Towards Reducing the Complexity of Enterprise Architectures by Identifying Standard Variants Using Variability Mining

  • For decades, Enterprise Architectures (EAs) of car manufacturers have been constantly evolved to respond to growing requirements. As a consequence, EAs have often reached a very high level of complexity, which leads to problems in adapting EAs to new environmental condi­tions. Such a new condition is, for instance, digitalization of society (e.g., social media, Internet of Things) which has a huge effect on the automotive industry and the grown EA. Resulting changes in complex EAs have long implementation cycles, require enormous communica­tion efforts, and lead to high development costs. To alle­viate these problems, in this paper, we present a concept to reduce the complexity of grown EAs by adapting the Family Mining approach. This approach is originally used to compare block-oriented models, such as MATLAB/Si­mulink models, and to identify commonalities and diffe­rences between these models. In our concept, we utilize the Family Mining approach to analyze the variability of a particular EA and to identify the contained variants. All information about the variability and the variants will be used to derive standard variants representing default so­lutions for different issues. Using these standard variants, the existing EA will be restructured involving economic considerations (e.g., which standard variant yields best benefits under certain circumstances). Hence, applying this concept to a complex EA should allow reducing the complexity of the EA, alleviating related problems and making suitable design decisions for future extensions.

Download full text files

Export metadata

Additional Services

Share in Twitter Search Google Scholar Statistics
Metadaten
Author:Kenny Wehling, David Wille, Martin Pluchator, Ina Schaefer
URN:urn:nbn:de:kobv:526-opus4-5546
DOI:https://doi.org/10.15771/ASW_2016_6
Parent Title (German):1. Automobil Symposium Wildau: Tagungsband
Editor:Stefan Kubica, Hagen Ringshausen, Jörg Reiff-Stephan, Marius Schlingelhof
Document Type:Conference Proceeding
Language:English
Year of Publication:2016
Year of first Publication:2016
Publishing Institution:Technische Hochschule Wildau
Release Date:2016/03/31
First Page:37
Last Page:43
Dewey Decimal Classification:3 Sozialwissenschaften / 33 Wirtschaft / 338 Produktion
proceedings:Automobil Symposium Wildau / 1. Automobil Symposium Wildau
Licence (German):Creative Commons - CC BY-NC-ND 3.0 DE - Namensnennung - Nicht-kommerziell - Keine Bearbeitung 3.0 Deutschland