IWS - Maintenance tool with algorithm feel
Maintaining production equipment is an often under-estimated cost factor. Chaotic warehouses and huge maintenance crews can result in wasted resources. Moreover, rigid maintenance planning according to fixed maintenance intervals based on manufacturer’s specifications for the manufacturer’s are hardly able to react to unexpected incidences. Manufacturing efficiency suffers from the negative impact on equipment availability, temporary storage and raw materials consumption.
Big Data helps the system learn
The IWS project has verified both the technical and economical feasibility of IPN’s intelligent maintenance planning tool. Using this real-time predictive method, the maintenance intervals of industrial machinery can be defined based on the actual future state of the machinery. The maintenance tasks can be integrated into production flows.
Adaptive algorithms able to recognize patterns and automatically detect the slightest changes in the equipment during production are responsible for this. They autonomously detect the factors relevant to the future state of the machines. The maintenance planning system continuously learns from the sensor and operational data streams fed to it from the company. Pattern recognition also takes into account intraplant expertise, securing it for the company in the long-term.
“Big Data” refers to the large volume, high frequency and highly unstructured data streams supplied. To ensure that the information is analyzed efficiently, a form of “aging processes,” in combination with bundling into statistical data, is applied to the existing data.
All results serve as input for downstream planning and monitoring systems, and can be displayed on standard PCs and mobile devices. The intelligent maintenance planning tool replaces costly, external maintenance experts and cuts companies’ personnel expenses. IPN has already brought in a large, Austrian industrial company that wants to offer Big Data applications for Industry 4.0 as a purely Austrian package.