Uncertainty Management for Automated Diagnostics of Production Machinery
Time: Fri 2020-06-12 14.00
Location: Publikt via ZOOM, (English)
Subject area: SRA - Production
Doctoral student: Károly Szipka , Industriell produktion, KTH IIP
Opponent: Assoc. Prof. Giuliano Bissacco, DTU Technical University of Denmark
Supervisor: Prof. Andreas Archenti, Design and Management of Manufacturing Systems, DMMS, Industriell produktion
Neither production machinery, nor production systems will ever become completely describable or predictable. This results in the continuous need for monitoring and diagnostics of such systems in order to manage related uncertainties. In advanced production systems uncertainty has to be the subject to a systematic management process to maintain machine health and improve performance. Automation of diagnostics can fundamentally improve this management process by providing an affordable and scalable information source. In this thesis, the important aspects of uncertainty management in production systems are established and serve as a basis for the composition of an uncertainty-based machine diagnostics framework. The proposed framework requires flexible, fast, integrated and automated diagnostics methods. An inertial measurement-based test method is presented in order to satisfy these requirements and enable automated measurements for diagnostics of production machinery. The gained insights and knowledge about production machine health and capability improve transparency, predictability and dependability of production machinery and production systems. These improvements lead to increased overall equipment effectiveness and higher level of sustainability in operation.