Magnetron sputtering system (magnetron) by Inmodus licensed under https://creativecommons.org/licenses/by-sa/3.0/deed.en

Motivation

Research on applicability of machine learning in industry often focuses on specific applications (predictive maintenance, quality assurance, process control, etc.) and/or specific processes/process classes. One possible process class to investigate are vacuum deposition processes. These find application in a variety of promising industries and a wide range of products, such as semiconductors, solar cells, optical coatings, and many more.

Keywords
Machine learning, deep learning, time series, vacuum deposition, physical vapor deposition (PVD), thin-film coatings

Task

The goal of the thesis is to investigate the application of state-of-the art machine learning concepts and models on industrial time series data, esp. from vacuum deposition processes, for various use cases. Depending on the scope of the work (BT/PT/MT) and preferences, a possible assignment would be:

• Technical familiarization and description of the state of the art in research related to multivariate industrial time series data for product quality prediction and process control
• Identification and characterization of current attention mechanisms as well as evaluation of the potential applicability to industrial time series
• Development of a machine learning/deep learning model based on the insights gained from the previous familiarization process
• Evaluation of the model on the basis of various publicly available industrial time series data

Requirements profile and application information

• Interest in machine learning in an industrial environment, ideally already initial experience
• Highly motivated, perceptive and structured way of working as well as good communication skills
• IT affinity and good knowledge of at least one high-level programming language (ideally Python) desirable
• Start of work is possible immediately
• Scope and content can be individually agreed depending on the type of thesis (BT/PT/MT) and personal preferences
• Please send your application with CV and current subjects/grades by mail to the contact below

Categories:

Contact: