English (UK)


We cannot change what was, but we can still control what is to come. Whether it's the future of climate change, the future of a company, the future of a home, or the future of all the systems we maintain, it is important to know what the future should look like. When it comes to the systems we install, we all agree that they should work in the long term. The magic word is predictive maintenance. It makes it possible to anticipate necessary maintenance work. This saves costs, time, and spare parts and ensures the continuous functioning of the machines and systems.

What is Predictive Maintenance?

Predictive maintenance is a proactive upkeep process based on permanent monitoring and analysis of machine and process data. Its purpose is to predict future maintenance needs, thereby avoiding interruptions and increasing the efficiency of the maintenance process. The condition of machines and plants is determined by real-time analyses and in combination with Big Data. In combination with other information, the aim is to predict  the best time to carry out maintenance work. Ideally, a maintenance technician repairs the equipment before the failure occurs - but only when necessary. In this way, costs are saved compared to preventive maintenance, where regular servicing is carried out.

Prevention instead of Repair: Why Predictive Maintenance?

To make maintenance calculations as accurate as possible, a large amount of data is needed. Therefore, predictive maintenance programmes are particularly worthwhile for companies that use many machines of the same kind or for manufacturers of these machines who want to use predictive maintenance not only for their machines but also for the machines they sell. Manufacturers and operators of machines and plants can achieve numerous advantages through the correct use of Predictive Maintenance::

  • Reduction of downtime
  •        Reduction of unplanned downtime 
  •        Prolonging the service life of machinery and equipment
  •        Calculate optimal maintenance time
  •        Avoid unnecessary routine maintenance
  •        Improved scheduling of maintenance and service technicians
  •        Efficient spare parts management
  •        Improved productivity and performance


The implementation: How does Predictive Maintenance work in practice?

To begin with, it is crucial to prioritise the machines and systems where failures are frequent and whose failing causes high costs. These machines must then be equipped with the appropriate sensors and networked with each other. The sooner this happens, the more current and historical data can be collected and evaluated either periodically or continuously. This data includes condition data (e.g. temperature) and static properties (e.g. manufacturer date) or event-related data, such as repairs or other service data.

A central system that provides a 360° view of the machines and plants is needed to store this data.

The processing of the collected data is then done by cleaning and deleting wrong values. Missing values are added.

The now cleaned data set is analysed and interpreted, e.g. with the help of machine learning algorithms, so that correlations can be derived. Among others, the following questions are addressed::

  •        Which measured variables are relevant for the machine?
  •        Which status data provide information about which component could fail in the foreseeable future?
  •        Which thresholds are relevant for which data types?


Once these questions are answered, necessary maintenance and spare parts can be planned foresighted.

The longer a predictive maintenance algorithm is in use, the more it learns and the more valid statements it can make. Predictive maintenance should therefore be seen as a long-term maintenance strategy.

Predictive Maintenance at Salvia

Why have we chosen Predictive Maintenance as a service? - Because maintaining the satisfaction of our customers and the preservation of the installed systems is close to our hearts. Ultimately, predictive maintenance goes beyond its maintenance task. Through this service, we not only help our customers but also extend the life cycle of the installed equipment and thus protect the environment.