There has been a shift from the initial, somewhat idealistic, objective to monitor everything and anything on a machine. It makes a lot more sense to identify priorities and focus attention on optimising the machine learning algorithms to spot data pattern anomalies in critical areas first, identifying the areas with the fastest ROI.
Predictive maintenance based on AI offers additional advantages compared to traditional condition monitoring approaches. Increasingly data from the machinery can be merged with process data and evaluated using analysis models and cloud-based solutions. For example, AI can be used to detect deviations at an early stage from the normal state of your production machinery. This reduces unplanned downtime, lowers energy costs and increases efficiency. Reducing unplanned production stoppages directly also increases overall equipment effectiveness.
Delivering the future
Delivering this new vision for predictive maintenance is an interesting challenge. Today’s manufacturers of automation machinery need to look for new ways to support their customers in achieving predictive maintenance based on real time data mining. For companies within the automation technology sector, this requires the ability to merge mechatronics expertise with digital analytical solutions, which is no simple task and requires a whole new set of skills.
At Festo, experience has taught us that it is critical to an AI project's success that we not only provide software expertise, but we also have the knowledge to integrate it into the production environment and provide the experience to interpret the data in the terminology of the application. Our strategic acquisition of specialist software company Resolto has enabled us to fast-track this synthesis and develop a solution that makes predictive maintenance a seamless part of the machine.
Deeper data dives
As the benefits of data analysis become better understood, the need for greater insight grows. Festo has developed its offer in this regard from connectivity alone and visualisation through dashboards to additional tools that can help users discover new correlations in their data. For example, a root cause analysis can be automatically generated for every abnormal behaviour of the machine. It shows which sensors are crucial for the anomaly detection. The additional data visualisation enables users to get to the bottom of the anomalies and to recognise important correlations. Further customisation to match a specific application is also possible by setting individual parameters to optimise the algorithms integrated in Festo AX.
Festo can supply suggested default parameters, but customers can easily adjust these in the user interface without in-depth or data science knowledge. With the notification function, users can document, manage and forward anomaly detection outputs. When an anomaly is detected, a notification is generated including the following functionalities: data visualisation of the anomaly, automatic root cause analysis, diagnostic and classification tool.
Depending upon the application, customers can choose between On Premises, Cloud or Hybrid installations. With an On Premises solution, the learning of the models takes place within the customer’s local control installation. For cloud-based solutions the user can select their preferred environment or utilise space within the Festo Cloud. A hybrid solution can also be selected where complex computing tasks and large quantities of data are handled locally and only defined outputs and the algorithm model training itself are outsourced to scalable cloud installations.
One final but essential criterion for this type of predictive maintenance tool is that the data analysis is not limited to components and modules from the solution provider. The Festo AX software makes it possible to reliably analyse machinery incorporating components from many other manufacturers.
Final observations
The use of AI in predictive maintenance will continue to evolve extremely quickly. This in turn means that learning and experience needs to be acquired as early as possible to take advantage of the technology. For end-users, it doesn't make sense to try to justify big, all-encompassing block-buster installations. Instead, take an Agile approach and identify the quick wins with the quick paybacks. The most successful predictive maintenance projects based on AI and other emerging technologies have taken a staged approach – proposing and testing a hypothesis with a pilot evaluation and then upscaling from the learnings gained.
Essentially, the application of machine learning and data analysis using AI helps lift the fog within large data lakes and bears down on the areas with the fastest ROI. This enables manufacturers to pursue a prioritisation strategy to benefit from the quick wins and to work down the priority list with the payback examples already in-hand. Without a doubt this is an exciting development area and I look forward to seeing how it evolves in the next few years.