Design of Experiments (DOE) is an efficient tool to develop process understanding and to construct a design space. DOE varies the values of studied factors in parallel and the experiments are selected to adequately cover the entire range of operations. The DOE protocol demands statistical validity, so costly development can be restricted to significant results with great confidence. Pharmaceutical manufacturing typically involves a series of process steps, each with many variables that can have a major impact on critical responses. Principal Component Analysis (PCA) is to decompose multivariate data with correlated measurements into a new set of uncorrelated principal components. The PCA results are often presented with graphs plotting the projections of the observations onto the components as scores, and the projections of the measurement variables as loadings. The importance of each component is expressed by the proportion of the variance explained.

This paresentation utilized a pharmaceutical active ingredient (API) synthesis as an example to develop process understanding by DOE and PCA. The purpose of this study was to identify critical process variables and detect possible interactions between variables such as temperature, reagent concentration and solvent type, which would affect yield, purity, isomer ratio, and Raman measurement quality. Main effects and interactions were assessed by ANOVA, followed by Response Surface modeling to study the relationship between process variables and property responses. Principal component analysis was used to analyze the variability in the process and isomer product physical properties.