This dataset was used in previous research projects, for example in IMPROVE. It contains data of a new and worn cutting blade for comparisons.
The Vega shrink-wrapper from OCME is deployed in large production lines in the food and beverage industry.
The machine groups loose bottles or cans into set package sizes, wraps them in plastic film and then heat-shrinks the plastic film to combine them into a package.
The plastic film is fed into the machine from large spools and is then cut to the length needed to wrap the film around a pack of goods.
The cutting assembly is an important component of the machine to meet the high availability target.
Therefore, the blade needs to be set-up and maintained properly.
Furthermore, the blade can not be inspected visually during operation due to the blade being enclosed in a metal housing and its fast rotation speed.
Monitoring the cutting blades degradation will increase the machines reliability and reduce unexpected downtime caused by failed cuts.
The following image shows results from two snippets of data comparing a new blade to a completely worn out blade.
The figure shows the comparison from a new cutting blade and data from a worn out cutting blade, recorded on the same machine.
A small portion was randomly drawn from the training data to validate the learned model.
The deviations detected for the validation data should be similar to the training data.
The prediction with the worn blade shows a higher average error and especially the peaks in the error response, which happen when the cutting blade makes contact with the plastic film, are a good indication for the wear of the cutting blade.
Here you can see the component degradation over a Year.
For more information see reference below.
This dataset is publicly available for anyone to use under the following terms.
von Birgelen, Alexander; Buratti, Davide; Mager, Jens; Niggemann, Oliver: Self-Organizing Maps for Anomaly Localization and Predictive Maintenance in Cyber-Physical Production Systems. In: 51st CIRP Conference on Manufacturing Systems (CIRP CMS 2018) CIRP-CMS, May 2018.
Paper available open access: https://authors.elsevier.com/sd/article/S221282711830307X
IMPROVE has received funding from the European Union's Horizon 2020 research and innovation programme under Grant Agreement No. 678867
Your data will be in front of the world's largest data science community. What questions do you want to see answered?
