The presented study shows the impact of changes in the content of continuing education programs on the assessment of the quality of courses by students in the center of additional vocational education. The developed system of evaluation and input and output questionnaires made it possible to identify point changes that need to be implemented to improve the quality of both the course attendees and the customers. The presented system of impact on the quality of professional development through continuous balancing of the thematic content of the program of additional professional education allows managing the development of the competencies of employees. Point changes in the proportions of the content blocks of the programs made it possible to achieve the optimal set of topics that satisfies the needs of the customer in terms of developing certain competences in the employees, as well as the course participants themselves in terms of interest in learning and personal growth. The developed approach makes it possible for the company of the customer to introduce a system of continuous professional training, advanced training and retraining of personnel by controlling the level of competence of trained employees to improve performance.

The problem of DEM contact models choice is described in the article. An analysis of the input data calibration, including a choosing the contact model, is performed. The main theoretical positions of the most popular models are considered, their comparison is made on the basis of a specially designed stand. Description and justification of the designed stand are considered. The analysis of the obtained results is carried out. Recommendations on the choice of the DEM contact model are given. The results show that it is time-efficient to use non-linear contact model for future model calibration.

The paper presents a methodology for training neural networks for vision tasks on synthe-sized data on the example of steel defect recognition in automated production control systems. The article describes the process of dataset procedural generation of steel slab defects with a symmetrical distribution. The results of training two neural networks Unet and Xception on a generated data grid and testing them on real data are presented. The performance of these neural networks was assessed using real data from the Severstal: Steel Defect Detection set. In both cases, the neural networks showed good results in the classification and segmentation of surface defects of steel workpieces in the image. Dice score on synthetic data reaches 0.62, and accuracy—0.81.