As a comprehensive methodology for quality improvement Six Sigma began its rapid development in the 1980s in Motorola, and today is successfully applied for improvements in the processes of large enterprise systems. The application of the Six Sigma methodology resulted in a high level of compliance an increased process capability, the reduction of costs and increased profitability as a fundamental feature of competitiveness. Considering the fact that the analysis of a theory and previous research observed the lacks of models for total quality control that would result in optimizing the printing process, and that the Six Sigma methodology has proved suitable for optimization in various production industries, the aim of this study was to create a new model for optimization of the printing process based on the principles of Six Sigma. This model allows continuous achievement of stable and predictable results, and reduces process variability. Design of a new model based upon the principles of Six Sigma rests upon the detection and grouping of possible points in the printing process (Critical To Quality), leading to variations in production, and thus the product itself. With regard to the fact that Six Sigma is a statistical measure of process functioning i.e. it is an indicator of variations in the process, and it is the basic intention of measuring the variability of the process, the monitoring the changes in the value of sigma has served as an indicator of changes in the process. It was therefore necessary to define the limits of possible variation in the process, so functional requirements of the products were determined for this purpose, defining the methods of control and the critical characteristics of the products which was analyzed and grouped according to cause categories. Defined is also a maximum range of each characteristic determined by the lower and upper limit of tolerance, and critical to quality points. The first part of the research focuses on the analysis of the zero process to determine whether the process is in a state of statistical process control, and whether it is normally distributed and properly aligned with respect to a defined and observed characteristics of the product. In the analysis of the zero process the accent was set upon the calculation of the current level of sigma and preliminary process capability. In the second part of the research functional requirements of the products were defined, critical products characteristics and critical to quality points described, and 27 – pooled process analysis conducted. Finally, a new optimization model was designed based on the results of the research, as well as tested. The backbone of model design was to generate an equation where the generation function is taken as a function of three variables. The chosen method of least squares is established as a good approximation of the curve with deviation less than 1%.