, in a mixed paper stream, there is a lot of small OCC and plastic bags mixed in with the paper, and pulling this out without generating a huge amount of mess on the floor would require mapping robot ‘joints’ to act in a algorithmic, precise way, differently for every single piece of cardboard and plastic bag that comes down the line in slightly different orientations
A bot goes down for repairs, time and money lost. Humans self-repair every night and are more interchangeable.