Jump label

Service navigation

Main navigation

You are here:

Sub navigation

  • News+

Main content

IPS presents two Data Science projects at the 10th International Conference on Applied Human Factors and Ergonomics (AHFE 2019) in Washington D.C.

Motion Capture and Image Mining in the context of Management of Productivity in Smart Manufacturing/Industry 4.0

IC-Applied Human Factors

As in 2018, the Institute of Production Systems participated in the AHFE 2019 with contributions and lectures entitled "Automatic generation of methods-time measurement analyses for assembly tasks from motion capture data using convolutional neuronal networks - A proof of concept" as well as "Automatic particle classification through deep learning approaches for increasing productivity in the technical cleanliness laboratory".

 

The first lecture describes the research hypothesis that motion data can be used to derive MTM analyses. In a first step, manual assembly work with motion detection systems is recorded and motion data is generated. This data is used to train a neural network, which aims to classify these movements as MTM.1 basic operations. This significantly reduces the effort that is required to generate MTM analyses.

 

The second article describes an alternative approach to the established metal/non-metal classification of particles with polarized light. By training a neural network with 70,000 images of metallic and non-metallic particles, comparable results can be achieved compared to the established approach of VDA 19.1. The advantage of this approach is the minimized sensitivity of the process to gloss and the adaptability of the classification model.

 

Detailed information on the presented contents can be found here.



Sub content

Contact

Secretariat IPS
Tel.: 0231 755-2652