YUNA-Module:
Condition Monitoring &
Predictive Maintenance
„With YUNA eoda enables us to fix problems before they actually occur. Thanks to real-time data evaluation, we reduce downtime, optimize processes and increase machine availability at the same time“
– Marco Holzer |
Head of Product Management & Logistics Services | TRUMPF
With the combination of AI and expertise, YUNA enables you to take the step from condition monitoring and continuous monitoring to true predictive maintenance. Using state-of-the-art algorithms, YUNA analyzes not only sensor data, but the entire spectrum of your machine cosmos – from conditions, to system messages and production data, to information on individual components.
On this basis, machine failures and malfunctions can be detected, predicted and prevented at an early stage.
Avoid failures
YUNA identifies factors that can lead to failures and alerts early so that failures and downtime are minimized.
Cost reduction
Avoid unnecessary costs due to sudden breakdowns. Reduce the cost of maintenance visits through proactive maintenance.
Optimize
Increase machine availability and thus safeguard not only your production but also its quality at critical moments.
Starting point of digitization
As a central platform, YUNA enables the development and operation of far-reaching digital services in companies. Low-maintenance and with state-of-the-art capabilities.
YUNA – Discover the potential
- Anomaly detection for machine failure prediction
- Central database for all participants generates quick insights/li>
- Intuitive operation and strengthening of collaboration between different teams
- Rich integration and automation capabilities
- Generation of data (sub)sets for the development of new models by users
- Integrated workflow for designing, managing, and going live with new analytics
- Intelligent rights and role management
This video is currently only available in german
YUNA how it works
Based on historical machine, sensor and system data, AI models are trained in YUNA to create a generalist profile of the machines. Then, anomaly detection can be used to identify various factors that indicate machine failure. Not only does YUNA identify these anomalies, but the models themselves can be further optimized in productive use by YUNA so that they become increasingly accurate. Even in the case of sudden or completely new situations, reliable operation is thus possible.
Fast implementation
From initial planning to reliable use as a condition monitoring or predictive maintenance system, we and YUNA accompany you every step of the way.
Assessment and model development
Together, we survey the existing database, discover potentials and provide support for possible processing. Here we collect the relevant information and then train the machine learning models.
Test run and initial operation
Together, we identify various factors and limit values for classifying harmless machine states and possible causes. Thanks to the full-managed workflow in YUNA, various test runs can be parallelized and transferred directly to the production system after final testing.
Automation and optimization
YUNA offers various possibilities to automate numerous actions. Send reports, let YUNA directly order maintenance when anomalies occur or give your employees the possibility to design new analyses or optimize existing ones.
Assessment and model development
Together, we survey the existing database, discover potentials and provide support for possible processing. Here we collect the relevant information and then train the machine learning models.
Test run and commissioning
Together, we identify various factors and limit values for classifying harmless machine states and possible causes. Thanks to the full-managed workflow in YUNA, various test runs can be parallelized and transferred directly to the production system after final testing.
Automation and optimization
YUNA offers various possibilities to automate numerous actions. Send reports, let YUNA directly order maintenance when anomalies occur or give your employees the possibility to design new analyses or optimize existing ones.
Deployment
YUNA offers two options: On Premise and Cloud.
On premise
Easily expand your tech stack by leveraging YUNA with applications and structures already in use.
- Perfectly integrated into your infrastructure
- Directly operable
- Use also as additional authentication layer
- No administrative overhead
- Unlimited scope of service
- Fast connection of data pools
Cloud
The easiest way to quickly deploy YUNA. Use already existing cloud structures and benefit from the flexibility they offer.
- Complete control over instances
- Directly operable
- Low latency
- No administrative overhead
- Unlimited scope of services
- High availability
Case Study:
Predictive maintenance – TRUMPF Lasertechnik
With YUNA, we at TRUMPF Lasertechnik have set up a condition monitoring portal that predicts machine failures and serves as the basis for further functions in the area of predictive maintenance, as well as enabling a new service model.
Case Study:
Smart rotary indexing table with condition monitoring and predictive analytics for WEISS
With YUNA and the use of cluster/density-based methods to identify iO and niO states, we have helped WEISS to further increase the reliability of rotary tables for customers.
Case Study:
Predictive maintenance for new customer services: Schenck Process Europe GmbH
Using algorithms to detect changes in the condition of the machines: Derivation of a "health indicator" based on sensor data that is comprehensible for users.
Solution:
YUNA for Machine Learning
The new way turning ideas into data science projects Quickly capture, reuse structured components, work collaboratively with internal and external parties – while maintaining compliance and performance.
Jump start now – we will walk you through it!