Pallet users have always said that they want better quality pallets. But frequently they have not been willing to pay for the difference. Things are starting to change thanks to warehouse automation and the need for higher quality pallets to ensure this automation functions smoothly. Automation may be opening the door for new conversations about pallet quality with some customers.
Judd Michael, professor of biorenewable Systems at Penn State University, spoke earlier this year at the National Wooden Pallet & Container Association Annual Leadership Conference on pallet scanning and inspection. He said, “We are in the golden age of supply chain automation, and this trend is driving demand for higher-quality pallets. At the same time, labor shortages are making it harder to find people who can monitor and produce consistent quality.”
This golden age is causing pallet users to re-evaluate their standards and impose stricter requirements. After all, why invest millions in automation and allow a low-cost pallet to disrupt operations? Big companies are waking up to the importance of a new focus on pallet quality. Will white-wood suppliers see this opportunity and take advantage of the changing landscape or cede much of this market to rental options? When it comes to automated storage and retrieval systems (ASRS) or automated palletizers in the fast-moving consumer goods supply chain, the cost of downtime can be immense. The average downtime cost for FMCG manufacturers caused by bad pallets is $15,000 or more per production stoppage incident. Typically, the cost is $4,000 per minute for disruptions.
Michael discussed the impact of downtime with a number of major FMCG manufacturers. He found that the biggest cost associated with downtime could run as high as $480,000 per day. He said, “Let’s compare $480,000 a day to an incremental cost increase per pallet, let’s say maybe 10%. It just doesn’t make sense to buy cheap pallets if your downtime is costing you so much money.”
In addition, automated inspection and sortation is much more accurate and reliable than the manual approach. Human/manual sortation is only 60-70% accurate, according to what Michael found. When you combine advancements in electronic imaging and scanning with artificial intelligence (AI), the accuracy is impressive. He commented, “AI does eliminate the variability that you see with human graders and inspection. But perhaps the biggest reason for AI is the difficulty finding good labor – people who will consistently apply a standard. Plus, the increasing cost of labor is going to drive the industry towards more technological solutions.”
So, what are the business incentives for different players when it comes to AI pallet inspection? Michael said, “On the FMCG side, if they buy an IVSYS, Pallet IQ, or Neocortex system or one of the other competitors, and they start scanning at a DC, then they can do their own sorts. The pallet user will keep the goods internally, reuse those within the supply chain, and then return the low-quality pallets to a recycler. This has implications for what you get as a recycler from your dock sweeps; instead of this nice mix of good to bad, all you’re getting is low-quality pallets.”
This could radically change the balance of power in the pallet market. Michael warned, “I think that would really hurt the business model for recyclers…I suspect that the FMCG guys and the big retailers, if they have these scanners, they’ll find a way to game the system and then everybody else loses including the recycler and the rental companies.”
While many DC managers don’t want to be in the pallet business, they can save a lot of money if they do give pallets more attention. Michael suggested, “I’m pretty convinced that if DCs could put in one of these relatively inexpensive scanners, they could drastically change their cost structure for pallets.”
If a pallet recycler puts in a pallet inspection system, it can monitor quality and production output as well as core supply dynamics. And the reality is that whoever leads the installation process will likely be in a better position to guide the discussion when it comes to pallet quality, core supply and pricing. A recycler who is armed with data can better negotiate true core pricing based on quality delivered.
Also, the pallet recycler can guarantee quality to meet any specification. Michael explained, “It’s a competitive advantage for me as the pallet producer or recycler because I can go to you as the FMCG and say, ‘Instead of these other ding-dongs that are providing you 70% of what you want, I’m going to give you 99% grade A pallets. You will not have problems in your automated systems, and I can promise you that.’”
Researchers at Penn State University have worked with a start-up sawmill in the South that is piloting AI-based scanners that can automate inspection of industrial hardwoods and softwoods. This team has also worked with IVISYS and iGPS on pilot programs for pallet inspection. The sawmill project uses a Microtech/Lucidyne scanner and Nvidia processor running an AI model. It can grade based on eight different classifications at a rate of 20 boards per second, producing a 99.8% accuracy. This system only takes up a small footprint and can print the grade on each board. The developer behind this system is looking to sell it to other companies beyond the pilot programs with Pacific Northwest and Southern lumber producers.
Michael believes this technology opens up an entirely new opportunity for industrial grade lumber producers. He stated, “I think that the ability to scan using artificial intelligence and then sort effectively with a high degree of accuracy is going to be very valuable to multiple parts of the industrial packaging supply chain.”
Lumber grading is connected to pallet quality and is the basis for efficiently producing higher-quality pallets at lower costs. If a pallet producer is vertically integrated (runs its own sawmill), the company can optimize production and yield based on specific customer needs and existing inventory. Since there are no specific grades for low-grade lumber at the moment, an AI system does not have to follow a specific protocol, such as the NHLA lumber grading standards. Michael suggested, “I believe low-grade lumber is easier to grade than fine appearance grades for furniture and flooring. The threshold for acceptance and what the rules are with industrial grade are much lower, and there oftentimes aren’t official rules.”
Some advantages of AI include the ability to get better over time, whereas human operators face fatigue and struggle to stay consistent in applying a standard. This leads to errors and inconsistency. Michael admitted, “The accuracy is very high, sometimes over 99% for redwood and cedar boards.”
Looking to the future, Michel projected, “I think with the speed that these guys have plus the accuracy, it’s going to displace human graders.”
These same benefits apply to pallet inspection as well. Consistency is critical to improve pallet quality and performance in automated warehouses. Fred Vrugteveen, visionary at Niagara Pallet, noted, “When you have 40-50 people in your plant making decisions on how to grade a pallet, it becomes very challenging to provide consistent pallets to your customers.”
When iGPS deployed automated inspection (IVISYS) in a test environment, the plastic pallet pooler discovered that the automated system improved accuracy, reduced waste and shrunk the inspection time from 45-60 seconds to 7-10 seconds. Michael stressed, “Automated inspection has been very successful for iGPS.”
Automated inspection can also be tied to pallet design software to evaluate if a pallet meets a specific design parameter, such as load-bearing capacity. The types of factors that can be analyzed include: the dimensional accuracy of the pallet, protruding nails, cracks in boards or blocks, misaligned blocks, missing, broken or dislocated boards or blocks, logo detection and verification, and plastic/paper residue.
Automated systems also reduce worker stress and physical strain associated with lifting and flipping pallets. Michael questioned, “Maybe this scanning technology makes the most sense in the used pallet world because it provides the best opportunity to upgrade a pallet and improve overall pool quality?” But there are some challenges that the industry faces. For starters, there are not clear, consistent grading standards for what makes a pallet an A-grade vs a B-grade or a premium pallet. Every recycler and maybe every core supplier has its own definition. The pallet industry and its core customer base need to develop better consensus standards. But this is something that many recyclers will reject because the lack of consistency provides greater profit potential. Also, pallet users may have reasons for not holding to consistent standards as well.
Michael explained, “I don’t think that all of the retail users want to check or test the grades. For example, I’m working on two lawsuits now where a little old lady tripped and fell in a grocery store. The grocery chain had set standards for pallet quality, and yet internally, they ignored those standards. I believe it’s partly because of the cost to police standards. Pallet users don’t want to hassle with inspecting pallets, offloading and reloading loads on better pallets, etc. Some retailers claim to want a great pallet, but in reality they don’t always police that requirement.”
Also, pallet companies have been willing to deliver high-quality pallets all along. The problem is that many customers were not willing to pay for it. The reality is that in most supply chains, the person who buys the pallet is not the one who has to deal with the consequences of low-cost pallets. Michael agreed, “There’s a disconnect between the people who run the automated system where they do palletizing and the people whose bonus is based on the cheapest pallet possible. And that’s the fault of the FMCG because they are operating in silos.”
Why has it taken so long for pallet users to truly understand and embrace the importance of pallet quality? Maybe it is the lack of automation in many U.S. warehouses that allowed inferior-quality pallets to suffice in the past? Poor pallet quality led to more damaged products and more unsaleable items.
Things are starting to change as pallet users begin to install inspection systems in facilities. This is maybe the biggest indicator that pallet quality is becoming a much greater concern. What customers want will eventually drive the market because pallet providers don’t want to lose customers. Michael suggested, “I think we’re getting very close to that tipping point where there are enough of these pallet scanning systems in place, either within the lumber side or pallet manufacturing, recycling or somewhere down the supply chain where people will be able to evaluate the true benefits.”
Some companies will choose to just improve pallet inspection processes without installing scanning or AI software solutions. But as the technology improves, greater adoption may occur. Michael predicted, “As the technology becomes better and better and cheaper, it is going to make a lot more sense for companies to adopt it. And then we’ve got the driver of the need for pallet quality, which is these automated systems.”
To find out more about the pallet quality, scanning and AI inspection research that is going on at PSU, contact Judd Michael at packaging@psu.edu or call (814) 863-2976.
Key Questions to Ponder
1.) What criteria do you think should become standard for pallet recycling grades?
2.) Will AI scanning expose the “warts” in the system?
3.) Could your business model be changed by new AI-driven pallet inspection technologies?
4.) What parts of the supply chain are most likely to benefit from, or get hurt by, the adoption of new scan/sort systems?
