Title
Data-and Expert-Driven Analysis of Cause-Effect Relationships in the Production of Lithium-Ion Batteries
Abstract
The development of lithium-ion batteries (LIBs) is characterized by a unique level of complexity in the manufacturing process. In particular, cause-effect relationships (CERs) between process parameters have a strong influence on the quality of a manufactured cell and thus on the ramp-up time. First approaches for discovery CERs in LIBs were expert-based and thus afflicted with a high degree of uncertainty. Therefore, data from a real battery production line has for the first time been systematically processed and analyzed using CRISP-DM. However, the approach shows shortcomings in the involvement of domain expert knowledge as well as in the accuracy of the applied models. Addressing these shortcomings, an interdisciplinary data analytics framework is presented using human-computer interaction (HCI). Moreover, the framework aims to improve data analysis with the help of expert knowledge and, conversely, sharpen the knowledge of experts through data analysis. Thus, the model provides a basis for automated fault detection, diagnostics, and prognostics. Implementation and validation of the framework was conducted using the data of an assembly line for prismatic LIBs at the BMW Group in Munich.
Year
DOI
Venue
2019
10.1109/COASE.2019.8843185
2019 IEEE 15th International Conference on Automation Science and Engineering (CASE)
Keywords
Field
DocType
manufacturing process,cause-effect relationships,process parameters,manufactured cell,battery production line,domain expert knowledge,interdisciplinary data analytics framework,lithium-ion batteries,LIB,data-and expert-driven analysis,CER,CRISP-DM,HCI,human-computer interaction,automated fault detection,BMW Group,Munich
Data mining,Data modeling,Prognostics,Data analysis,Fault detection and isolation,Subject-matter expert,Computer science,Production line,Battery (electricity),Manufacturing process
Conference
ISSN
ISBN
Citations 
2161-8070
978-1-7281-0357-0
0
PageRank 
References 
Authors
0.34
0
5
Name
Order
Citations
PageRank
Thomas Komas100.34
Rüdiger Daub200.34
Muhammad Zeeshan Karamat300.34
Sebastian Thiede473.49
Christoph Herrmann537344.21