DescriptionCompression is the final unit operation in a pharmaceutical tablet manufacturing process scheme that produces the compact. Since compression determines some of the major critical quality attributes (CQA's) of tablets, such as hardness and disintegration time, understanding the effect of compression parameters on tablet quality is essential. The objective of this study is to develop a proof-of-concept methodology to correlate material properties to equipment and process performance using semi-empirical models, specifically compression models, and predict model coefficients. In this study, experiments involving some commonly used pharmaceutical ingredients such as lactose, microcrystalline cellulose, and acetaminophen was performed. The excipients were blended with varying levels of magnesium stearate ranging from 0.25 - 1.5% and the blends were characterized. The material properties measured for the blends were compressibility, permeability, cohesion, density, and particle size. Principal Component Analysis (PCA) was performed to understand the operating material design space. After tablet compaction, the compression data values were regressed to the unknown coefficients of the Kawakita compression model and the Kuentz hardness equation. The parity plots, R-Squared (R2) and RMSE values showed a good fit between experimental data and the model output obtained using the regressed coefficients. Partial Least Square (PLS) regression was performed using the regressed coefficient values to obtain a linear correlation between the regressed coefficients and the original blend material properties. The PLS model regression presented less than 10% error for most of the calibration points and a decent prediction of the model coefficients for the validation points. The results obtained indicate that correlations between material properties and semi-empirical model coefficients are feasible and it is possible to predict the response of model coefficients with decent accuracy. This work can be used as a basis to expand material property and process parameter correlations to semi-empirical models of other unit operations involved in pharmaceutical processing in the future.