|Bibliography||Kassner, Laura; Gröger, Christoph; Mitschang, Bernhard; Westkämper, Engelbert: Product Life Cycle Analytics - Next Generation Data Analytics on Structured and Unstructured Data. |
In: Proceedings of the 9th CIRP Conference on Intelligent Computation in Manufacturing Engineering - CIRP ICME '14.
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology.
pp. 1-6, english.
Naples: Elsevier, July 23, 2014.
Article in Proceedings (Conference Paper).
|Corporation||CIRP Conference on Intelligent Computation in Manufacturing Engineering|
|CR-Schema||H.3.1 (Content Analysis and Indexing)|
H.3.4 (Information Storage and Retrieval Systems and Software)
J.2 (Physical Sciences and Engineering)
J.6 (Computer-Aided Engineering)
|Keywords||analytics, big data, unstructured data, text analytics, product life cycle management, PLM, data warehousing, product life cycle analytics, data integration|
Enormous amounts of unstructured data, e. g., emails, failure reports and customer complaints, are abundant around the product life cycle and provide a huge potential for analytics-driven optimization. However, existing analytics approaches on unstructured data are fraught with three major insufficiencies limiting comprehensive business improvement: (1) they focus on isolated data sources from a single life cycle phase â€“ for example, data from a customer relationship management system are mined for frequent complaints without considering manufacturing failure reports related to the same product; (2) they do not make use of structured data for holistic analytics, e. g., to automatically correlate unstructured failure reports with structured performance data of a manufacturing execution system; (3) existing implementations of data integration and analytics components are typically cost-intensive, manual and case-based, without a general framework.
To address these issues, we present our Product Life Cycle Analytics (PLCA) approach, a platform and a reference architecture for the holistic integration and analysis of unstructured and structured data from multiple data sources around the product life cycle. For this purpose, we survey structured and unstructured data sources around the product life cycle and discuss limitations of existing analytics approaches like traditional Business Intelligence applications. Moreover, we develop use cases for holistic life-cycle-oriented analytics and give examples based on case study investigations, e. g., for the holistic analysis of unstructured failure reports in the automotive industry. On this basis, we discuss technical requirements and components of our reference architecture, such as a versatile, plug-and-play Natural Language Processing pipeline and mechanisms for linking structured and unstructured data in a holistic data warehouse. Finally, we analyse implementation issues and investigate underlying technologies from the areas of text analytics and data mining in order to evaluate our architecture with respect to the identified use cases.
|Contact||Per Mail an firstname.lastname@example.org |
|Department(s)||University of Stuttgart, Institute of Parallel and Distributed Systems, Applications of Parallel and Distributed Systems|
|Entry date||August 11, 2014|