Gasoline adulteration remains a global problem due to its economic, health and environmental impacts. The gasoline adulteration monitoring methods used in the present study were not effective in most developing countries due to the associated implementation costs. Therefore, a fast, reliable, and cheaper method of vibrational spectroscopy in combination with chemometric tools was developed. In this study, sensitive FTIR spectroscopy was proposed in combination with multivariate techniques to analyse adulteration in Mizoram gasoline fuels. Standard solutions were prepared by mixing the gasoline with different proportions of kerosene and methyl tert-butyl ether (MTBE), which are then used for model calibration. The chemometric techniques used were principal component analysis-linear discriminant analysis (PCA-LDA), classification of partial least squares discriminant analysis (PLS-DA) and the regression method (PLS). The developed PCA-LDA classification model has an overall error rate of 12.5%, while PLS-DA has an accuracy of 100% for model. The PLS regression models of kerosene and MTBE with a high significance value of R2 0.996 and 0.986, RMSEP 0.275 and 0.291 were able to predict adulterant and MTBE concentrations, respectively. However, PCA-LDA achieved better accuracy than PLS-DA in classification analysis, especially on oxygenated gasoline samples.