The use of automated systems for monitoring or surveillance of the condition of machinery for an industrial plant is becoming more common. Variables or indicators are transmitted to a database connected "on line" with such systems, allowing us to correctly track the condition parameters included in the PdM (Predictive Maintenance) Program. A practical example of an installation of 500 measuring points in a paper mill is presented in this paper. Paper mill plants show a wide variety of defects and conditions related to problems in rotating machinery bearings, gears, low speed operation, variable speed machine due to harsh conditions in the environment for elements of the installation and other adverse factors. Therefore, the vibration analyses are presented as physical indicators of the technical system for this type of machinery, on which the maintenance performance metrics are built. The article concludes presenting the "state of the art in data transmission to remote facilities engineering, for predictive diagnostic in order to detect potential problems in machinery which facilitate decisionmaking process in maintenance departments. These systems allow the rapid construction of the KPIs at different hierarchical levels. Therefore an approach to decision making based on the Maintenance methodology has been developed as a result of technological advances in the collection of physical data