SmarterMaintenance
The content on this page was translated automatically.
Initial situation
Nowadays, increasing demands on the reliability of production lines also place high demands on the maintenance of machines and systems. Due to the high cost of skilled labor, the lack of networking for problem-solving information and high material costs for spare parts, maintenance is the third-largest cost item in the steel industry, for example, and the cause of several days of production downtime every year.
Objective
As part of the industrial progress towards Industry 4.0, the new "SmarterMaintenance" analysis tool aims to fundamentally optimize the maintenance of process systems. By evaluating sensor data with complex data mining analysis, a self-learning correlation between measured values is to be detected and used to enable predictive maintenance. Furthermore, a text mining process based on a semantic model with keyword extraction and part-of-speech analysis realizes the automated evaluation and logical information linking of operating instructions, etc. The new tool should be able to reduce maintenance costs and downtimes in production lines by 30 % to 50 %.
As part of its own sub-project, the Department of Business Informatics is developing a text evaluation method based on a semantic model as well as keyword extraction and part-of-speech analysis methods. With the help of this algorithm, it will be possible to link machine problem information with corresponding solution information from electronic documents.
Promotion
The SmarterMaintenance project is funded by the Federal Ministry for Economic Affairs and Energy under the project management of the AiF.
Funding reference number: ZF4189701BZ5
Duration: 01.05.2016 - 31.10.2017
Project participants:
eoda GmbH(https://www.eoda.de/de/)
Contact person
- Dr. Philipp Bitzer
QuAALi