A quick search of the MEAT+POULTRY website produced 17 stories from 2019 involving recalls due to foreign matter contamination of meat and poultry products. Foreign matter recalls have spiked recently for several reasons, from government pressure on facilities to react and eliminate problems quickly, to an inaccurate understanding of the purpose of detection equipment and its placement in the production line, to the fact that some foreign materials are just plain difficult to detect.

Preemptive detection

The first step to eliminate foreign material contamination starts long before any X-ray machine or metal detector finds it in the product. While detection equipment at the end of a processing line certainly makes sense, advantages to placing equipment early in the process also has value.

The end of the line creates challenges in the way of packaging. Robert Rogers, senior advisor for food safety and regulations at Columbus, Ohio-based Mettler Toledo, says by that point in the line product is already in a package of some sort.

“The packaging material could potentially have an effect on sensitivities,” Rogers says. “It could be the size of the packaging is larger at that point because it may be in a final box or final case, inspection. So, the inspection technology has to be able to handle that larger product profile.”

Another issue arises with detection at the end of the processing line. Once equipment detects foreign matter at this point, the company has added all their value to the product. A true root cause analysis of a contamination now entails going all the way back through the value chain, internal processes and even the raw material supply chain.

“One of the solutions is to go through the process and understand where installing an inspection device would provide value,” Rogers says. “Not only in the detection capability, but also in the potential to reduce waste and effect equipment.”

Rogers gives the example of placing a detection device prior to a grinding stage in the line to remove any contamination before entering the grinder. This prevents contamination being ground into smaller and harder to detect sizes, as well as potentially damaging the equipment.

“If you have somebody that doesn’t understand the overall program and what to do when a contamination or a reject occurs, or how to handle that rejected product and to get into the lab to do a root cause analysis and identify where the sources of those procedures aren’t in place, then things can get by,” Rogers says.

Erik Brainard, global vice president, Anritsu Infivis Inc., agrees with Rogers and adds that detection equipment is a tool and operators must know how and where to use the tool correctly.

“An example of that is where someone might install an X-ray machine and think all of their metal or other foreign material issues are going to go away because they now have protection,” Brainard says. “When in fact, they may need another unit in a different location to truly maximize contaminant detection.”

The problem with plastic

Plastic contamination poses a distinct and difficult challenge due to its low density and the fact that it’s not magnetic or conductive like metals. Apart from some plastics that are manufactured with metals or other materials in them to make them detectable, or a few that are significantly dense, Brainard says plastics are virtually undetectable by X-rays or metal detectors.

“It has to come back to the customers protocols, making sure they actually check all of their equipment ahead of time to make sure there isn’t any damage to any plastic materials or packaging materials,” Brainard says. “The joke is, if we could find plastic, we’d get rid of our logo and just say ‘We Find Plastic.’ Given the challenges in detecting plastic, customers must be rigorous in regularly examining their line equipment for damage that might inject plastic pieces into production.”

While the advanced detection technologies of today do help in the fight against contamination, the nature of contaminants and the machines that detect them can only do so much.

“Certainly, there are some leaps forward in technology, capabilities, processing speed, things of that nature,” Rogers says. “But when you really look at, how a metal detector works or how an X-ray inspection system works and what it can and can’t detect, a lot of those capabilities, and more so limitations, are really based on overall physics.”

The physics of plastics give them the ability to go undetected. The same goes for certain metals. Stainless steel is not magnetic and is less conductive, making it difficult for a metal detector to identify it. Aluminum is less dense, so X-ray machines may have more trouble with it.

Detection in the ‘20s

Brainard notes that throughout his career he’s seen a steady move toward X-ray detection versus metal detection, although both technologies have continually evolved over the years and metal detection is still a high-value market. The machines have become more accurate and gained the ability to detect more materials.

He also says the technology needs to evolve further and continue to become more sensitive. The ability to pick up the less dense materials will enhance food safety in the future, and the technology is already moving toward that goal.

“We can see certain types of plastics today,” Brainard says. “PVC and Teflon are detectable because they’re dense enough for the technology. But as the technology evolves, hopefully we’ll start seeing other materials, including bone, at a higher levels.”

Rogers talks about the future being data collection, and more specifically, how to use the data. Basically, detection equipment currently operates on a pass/fail spectrum. When tested, a machine detects or it doesn’t, but according to Rogers, there’s more to it than that.

As an example, he explains that machines can be tested on a scale of 1 to 10. Rather a machine gives a detection signal strength of 1 or 10, it’s a pass. But closer testing and data collection might indicate a machine repeatedly gave a signal strength of 10, and suddenly one day the strength dropped to 9.

“It’s still detecting because I only need one to detect, but I see a variation that’s occurring, Rogers says. “Now let me during a non-production time, schedule maintenance out there to go figure out why I’m only getting nine instead of 10. Things like that is where we start to impact the overall bottom line. Not only ensuring that all the products going through the device are safe, good quality products, but the data that comes out of the machine can provide analytics to affect the overall bottom line.”