A model for predicting the asphaltene stability class index (ASCI) was developed from the data obtained by Fourier transform infrared spectroscopy attenuated total reflectance (FTIR-ATR) coupled with partial least-squares regression (PLSR). The precipitation onset of the asphaltenes present in each sample with different ratios (n-heptane/toluene) of solutions was obtained for eighty-two different Colombian crude oils, generating a database of the asphaltene solubility class index. FTIR-ATR spectra were recorded in the mid-infrared region (4000 and 400 cm–1) for each crude oil sample. With the information collected, the prediction model was developed by the PLSR method, which included the criteria to choose adequate number of latent variables (LVs) and Monte Carlo cross-validation (MCCV). Results provide standard error of cross-validation (SECV) of 1.42. The model obtained by chemometrics could allow overcoming the problems of time of response and subjectivity in the determination of the stability of crude oils.