The paint quality is among the most influential factor that the person will notice when looking at the car. Since painting process is one of the most complicated technological processes in car manufacturing it is necessary to have supporting procedures to evaluate paint quality. Despite the high degree of automation in production some evaluations and audits must still be done by employees due to the high technological difficulty of evaluation systems and versatility of employees. Therefore, we have focused on the differences between the evaluations of paint structure quality performed by the automated system in comparison with the paint structure quality evaluation performed by the paint auditors. The main aim of our proposal is the unification of paint quality audit results performed by the multiple paint auditors and automated system. The result is the proposal of Decision Making Platform that utilises the integrated data set from both evaluations and neural network for decision making support. Tableau software was used as interactive visualization tool for the obtained results. The proposed Decision Making Platform has been deployed in a real production process as a proof-of-concept and initial feedback results confirm the correctness and suitability of this solution.

Authors: Lukas Spendla, Michal Kebisek, Pavol Tanuska

Published in: World Congress on Industrial Control Systems Security (WCICSS-2020)

  • Date of Conference: 8-10 December 2020
  • DOI: 10.20533/WCICSS.2020.0001
  • ISBN: 978-1-913572-26-6
  • Conference Location: London, UK