The Formalization and Representation of Expert Knowledge with Visual Analytics to Improve Product Quality in Serial Manufacturing Processes
December 09th, 2021 | 11:00 pm | online
Advances in new technologies, such as the Internet of Things (IoT) and big data analytics are enabling a new generation of smart manufacturing processes. These technologies allow the efficient tracking of the quality of produced parts along every step in the manufacturing process.
Here, sophisticated measurement equipment forms interconnected IoT networks, where the “smartness” level depends to great extent on how organizations can leverage data-driven approaches to create value of resulting big datasets and to support human experts with their high-cognition analysis tasks. In this regard, Visual Analytics (VA) can help to integrate human experts into the visual exploration of IoT data by augmenting analytical reasoning capabilities together with model-supported and custom-designed visualization interfaces. Thus, in this thesis, we analyze the role of VA in serial manufacturing processes. Specifically, we aim to examine how such visualization approaches can be used to formalize and represent expert knowledge to make it readily accessible for the development of data-driven approaches, which show great potential in improving serial manufacturing processes. In doing so, we present results from four design study projects in collaboration with a German manufacturer of electric vehicles. Our results indicate that VA shows great potential in augmenting human analytical reasoning processes of knowledge-intensive tasks. Furthermore, we outline how features, models, and especially labels can be leveraged by organizations as explicit knowledge products to be shared among organizational members and leveraged to engage in sophisticated data-driven approaches.
About Joscha Eirich
Joscha Eirich holds two bachelor's and master's degrees in business administration and information systems from the University of Bamberg.
At the moment he is working on his PhD at BMW Group. His research activities focus on the integration of domain knowledge into machine learning pipelines as well as visual analytics for the analysis of large-scale IoT data from serial manufacturing processes.
tugraz.webex.com/tugraz/j.php
Thursday, Dec 09, 2021 11:00 am | 1 hour | (UTC+02:00) Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna Meeting number: 2733 247 3401
Password: UJhxpqJ4M28
Join by video system
Dial 27332473401@tugraz.webex.com
You can also dial 62.109.219.4 and enter your meeting number.
Join by phone
+43-720-815221 Austria Toll
+44-20-3478-5289 United Kingdom Toll
Access code: 273 324 73401