A classification of raw bovine’s milk samples according to theirgeographical origin in Peninsular Malaysia by Makmal Kesihatan Awam Veterinar, Department of Veterinary Services Malaysia (DVS). Six hundred bovine milk samples were collected from Perlis, Kedah, Perak, Selangor, Pahang, Negeri Sembilan, Melaka and Johor states by 26 milkcollecting centres under DVS. This study was carried out directly using attenuated total reflectance Fourier transform infrared(ATR-FTIR) spectroscopy method coupled with a multivariate principal component analysis (PCA). The spectra generated by ATR-FTIR were analysed and regions of interest were found in between 3851.651cm-1 until 2700.819 cm-1 and 2419.173 cm-1 until 977.368 cm-1. The absorbance and wavenumber data of the regions were then analysed using PCA and the results show presence of clustering towards theirgeographical origin. ATR-FTIR coupled with multivariate PCA has potential for classifying the geographical origin of raw milk produced within Peninsular Malaysia. This method provides a rapid and nondestructive secondary methodology in milk classification without further sample preparation.