React at the right moment.
When it comes to maintenance, machines and plants are often subject to fixed cycles. This suggests reliability and safety. Experienced maintenance staff however know that: a fixed cycle never really meets the requirements of every day production. It doesn't react to actual demands or pressures. Fixed maintenance which always takes place at the same intervals can mean that components are often switched too often or even to late. Both increase costs. Predictive maintenance can direct maintenance efforts to the real requirements. You can thereby reduce costs and make production significantly smarter.
What is Predictive Maintenance?
Industry 4.0 and Smart Factory demand new ideas for improved efficiency. Predictive maintenance – forward thinking maintenance – is a future-oriented concept which can already be productively implemented today. Predictive maintenance offers a forecast into the future of a machine. Predictions are made when a machine should be serviced or a component switched based on data gained from experience and learning models. Forward thinking maintenance continuously collects data in running production and analyzes it. The system is thereby constantly learning and developing and allows interpretation of live data via a model. Rigorous maintenance cycles can thereby be replaced with individual service times for each machine and every component. The analysis and creation of the models can take place directly on-premise at the company or via a service in the Cloud.
Although predictive maintenance may not always be able to give 100% exact claims for the future, it can deliver very good indications for the most suitable time to carry out maintenance tasks.
zenon and Predictive Maintenance
zenon is very well suited – such as in connection with Azure Machine Learning from Microsoft – for a Predictive Maintenance System. The possible process:
- zenon collects sensor data in real time.
- zenon communicates with Azure Machine Learning or a similar application and also provides permanent storage of data.
- In the external application this data is used for machine learning. The system continually expands its experience and continues learning.
- The model construction takes place in the external application: When does which machine or component require which intervention based on the given data?
- zenon takes over the model and displays it in the SCADA system. The corresponding interaction is planned and communicated in the user interface.
zenon Predictive Maintenance: More flexible for improved competitiveness
Maintenance at the right moment offers a number of advantages compared to having a fixed cycle. Above all: the machine can be maintained according to your requirements. Increased loads will shorten maintenance periods and thereby prevent machine damage. Less load application will postpone the time of maintenance and thereby save unnecessary costs and downtimes. Spare parts can thereby be ordered at the right time and in the right quantity and engineering time can be optimally planned.
Predictive maintenance can be your perfect first step to the Smart Factory. Start with an E-Mail today to email@example.com. We're happy to share our know-how!