Abstract

The current volatility on the global market presents new challenges in production system engineering and control. On the one hand, packaging and packaging production systems must support increasing production volumes. On the other hand, market volatility requires more flexible production that is fragmented into smaller batches, which causes increased time, effort, and overall workload implied by changeovers between different production setups. Typically, it is not feasible to build a green-field generation of production sites that conforms to all design principles of Industry 4.0. Therefore, the addressed approach deals with a “smartification” of legacy hardware components, bridging multiple levels of the traditional automation pyramid, improving human-machine collaboration, and transforming human experience into a common shared knowledge base. In more detail, we are applying unsupervised learning methods to generalize individual machine settings into reusable and automatically applicable control setups and strategies. The core components of the addressed packing lines are counting machines that deliver the required portions of elements. The developed Smart Counting Machines retain current interfaces that are enhanced with new enhancements providing OPC UA communication and modular control layer. The realized solution has been deployed and tested in a real packing line of our industrial partner. The lessons learned in the presented mitigate barriers on the journey from the proof-of-concept phase to the full production deployment.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call