Abstract
Fourier-transform infrared (FTIR) spectroscopy is widely used instrumental technique in pharmaceutical analysis. In this study, three pharmaceuitcal active ingredients representing different clinical and therapeutic classes (diflunisal, diltiazem and hydroxyzine) were investigated using regional spectral descriptors, multivariate analysis and Monte Carlo robustness modeling. Absorbance values of the drugs were calculated over predefined three diffrenet spectral regions and descriptors were derived to structural discrimination. Principal Component Analysis (PCA) results showed that the first two components explained more than 92% of the total variance, and that discrimination was particularly observed in the 1210–1220 cm-1 and 1670–1680 cm-1 bands. Monte Carlo simulations were performed using FTIR data as input, and the obtained results revealed that the metrics were highly stable. Coefficient of Variation (CV) values for total AUC remained below 0.03%, centroid variability was below 0.01% and spectral correlation coefficients were found 0.9999. Descriptor components showed variability at analytically acceptable levels (CV < 0.12%). The obtained findings demonstrate that quantitative FTIR band integration combined with simulation-based robustness evaluation provides a rapid and accessible analytical screening approach for pharmaceutical quality assurance. With this method, robustness evaluation of different drugs can be simulated using analytical data, and its applicability to other pharmaceuticals has been demonstrated.