The »IoT Assistant« enables machine operators to use automated services for connecting devices, storing, and analyzing data. The services are run as temporary single purpose services, so unlike web services which run continuously, the IoT Assistant services only runs for the time of use. This enables a more efficient use of resources and a more flexible deployment.

This was made possible by using a microservice architecture. The increase in complexity with each service can be reduced and is made manageable by encapsulation in containers.

An important step, besides the implementation of the GUI and VPN access, was the move to the Fraunhofer Edge Cloud and a Kubernetes cluster. In addition to scalability and flexibility, this also offers a broad connection to the store floor, as the Kubernetes nodes can be complemented beyond the FEC, for example by IPCs or micro-controllers. The logical next step would be to automate the deployment of services and add the asset administration shell for better linkage. The »IoT Assistant« has potential to be widely used as infrastructure in other projects.