Feeling the heartbeat of your machines and plants.
The DXQanalyze product family allows for a comprehensive logging of all available process data with the objective of identifying possible quality defects on the product or imminent wear on the equipment in real-time. On a superordinate level aggregated data enables the system to draw conclusions about the operation of individual steps along the value chain based on documented product quality. In the future, this information will be used to automatically adjust the process to counter these changing conditions. Products within our DXQanalyze product family use artificial intelligence, i.e. machine learning, in order to identify anomalies and derive patterns.
- Comprehensive analytics offering for all levels of data scientists (newcomers, advanced users, experts)
- Increased equipment / plant availability and first-run rate through faster troubleshooting
- Integrated domain knowledge in analytics solutions
DXQequipment.analytics provides a deep insight into various process steps and involved equipment along the value chain. The software package aims at improving all aspects of the overall equipment effectiveness (system performance, production quality, equipment availability). In a first step, DXQequipment.analytics supports faster troubleshooting with root-cause-analysis visualizing critical situations, detected patterns, and exceeded thresholds. Secondly, an automated analysis is enabled by providing a drag-and-drop analytics builder to create own algorithms. With the deployment of such algorithms data can automatically be analyzed and a direct feedback to the machine in real-time can be realized. In future, the Advanced Analytics module will use historical data and machine learning to find the optimal parametrization of algorithms and to detect long-term trends and patterns. This AI application combines information technology with engineering expertise, identifies fault sources and determines the optimal times for scheduled maintenance. It finds correlations in the plant and adapts the algorithm using a self learning approach.
In combination with DXQplant.analytics self-learning algorithms will automatically be trained to identify quality issues. Alongside DXQequipment.maintenance further information on detected maintenance tasks is provided.
DXQequipment.analytics is based on Dürr’s expert knowledge and can be offered for various equipment types, e.g. application robots, ovens, PT / EC systems.
- Real-time streaming analytics to ensure production quality
- Self-learning quality anomaly detection
- Easy-to-use frontends for data visualization and analytics model creation
- Permanent acquisition and analytics of equipment data
- Machine learning algorithms to evaluate the painting process and predict equipment failures
- Faster troubleshooting to increase equipment availability
- Increased first-run-rate
- Decreased equipment downtimes
- Optimized identification of root-causes
DXQplant.analytics aims at improving the first-run-rate of a production system. In a first step, quality dashboards and reports show key performance indicators in order to improve transparency of the product quality status within a quality loop. Secondly, systematic quality problems are detected using smart pattern recognition. In future, systematic quality problems will be correlated to process anomalies derived by DXQequipment.analytics enabling a root-cause-analysis as well as early troubleshooting.
- Dashboarding for quality relevant plant KPIs
- Smart pattern recognition for systematic quality defects
- Visualization of product life cycle of affected workpieces
- Indication of root-causes based on big data analytics and expert rules
- Improved first-run-rate to increase the OEE
- Structured overview of workpiece-related quality and process data
- Support in finding correlations between quality defects and process causes
- Domain knowledge linked with data analytics