Three questions for Alexander Matt, Product Manager Prime Cube
The Industrial Internet of Things is generating more and more data that needs to be managed.
What do companies need to pay special attention to in this context?
Edge Computing is currently one of the most talked about trends in automation. It offers many new opportunities to increase productivity while being competitive and viable for the future.
As networking grows, the demands placed on machines, production lines and their manufacturers are increasing. The data world is becoming larger and more complex. However, the pressure to make optimal use of this data is also increasing. On the one hand, it needs to be collected and made available. On the other, and this is a totally different challenge, the data also has to be used efficiently and sensibly to add value for all stakeholders. It is important to work with a partner in systems development who has mastered the entire process and who can advise and support in terms of data utilisation.
The use of Edge Computing technology is the ideal solution in this regard. “Edge” is the term used to describe the periphery of a technical information network where the virtual and real worlds converge. In decentralised IT architecture, data that accumulates is not processed in the data centre, but rather directly at this intersection and, as required, moved to the cloud. Edge Computing enables real-time data pre-processing at precisely this juncture.
On the journey towards a well-functioning Edge Computing solution, there are many stumbling blocks to be avoided. In the preliminary stages, it is important to be clear about what added value you hope to derive from the use of data. Then, the required data can be determined. The next step is to decide whether data can be processed directly in Edge Computing – for example, for a virtual sensor or condition monitoring – and for which data the use of a cloud would provide a better solution. Predictive maintenance or complex data analyses are good examples. When using a cloud, the necessary data should be reduced to a minimum to avoid unnecessary costs while keeping the required bandwidth as low as possible. Once these questions have been answered, a good foundation exists for selecting the appropriate software and hardware resources.
When the conceptual questions have been addressed, the main focus of systems development is on the optimised interaction between hardware and software. But attention also needs to be paid to solution transparency, flexibility and scalability.
What can Edge Computing achieve in a modern manufacturing infrastructure?
A continuous, reliable flow of information and data is vital, especially when it comes to efficient manufacturing and greater productivity. In a modern infrastructure, data is generated and provided by various machines, lines and components from different manufacturers, which makes an Edge Computing solution especially challenging. The data generated from the control and process management level is used to evaluate and directly influence the relevant production processes and thereby increase production efficiency.
This is precisely where an Edge Computing solution comes in, positioning itself between a company’s production level (OT) and its IT to ensure reliable, efficient data usage across the entire line. It enables transparency in keeping with requirements in order to obtain an overview of data flows at all times and to protect proprietary know-how and with it, a significant competitive advantage. At the same time, Edge Computing provides enough data to enable manufacturer-supported maintenance and ensure line availability. It offers scalability and flexibility to respond to different influences and performance requirements.
Requirements that may seem self-evident. Yet, as of today, there is hardly any solution on the market that actually meets company needs for reliable data management and meaningful data use, and that can translate the above-mentioned challenges into a practically functioning solution.
What are the advantages of sensor-related evaluation/pre-processing of data?
By pre-processing the data directly at the sensor or edge, it is possible to respond to events in real time. This is indispensable when implementing a virtual sensor or an augmented reality application.
Another key advantage is the efficient use of data. By pre-processing the data, only relevant and resulting reduced data packets are transferred to the cloud. This leads to lower costs for data transfer and minimises the risk of its loss outside the company. Furthermore, know-how is better protected as critical data remains within the company.
The process-oriented analysis and processing of the data also simplify its further use for functions and services on site. For example, to use it in a virtual sensor, to implement condition monitoring or for quality assurance applications.
How can a coherent Edge Computing concept be realised with a service-oriented system architecture? The Multiservice Platform (MSP) of Schubert System Elektronik’s Prime Cube® product brand combines several services into a single system using container technology. The required data is collected via interfaces, then further processed via defined applications and transferred, for example, to the cloud or another destination. In addition to data pre-processing and cloud connectivity, other tasks can also be fulfilled in this manner, e.g. virtual sensor technology or remote access to the line.