| Abstract [eng] |
The master’s thesis examines the possibilities of applying smart technologies to equipment maintenance in order to reduce downtime caused by failures and increase production stability. The relevance of the topic is driven by the growing complexity of automated production lines and the need to restore their operation quickly under conditions of skilled labour shortage. The novelty of the research lies in the practical testing of an artificial intelligence-based conversational system in the maintenance department of a manufacturing company. This system integrates equipment documentation, failure history and technicians’ accumulated knowledge and provides real-time support for repair decisions. The object of the research is the work of the case company’s maintenance department when using artificial intelligence tools for equipment failure diagnostics and troubleshooting. The aim of the project is to assess the potential of artificial intelligence tools in maintenance activities in order to increase efficiency in failure diagnosis and repair processes. The theoretical part reviews maintenance department performance indicators, methods for assessing equipment reliability, and the challenges of knowledge transfer and onboarding of new technicians, which can be addressed through digital knowledge bases and intelligent support systems. The empirical research is based on a single-case study. Data from the computerised maintenance management system were analysed, including production volumes, failure frequency, mean time to repair and mean time between failures, as well as technicians’ work with the artificial intelligence tool in daily practice. The results show that, as production volumes increased, the artificial intelligence solution helped to keep failure levels stable, shorten the diagnostic stage and reduce the average repair time, which in turn lowered equipment downtime costs. The economic analysis indicates that, even under conservative assumptions about the contribution of artificial intelligence, the costs of developing and maintaining the system can be recovered within a relatively short period. The findings are limited by the focus on a single shop floor and a short observation period, but they indicate directions for further research by extending the solution to different equipment groups and assessing its long-term impact on equipment reliability. The thesis concludes that artificial intelligence can become an effective support tool for maintenance technicians if it is integrated into a clearly defined failure management process, with high-quality data, well-maintained documentation and continuous development of employee competences. |
