The study focuses on solving the current problem of planning the technological preparation of production (TPP) for large-scale products, that is characterized by high uncertainty, interdependence of stages, and competition for resources. Traditional deterministic methods, based on averaged estimates, often lead to schedule overruns and inefficient use of production capacity. The article proposes an approach to TPP optimization based on the integration of simulation modeling and virtual reality (VR) technologies. A comprehensive methodology has been developed, involving data collection and analysis by means of two methods: structured surveys of expert practitioners and controlled experiments in which participants performed technological design tasks in an interactive VR environment. The obtained empirical data on operation durations, failure probabilities, and logical dependencies served as the basis for building a detailed simulation model in the AnyLogic environment. The model architecture includes elements responsible for generating tasks, performing technological operations considering personnel availability, implementing feedback loops to account for design changes, and collecting resulting statistics. This structure allows for adequate reflection of the dynamics and stochasticity of the full TPP cycle within the adopted assumptions. The results of the simulation sessions were implemented to analyze the process time characteristics, the dynamics of queue formation and resolution, and the share of non-productive costs. The results demonstrate a significant positive effect from the use of VR tools at the technological preparation stage, expressed in the reduction of the total cycle time, decrease in the number of errors, and optimization of waiting time. This indicates an increase in the overall process efficiency, planning reliability, and improvement of resource utilization. The proposed approach has practical value for machine-building enterprises seeking to reduce time-to-market for new products and increase production flexibility.
Kudryavtsev et al. (Thu,) studied this question.