This study investigated the potential of visible near infrared spectroscopy (Vis-NIRS) to quantify the fatty acid (FA) composition of lamb meat under commercial abattoir conditions. Genetic algorithm based partial least squares (PLS) were used to develop regression models for predicting individual FA and FA groups such as saturated FA (SFA), monounsaturated FA (MUFA) and polyunsaturated FA (PUFA). Overall, the majority of the FA (C14:0, C16:0, C16:1, C17:0, C18:1 c9, C18:1 c11, C18:2 n − 6, C18:2 c9 t11 and C18:1 t11), intramuscular fat (IMF) and all FA groups were predicted with an R_CV², the squared correlation between observed and cross validated predicted values, which ranged between 0.60 and 0.74 and ratio prediction to deviation (RPD) values between 1.60 and 2.24. However the results for the remaining FA (C17:1, C18:0, C18:3 n − 3, C20:4, C20:5, C22:5, C22:6) were unsatisfactory (R² = 0.35–0.57, RPD = 0.76–1.49). This indicates that Vis-NIRS could be used as an on-line tool to predict a number of FA.