Near infra-red (NIR) spectroscopy with partial least square (PLS) regression of the results was developed as a rapid method for the determination of the hydroxyl value (HV) for palm-based polyols. The best NIR PLS calibration models to predict HV were developed in two types of spectral regions, Model 1 at 6500-8100 cm^-1 and Model 2 at 4500-10 005 cm^-1. For Model 1, the NIR-generated calibration plots had coefficients of determination (R²) of 0.991 and a standard error of calibration (SEC) of 3.23, while the validation plots had R² of 0.994 and a standard error of prediction (SEP) of 2.96. For Model 2, the calibration plot had R² of 0.997 and SEC of 2.32, while the validation plots had R² of 0.998 with SEP of 2.52.