Optimization of production machine maintenance using predictive analytics and real time machine data visualization

The aim of this thesis is to evaluate the current effectiveness of monitoring machine conditions to prevent potential failures and downtimes, based on theoretical knowledge, and to propose an optimal solution for monitoring machine data and predictive measures for the machine maintenance department using descriptive analysis. The thesis is divided into a theoretical part and a practical part. The theoretical section focuses on defining the essence of machine maintenance, describing various approaches, their characteristics, and their application in practice. It also deals with tools used for monitoring machine operational data and outlines methods of predictive maintenance. Special attention is given to the use of descriptive analysis to support effective evaluation of equipment condition and the design of appropriate measures. An economic evaluation of the company is then carried out, its business direction is defined, and the main issue around machine maintenance is described, specifically the problem of insufficient monitoring of machine data. The issue, which is already partially addressed by an external company, is thoroughly analysed and specified to clearly highlight the deficiencies that are subsequently addressed through a proposed solution using the Power BI reporting tool. Based on the conducted descriptive data analysis, a predictive measure is proposed and thoroughly discussed with a focus on further development and optimization.