Advantage
Creating added value for a new service module based on data and algorithms
Data
Information about machine conditions in real time
Methods
Methods for determining structural changes
Challenge
Schenck Process is a global technology and market leader in applied measurement technology, offering innovative solutions for weighing, conveying, and screening, among other applications. To continuously develop its service offerings for its customers, Schenck Process plans to implement a use case in the area of predictive maintenance. This is particularly relevant for Schenck Process, as unplanned machine downtimes are associated with high costs and very complex repair processes.
Prior to the project, Schenck Process had a limited database of historical data. This included numerous records of machines in good condition, but only a few records associated with specific failure scenarios. Due to individual specifications, each machine manufactured is unique.
Goal
Schenck Process's goal is to detect changes in the condition of machines using algorithms. Specifically, this involves deriving a "health indicator" that makes the machine's condition transparent to users based on sensor data.
Solution
To create an adequate database, experts from Schenck Process defined the current machine states, thus laying the foundation for training the analysis models. As part of the data management, eoda also established equidistances by interpolating missing measurement periods.
To generate the "Health Indicator," eoda used methods for detecting structural changes based on regression models.
Key aspects in the design of the analysis models by eoda:
- Efficient retraining of the models: Easy transfer of the analyses to other machines.
- Use of real-time data: State changes in the production system can be continuously updated.
- Existing toolset: Schenck Process relies on the commercial software MATLAB. Therefore, the algorithms and methods used can naturally be reimplemented in MATLAB.
Result
With the "Health Indicator," eoda is creating added value for a new service module for Schenck Process. The indicator helps automatically and proactively detect machine failures and significantly improves understanding of machine health. The "Health Indicator" can therefore make a decisive contribution to increasing machine availability and thus customer satisfaction.
(Image source: Schenck Process Holding GmbH)
Starten Sie jetzt durch:
Wir freuen uns auf den Austausch mit Ihnen.
Ihr Experte rund um das Thema Data-Science-Projekte:
Lutz Mastmayer
projects@eoda.de
Tel. +49 561 87948-370
