A data analysis framework for high-variety product lines in the industrial manufacturing domain
|Titel||A data analysis framework for high-variety product lines in the industrial manufacturing domain|
|Buchtitel||Proceedings of the 16th International Conference on Enterprise Information Systems (ICEIS 2014)|
Industrial manufacturing companies produce a variety of different products, which, despite their differences in function and application area, share common requirements regarding quality assurance and data analysis. The goal of the approach presented in this paper is to automatically generate Extract-Transform-Load (ETL) packages for semi-generic operational database schema. This process is guided by a descriptor table, which allows for identifying and filtering the required attributes and their values. Based on this description model, an ETL process is generated which first loads the data into an entity-attribute-value (EAV) model, then gets transformed into a pivoted model for analysis. The resulting analysis model can be used with standard business intelligence tools. The descriptor table used in the implementation can be substituted with any other nonrelational description language, as long as it has the same descriptive capabilities.