SIMULATED INSTRUMENT AUGMENTATION OF USDA YIELD GRADE APPLICATION TO BEEF CARCASSES

Because no instrument technology has been shown to predict beef carcas s composition better than USDA yield grades, this study was conducted to determine whether an instrument could be used to augment and improv e the accuracy of USDA yield grade placement. Adjusted preliminary yie ld grade (PYG), ribeye area (REA), estimated percentage of kidney, pel vic, and heart fat (KPH), hot carcass weight (HCW), and USDA yield gra de (called and computed) were determined by five on-line USDA graders and two USDA grading supervisors for beef carcasses (n = 550) selected randomly in a commercial beef packing plant. Data were compared (2,73 7 comparisons) to Gold Standard yield grades and yield grade factors d etermined by an expert panel of carcass evaluators (unrestrained in ac cess or time to evaluate carcasses). On-line USDA grader PYG were clos ely related (mean absolute error of .15 +/- .14 yield grade units; r = .91), and on-line REA and KPH were nominally related (mean absolute er ror of .51 +/- .35, .06 +/- .07 yield grade units and r = .48 and .66, respectively), to Gold Standard yield grade factors. On-line USDA gra ders determined adjusted PYG effectively, but they may require instrum ent assistance to evaluate carcass muscling traits and perform time-se nsitive computations. To explain why instrument technology may not est imate beef carcass fatness as accurately as USDA yield grades, the abs olute mean difference between Gold Standard measured PYG and adjusted PYG were compared. Only 5.6% of the sample population required no PYG adjustment, 94.4% required some adjustment, and 11.0% required over a .5 yield grade unit adjustment. Yield grades for beef carcasses, calle d by the USDA graders and supervisors at chain speeds, resulted in gre ater accuracy (absolute mean error of .24 +/- .43 yield grade units; r = .82) than when yield grades were computed for carcasses using the y ield grade factors determined by on-line USDA graders and supervisors at chain speeds (absolute mean error of .52 +/- .41 yield grade units; r = .75). Gold Standard yield grade factors were sequentially substit uted into the short-cut USDA yield grade equation for the yield grade factors determined at chain speeds by the USDA graders and supervisors . Results suggested that instrument augmentation would improve accurac y and precision of yield grade placement if on-line USDA graders deter mined PYG and an instrument determined REA and performed the necessary computations, incorporating KPH and actual HCW (P <.05).