Multiple unit drug delivery systems are nowdays a very common research topic in pharmaceutical drug development and production technology. An example of such are pellets that offer many advantages with their small size and structural independence. There are many technological methods for preparing pellets. Pellets, which are discussed in our study, are manufactured by extrusion and spheronization, a multistage process whose product quality depends on a number of process and formulation parameters. The main goal of the diploma was the application of multivariate method as a statistical tool in Quality by Design (QbD) environment form determination of potential critical material and process parameters in commercial pellet production. For this purpose, a sessile drop method was used to determine the physical parameter of penetration rate of liquid medium into the solid substance of a model active ingredient, microcrystalline cellulose and κ-carrageenan that were entering under different batch numbers in the final dosage form, e.g. immediate release pellets. Experimental values were combined with the data from previous research in order to upgrade a multivariate model. This enabled us to identify potential critical process and formulation parameters with the impact on the pellets' dissolution rate of a model active ingredient. Multivariate data analysis was performed using a partial least squares regression method (PLSR), where all 117 studied process and formulation parameters were included in the X matrix of independent variables. Dissolution rate of the model active ingredient in 45th minute was our observed response Y (dependent variable). Results from the final model, developed by SIMCA P+ 14.0 computer program, showed that dissolution rate is influenced by 5 process and 4 formulation parameters. Their positive or negative impact on the response was interpreted with the help of theoretical explanations found in literature. Fitting (R2) and predictive ability (Q2) values indicated a model with a moderate quality. Further research on physicochemical properties of input materials and production process parameters could improve the quality of the model.As part of the QbD approach, the parameters identified from our model have been defined as Critical Process Parameters (CPP) and Critical Material Attributes (CMA) with a potential influence on Critical Quality Attributes (CQA), i.e. dissolution rate of pellets' active ingredient.