One Computer Science MSc thesis to be presented on 2 April

Wednesday, 2 April there will be a master thesis presentation in Computer Science at Lund University, Faculty of Engineering.

The presentation will take place in E:4130 (Lucas).

Note to potential opponents: Register as an opponent to the presentation of your choice by sending an email to the examiner for that presentation (firstname.lastname@cs.lth.se). Do not forget to specify the presentation you register for! Note that the number of opponents may be limited (often to two), so you might be forced to choose another presentation if you register too late. Registrations are individual, just as the oppositions are! More instructions are found on this page.

11:15-12:00 in E:4130 (Lucas)

Presenter: Gustav Tindberg
Title: Deep Learning For Anomaly Detection in Manufacturing Equipment Research
Examiner: Jacek Malec
Supervisors: Marcus Klang (LTH), Robert Hansson (Tetra Pak Packaging Solutions AB)

In the manufacturing industry, equipments are designed for consistent and cyclic behavior, to perform the same task repeatedly with high precision over long lifespans. In line with the ideas of Industry 4.0 integrating digital technology and AI into manufacturing equipment this thesis proposes the PDA-RAE model which outperforms several comparable models for anomaly detection in cyclic manufacturing equipment tasks. The PDA-RAE model achieves an anomaly detection MCC score of $0.835$ and a reconstruction PSNR of $26.927$.

Additionally, manufacturing is an industry with highly experienced developers and operators on equipment that has been developed and used for decades. This provides an excellent opportunity to combine explainable AI methods with human domain expertise to extract new knowledge, increase confidence in the models predictions, and provide foundations for future equipment development. Using the PDA-RAE model in combination with experts has provided several new directions for future development of the equipment.

Link to popular science summary: To be uploaded