Premium
Expedition ins Datenreich: Wie kann man Informationen in Produkt‐ und Prozeßdaten umfassend erschließen?
Author(s) -
Estler Manfred,
Hilpert Ralf,
Soravia Sergio
Publication year - 1998
Publication title -
chemie ingenieur technik
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.365
H-Index - 36
eISSN - 1522-2640
pISSN - 0009-286X
DOI - 10.1002/cite.330701106
Subject(s) - computer science , product (mathematics) , process (computing) , data presentation , visualization , control (management) , data quality , quality (philosophy) , data acquisition , data management , data science , data mining , engineering , operations management , artificial intelligence , documentation , metric (unit) , philosophy , geometry , mathematics , epistemology , programming language , operating system
Exploring the Data World: How Can We Extract the Maximum Information from Product and Process Data? In various areas of the chemical industry a wealth of product and process data of various kinds is generated and recorded. In the majority of cases these data have only been evaluated fragmentarily and not systematically in terms of their information contained. Following the acquisition and possible short term visualization they end up in regular “data graveyards”. Furthermore, these data often stem from different sources such as distributed control systems (DCS), laboratory information and management systems (LIMS), or shift logs. There is a great potential for an improvement of product quality, process control, and plant safety, as well as for a reduction of inspection costs in the compilation and systematic analysis of these data. In addition to classical statistical methods, new data analysis techniques are increasingly used. After a brief explanation of the procedure used in data analysis projects and a presentation of several methods, an example illustrates which results can be obtained from an approach to data analysis that is not yet widespread in the process industry.