Methodical support for identifying and selecting data and analysis tools in the data-driven development of complex mechatronic systems

Publikationstyp
Conference Paper
Date Issued
2024
Sprache
English
Author(s)
Wagenmann, Steffen  
IPEK, Karlsruhe Institute of Technology
Rall, Jakob
Albstadt University of Applied Sciences Ravensburg-Weingarten
Weidinger, Felicia
IPEK, Karlsruhe Institute of Technology
Krause, Artur  
Smarte Entwicklung und Maschinenelemente M-19  
Sommer, Lutz  
Albstadt University of Applied Sciences Ravensburg-Weingarten
Bursac, Nikola  
Smarte Entwicklung und Maschinenelemente M-19  
Albers, Albert  
IPEK, Karlsruhe Institute of Technology
TORE-DOI
10.15480/882.13715
TORE-URI
https://hdl.handle.net/11420/52082
Journal
Procedia CIRP  
Volume
128
Start Page
114
End Page
119
Citation
Procedia CIRP 128: 114-119 (2024)
Contribution to Conference
34th CIRP Design Conference, CIRP 2024  
Publisher DOI
10.1016/j.procir.2024.04.002
Scopus ID
2-s2.0-85208785508
Publisher
Elsevier
Current research shows that analyzing usage data from reference systems holds a high potential for product development. The process model for data-driven validation of systems of objectives supports developers in performing such data analyses. However, developers still struggle in the process of identification and selection of data and analysis tools. Thus, this work aims to develop methodical support in this process. nine expert interviews are conducted from which 15 influencing factors on the process are identified. Based on this methodical support is derived. For successful process execution, three areas of expertise are identified that must be covered regarding development domain, data, and conducting analysis. Developers have differing skills and thus require distinct levels of assistance in different process stages. To guarantee support to as many developers as possible while ensuring applicability and efficiency, five personas are derived covering frequently occurring expertise patterns. Furthermore, metadata documentation is identified as a central lever to reduce complexity in handling data, which in turn decreases dependency on experts. Thus, a consumer-focused data catalog concept is developed. Activity profiles are used to guide developers through the process while providing support at different stages as needed. Finally, the evaluation of the solution by conducting workshops with 15 developers shows an improvement of 41.7 percent in task accomplishment.
Subjects
Data Analytics
Design methodology
Methodical Support
technologies
tools
DDC Class
004: Computer Sciences
005: Computer Programming, Programs, Data and Security
Publication version
publishedVersion
Lizenz
https://creativecommons.org/licenses/by-nc-nd/4.0/
Loading...
Thumbnail Image
Name

1-s2.0-S2212827124006565-main.pdf

Type

Main Article

Size

606.96 KB

Format

Adobe PDF