Hilscher, IBM, X-Integrate
Machine learning you can touch
Hilscher, IBM and X-Integrate showcased an intelligent cloud solution for automated quality management at SPS IPC Drives 2017 in Hall 2.
An advanced predictive maintenance solution that recognizes quality problems in advance in order to plan maintenance work, identify assembly errors or even stop production processes automatically - this is the task that Hilscher, IBM and X-Integrate have set themselves. The demo application is based on sensors that record digital data from machines and systems during production. The key figures are evaluated decentrally at the production machine using edge computing. This reduces latency times and the load on the networks and central IT systems in the cloud. The netIOT Edge Gateway from Hilscher is used to transfer data between the edge and the cloud.
The Industry 4.0 application can be used with machines that are integrated into the production lines of car manufacturers, for example. Equipped with sensors, the machines can measure operationally relevant machine key figures in real time, evaluate them via edge computing and transmit them to the IBM Cloud. There, the data is further processed by the IBM statistics solution SPSS. These results can be used to make further predictions as to when set tolerance values are likely to be exceeded and how this will affect product quality. This real-time analysis in combination with IBM business process and decision management solutions enables different scenarios for quality assurance, from maintenance and adjustment of production capacities to production stops.
Secure database for machine data
The IBM database solution Informix is used for audit-proof documentation of all events. All machine data is recorded and evaluated decentrally, the relevant events are time-stamped and written to the database. This makes it possible to track at any time which decisions were made based on which system statuses at which time. By constantly comparing the calculated wear data with the actual situation, the mathematical model used is continuously refined and the prediction of quality is improved with increasing operation. The system was developed by IBM business partner X-Integrate. The company is a specialist in digitized processes and decision support based on established methodologies, open standards and IBM middleware.













