Put yourself in the shoes of a federal meat grader in a commercial packing plant. You have about 10 seconds to assign each beef carcass a yield and quality grade, as thousands will pass by your station during a shift.
At that pace, coupled with variation in the carcasses streaming by and the demanding conditions of a packing plant, is there any wonder you get blamed for inconsistencies and errors associated with subjective evaluation? And, you have cohorts in scores of other packing plants around the country doing the same work.
For these reasons and more, the beef industry has searched since the 1970s for an answer — mainly a machine that would efficiently, accurately and objectively assess carcasses for both quality and yield grade attributes.
In February 2001, USDA's Agricultural Marketing Service (AMS), the agency that governs federal beef grading practices, approved the use of electronic vision instruments — basically computerized cameras — in applying official U.S. yield grades to beef carcasses (see: “Value-Vision,” December 2000 BEEF).
But the industry was only halfway there. Graders, who on average examine 450 carcasses/hour, still had to estimate the degree of ribeye marbling — the primary determination of quality grade — within each carcass.
Earlier this year though, AMS announced it would begin accepting carcass-marbling scores made by approved instruments for use in augmenting the determination of official USDA beef-quality grades.
Colorado State University (CSU) meat scientist, Keith Belk, along with colleagues Gary Smith and Daryl Tatum, is among the pioneers in the application of instrument grading in the beef industry.
“It's clearly a paradigm movement, but one that's taking us in a direction that will allow better management of cattle and carcasses,” he says. “It provides some real, value-adding opportunities during cattle production and within the packing and processing sector of the industry.”
Not really so new
Although instrument grading is just now entering USDA's official quality-grading spectrum, it's been used “in-house” at six of Cargill Meat Solutions' North American fed beef plants. Glen Dolezal, Wichita, KS, director of new technology for Cargill Meat Solutions, says his company has been exploring the use of instrument grading for 15 years.
“Today, we run our business with this technology,” Dolezal explains. “It's an excellent tool for supplier feedback, planning, sorting and providing innovative customer solutions.”
His company employs the machines to evaluate beef carcasses for backfat thickness, ribeye area, percent marbling, lean color and fat color. They also can be used to track cattle-buyer performance, as well as the consistency of USDA graders. Another use is to measure fabrication-floor performance within plants.
From a regulatory standpoint, USDA stresses the machines are intended to augment the job of a federal grader. The agency has taken a measured and deliberate approach in allowing the industry to implement instrument grading.
“We've used every bit of scrutiny in embracing the technology,” says Marty O'Connor, AMS Livestock and Seed Program Standardization branch chief. “With regard to the prediction of the marbling scores, we want to ensure reliability, accuracy and repeatability when applied at plant chain speeds of up to 500 carcasses/hour.”
The AMS requirements for instrument grading were established after consultation with representatives of USDA, National Cattlemen's Beef Association (NCBA), packing companies, producers, instrument manufacturers and academia. An industry working group collaborated on an instrument grading phase-in plan for individual plants:
Phase I — Demonstration of the repeatability of marbling score prediction on stationary beef carcasses.
Phase II — Demonstration of the accuracy of marbling score prediction at line speeds.
Phase III — Documentation of daily in-plant verification of the approved instrument to validate calibration and that correct data is presented to the USDA grader.
“Each plant that wants to implement this technology will have to demonstrate performance on a day-to-day basis,” O'Connor explains. “How soon they roll it out depends on how long it takes to get their quality manuals developed and outline the operational elements they have to address in their plants.”
Packing plants like Cargill's, which already use the cameras, may or may not be ahead of the game — depending on how cameras are being used internally, and how the uses match USDA's requirements for image capture, O'Connor says.
He emphasizes instrument grading is initially a voluntary means to augment the voluntary practice of quality and yield grading. It's up to individual firms to determine whether to use the technology in supplementing grading systems.
Dolezal says the requirements needed to implement instrument grading — and satisfy USDA — take a totally different approach to grading from what have been standard operating procedures in many plants.
“The government has outlined performance standards for this technology,” he says.
Therefore, along with a price tag of about $250,000/system — plus the cost of service contracts, software updates, training, etc., — instrument grading isn't likely to sweep the industry overnight.
“We at Cargill hope to roll it out in at least one or more of our regions by the end of this calendar year,” Dolezal says. “We're highly confident in the system, and expect to capture error-free images on a minimum of 98% of all our carcasses.”
Advancing a merit system
Bo Reagan, NCBA's vice president of research and knowledge management, says most of the focus on instrument grading has been on performance standards for the machines and the related software systems. While the instruments will provide the official call for marbling, it doesn't mean humans will necessarily be replaced.
“Federal graders will still have a full-time role in assigning the grades,” Reagan says. “But we're on the right track with this technology because of the consistency it provides.”
CSU's Belk says the source of most of the inaccuracy in assigning quality grades comes in the measurement of marbling scores.
“This system takes out the human factor that can move, and does move, in assessing marbling scores,” Belk says.
Dolezal doesn't hesitate when asked if instrument grading can remedy quality-grading inconsistency problems Cargill has experienced in certain regions.
“Absolutely,” he answers. “Instrument grading will help level the playing field from region to region, plant to plant, shift to shift, and grader to grader.”
Dolezal believes improved carcass-evaluation accuracy dovetails with technology trends in the production sector.
“We have some very progressive feedyards and alliances using cutting-edge animal sorting and live-animal evaluation systems,” he explains. “They get frustrated upon receiving grading results from the packers — either quality or yield grade — inconsistent with their pre-harvest instrument-based predictions.”
Consensus is that instrument-grading technology will help beef producers get paid exactly what their cattle are worth. This is especially important with an advancing “buyer's market” where price spreads widen as packers have more cattle from which to choose.
Belk certainly doesn't mince any words in predicting what this technology means to cattle producers. He says cattle producers participating in value-based alliances, at least initially, are likely to benefit the most from more accurate and consistent quality-grading systems.
“It's the next step toward true value-based marketing,” Belk maintains. He adds the increased grading precision will upset the “socialistic” system where producers of higher quality cattle are paid averages less than their carcasses are worth in order to “subsidize” the production of the poorer quality cattle.
“Producers who don't know how their cattle perform as a food product might not like it,” he adds. “If you're on the bottom end of the average in terms of carcass quality, you're going to get slammed.”
Managing 100+ sorts
With cattle being directed to fit more than 100 different customers' specifications, Dolezal says, at least for Cargill, instrument grading helps create “sameness” in sorting those carcasses.
But, while instrument grading is going to help meet those customers' needs, quality grading isn't a perfect indicator of eating satisfaction. Therefore, there's room for improvement in the evaluation of other end-product traits.
“We need to work on an objective measurement for other palatability traits, and there's still interest in on-line tenderness evaluation,” Dolezal says. “This means finding ways to further filter carcasses out of Select and low-Choice that, for one reason or another, cause problems with consistency in eating satisfaction.”
The challenge, he adds, is to find within that big block of carcasses those that need to be “enhanced” or in some way further processed to meet consumer expectations.
As a researcher, Belk jumps on the idea of using every available tool to add value to a beef carcass. He says there are a lot more possibilities for imaging technology than just helping to assign grades.
“With 22,000 possible products that can be derived from any given carcass, based on how muscular it is and how fat it is, there's an optimal way to fabricate each primal cut from each carcass,” he explains. “We know there's upwards of a $100/head value opportunity to packers depending how they decide to fabricate a carcass. If the industry could capture even half of that amount, it would be an enormous help.”
The technology allows packers more capability to sort each carcass to fit the “styles” to be fabricated at a time — maximizing the carcass's value potential. Belk says that gives packers more opportunity and incentive to send signals back though the system as to what kind of cattle are needed in fabrication.
The move to objective carcass evaluation is an area the beef industry has worked on for a long time. With nearly 25 million beef carcasses to grade this year in the U.S., federal graders have their hands full. Even given a 95% agreement rate with independent program evaluations and reviews, there's little argument there's room for improvement.
“We're excited to see the concepts finally coming to fruition,” Reagan says. “Once individual packing plants develop their standard operating procedures for instrument evaluation of carcasses, there's no telling where it will all end — but it's coming, soon.”