The digital transformation in the industrial environment is continuously evolving. Anyone who wishes to be successful in this process relies on AI-based data analyses in their production in order to optimise processes automatically or to develop new digital business models.
Our AI solution Festo AX provides you with a wide range of features - cross-industry and highly scalable.
While the computationally intensive training of the models is carried out by Festo AX on the main component, the trained model evaluates the data directly on the edge component. This means even large and high-frequency data volumes are no problem.
- Better scalability
- Cost savings when using cloud-based services by minimising data transfer
- Latency minimisation
The choice of hardware for the edge component is completely up to you. There are minimal hardware and software requirements to ensure smooth operation.
Global Health & Festo AX visualisation tools
The trained digital model monitors, interprets and evaluates the input signals of your asset in real time. The results of the predictive maintenance and predictive quality analysis are available to you in a single visualised value: Global Health. An anomaly is triggered from a critical range that you can define, and a notification is sent.
Festo AX offers you further useful visualisation tools. Make trends, patterns and correlations in your data visible. Thus gaining new insights.
Data connectivity & data sets
Connect your assets via standard industrial protocols (e.g. MQTT or OPC-UA) by using an intuitive configuration. You can flexibly define where the data comes from and where the analysis results are sent. The data from your assets can be collected, visualised and analysed. You can use the collected data or data sets from other systems as a training basis for the AI. The data set manager is available to you for this purpose with all the necessary functionalities:
- Data visualisation
- Feature selection
- Data annotation
On Premises | Cloud | Hybrid
The modular structure of Festo AX allows great flexibility in the software architecture:
- On Premises: In an on premises solution, the model training takes place on one of your local computing instances. The instance communicates with the local component at the plant or on the asset and provides information for user interfaces.
- Cloud: The main component of Festo AX runs in a cloud of your choice.
- Hybrid: The Festo AX components are operated on premises as well as in the cloud. In doing so, for example, complex computing tasks performed by the locally installed component for model training can be outsourced to scalable cloud instances.
Festo AX automatically detects anomalies in your assets and makes them available for manual diagnosis and classification. Detected anomalies can, for example, be classified as a malfunction or maintenance requirement. The additional integration of the experience and knowledge of your technicians, engineers and process managers creates a constantly growing knowledge base for the Festo AX algorithms.
The notification function enables you to document, archive and manage the most important activities and forward them for example by e-mail. Notifications of anomalies are particularly important: Every time an anomaly is detected, a notification including the following functionalities is generated:
- Data visualisation of the anomaly
- Automatic root cause analysis
- Diagnostic and classification tool
Parameterisable artificial intelligence
You can optimise the algorithms integrated in Festo AX for your application by setting individual parameters. We supply the default parameters that are optimally pre-configured for you. You can easily adjust these yourself in the Festo AX user interface as required. No in-depth data science knowledge is required for you to do so.
Root cause analysis
A root cause analysis is automatically generated for every abnormal behaviour of your system. It shows which sensors are decisive for the anomaly detections. The additional visualisation of the data helps you to get to the bottom of the anomalies and to recognise important correlations.