Prediction of pork quality with near infrared spectroscopy (NIRS) 2. Feasibility and robustness of NIRS measurements under production plant conditions

Longissimus dorsi samples (685) collected at four processing plants were used to develop prediction equations for meat quality with near infrared spectroscopy. Equations with R2 > 0.70 and residual prediction deviation (RPD) ≥ 2.0 were considered as applicable for screening. One production plant showed R2 0.76 and RPD 2.05, other plants showed R2 < 0.70 and RPD < 2.0 for drip loss %. RPD values were ≤ 2.05 for drip loss%, for colour L* ≤ 1.82 and pH ultimate (pHu) ≤ 1.57. Samples were grouped for drip loss%; superior (< 2.0%), moderate (2-4%), inferior (> 4.0%). 64% from the superior group and 56% from the inferior group were predicted correctly. One equation could be used for screening drip loss %. Best prediction equation for L* did not meet the requirements (R2 0.70 and RPD 1.82). pHu equation could not be used. Results suggest that prediction equations can be used for screening drip loss %.

► Samples measured at production plants could be used for prediction equations. ► The best equation for drip loss % was developed on data with the highest CV value. ► The results suggest that prediction equations can be used for screening drip loss %.