Predictive Maintenance
Smart Predict
AI-based failure prediction for the RAS Multibend-Center
Smart Predict analyses machine data in real time and detects critical trends before a failure occurs. This lets you respond early and avoid unplanned downtime.
What Smart Predict does
The platform combines continuous sensor data analysis with AI-based decision support, laying the foundation for proactive machine monitoring.
Anomaly detection
Unusual patterns in the machine data are detected early, before critical conditions arise.
Failure prediction
AI models calculate the probability of an impending failure and provide precise recommendations for action.
Sensor data analysis
Continuous, automated processing of 5 machine sensors in real time:
- Upper beam gearbox
- Folding beam gearbox
- Upper beam drive
- Main manipulator gearbox
- Hydraulic unit
More availability. Less downtime.
You get all critical machine data in one clear interface. Anomalies become visible immediately and can be assessed quickly.
More response time
Early warnings create room for targeted countermeasures long before a failure occurs.
Fewer unplanned failures
Critical trends are detected and can be addressed proactively instead of reacting after the damage is done.
Higher machine efficiency
Stable running times and optimised maintenance intervals increase overall equipment effectiveness (OEE).
Automated monitoring
Real-time dashboards show the current machine status at any time, without manual effort.
Freely selectable update interval:15, 30 or 60 minutes, depending on process requirements.
Five values that show the condition of your machine
For each position, Smart Predict shows the current values and the trend over the selected period. Limit values are marked in the chart.
Prevent downtime before it happens
Talk to us about how Smart Predict can be used on your Multibend-Center.
Optimised for machine operators
The Smart Predict interface is designed for everyday production. It enables quick orientation, even under time pressure and in complex production environments.
Intuitive user interface
Clearly structured views without unnecessary complexity.
Several machines at once
Parallel monitoring of the entire machine park in one view.
Live data at any time
Current machine values available at the push of a button.
Scalable for the future
Smart Predict is designed as an open, expandable platform. The solution grows with your requirements, from the first machine to fully networked production. Smart Predict is currently available for the Multibend-Center.
Additional machines
Smart Predict can be extended to further machines as required.
System integration
Can be connected to existing ERP, MES and quality systems as required.
Expansion
The algorithms are continuously trained and become more precise as the volume of data grows.
Predictive maintenance
Smart Predict provides the technological basis for a complete predictive maintenance strategy.
Request Smart Predict
Would you like to know how Smart Predict works on your Multibend-Center? Write to us and we will get back to you.
Frequently asked questions
Predictive maintenance is forward-looking maintenance based on machine data. Smart Predict analyses sensor data in real time, detects anomalies and uses AI models to calculate the probability of an impending failure.
Smart Predict is currently available for the Multibend-Center. The platform is expandable and can be extended to further machines as required.
Smart Predict analyses five measured values: temperature (heat), fatigue (material fatigue), crest (gear backlash), impact (shock load) and friction. For each position, the dashboard shows the current values and the trend over the selected period, with limit values marked in the chart.
The update interval is freely selectable: 15, 30 or 60 minutes, depending on process requirements. Current machine values can also be called up at any time at the push of a button.
With reactive maintenance, repairs only start once a fault has occurred and production has already stopped. Predictive maintenance detects anomalies early, so maintenance can be planned within a maintenance window and unplanned downtime is avoided.