Edge Computing provides users with the opportunity to make optimum use of the huge volume of data produced by intelligent plants for both analytical purposes and the evolution of their plants. To achieve this, Edge Computing relocates the data processing from the cloud to closer to the data source – the plant itself.
Edge Computing thus provides new opportunities to use the data in production or manufacturing while IT-Capabilities and methods such as machine learning also move closer to the data sources. Siemens is already using the enormous potential of Edge Computing in its own manufacturing, for example, at the factory in Amberg, and is saving approximately 12,000 Euros in downtime costs per annum per equipped machine by analyzing data close to the process.
Up to a petabyte (one million gigabytes) of data is produced by an intelligent plant every day. This data is captured directly at the point of processing by Edge Computing through components distributed throughout the process, i.e. where the data is generated. Edge Computing allows the data and information from a range of sources to be brought together, analyzed, and translated into instructions.
Edge Computing has been used for a long time now at the Amberg plant. A circuit board cutting machine used on the Simatic production line creates dust during the cutting and milling process which can lead to the spindle bearing jamming and thus to machine downtime. By using Industrial Edge Computing to combine artificial intelligence (AI) and data analysis, both in the cloud and close to the process, it is possible to predict an imminent failure up to 36 hours in advance. Machine spindle changes can thus be scheduled, saving Siemens approximately 12,000 Euros per annum per machine in downtime costs.
Easy to integrate
The open solutions for Edge Computing can be easily integrated into existing automation landscapes and do not require any specialist IT knowledge to use. Siemens also offers a central management infrastructure to manage and administer the distributed Edge devices including mass software rollouts. One possible future solution is, for example, a Simatic TM NPU technology module for Simatic S7-1500 controls and ET 200MP I/O systems.
This unit uses a AI-enabled chip for efficient design of the neural network. In addition to the hardware, the portfolio also includes the so-called Edge apps such as the open source-based Node-RED app for implementing data capture workflows or the Inventory app for capturing built-in components according to their Hardware-Version, Firmware Version and ordering number. Furthermore, Siemens enables his customers with an open system to build own Edge applications based on high level programming languages.