Predictive maintenance is one of the key aspects of Industry 4.0. The article presents the results of experimental tests of nitrogen purification filters in the installation of a low-pressure, metal processing device. The aim of the research was to develop a predictive algorithm for making decisions regarding the replacement of used filters, based on flow analysis and measurement of the pressure difference in front of and behind the tested filter. For the purposes of the research, a special test stand was constructed, which made it possible to determine the operating characteristics of three selected filters. Based on the tests carried out, the limit characteristics of the parameters measured were determined, identifying the need to replace filters in the gas installation.