The paper defines the patterns of stochastic scattering of parameters of casting processes, the determination of which is a necessary condition for the creation of robust casting technologies and methods for controlling the processes of obtaining high-quality castings. It is shown that an effective quality management system must meet two requirements: firstly, it must take into account the presence of a random scatter of quality parameters, and, secondly, it is necessary to quantitatively take into account the patterns of dispersion of controlled technology parameters and quality characteristics of castings. The work aims to study the patterns of stochastic fluctuations and distribution functions of casting technology parameters and determine their influence on the quality of castings based on extensive production and technological observations. To adequately describe distribution functions, universal systems of distribution functions of random variables with wide possibilities are considered. These are systems of distribution functions of Pearson and Johnson, which cover all cases encountered in practice. The possibility of using these systems to describe foundry processes was studied based on a numerical study of the characteristics of empirical distributions. It is shown that according to the meaningful meaning and accuracy of describing the spread of the characteristics of technological processes and parameters of castings and cast parts, the Johnson function Sb is most suitable. The results obtained can be used to improve the efficiency of problem-solving in casting quality control systems.