|
6A - LCIA - Toxicity/RA (WEP/221) The Data Quality Foundation in OMNIITOX Information System. Carlson, Raul1, Erixon, Maria 1, Pålsson, Ann-Christin1, 1 Industrial Environmental Informatics, Chalmers University of Technology, Göteborg, Göteborg, Sweden ABSTRACT- This poster presents the data quality foundation, i.e. the definition, requirements, and review process applied in OMNIITOX information system. The system contains information about characterization models and the data input to the models, e.g. substances and their properties. The information is acquired manually by OMNIITOX participants or automatically from databases. The imported information is reviewed by the information system developers in the OMNIITOX-project, based on fundamental data quality dimensions, explicit data quality requirements, and according to a well defined review process. The information system developers are responsible for the concept analysis and synthesis in the OMNIITOX-project. OMNIITOX concept model supports the data quality as in the fundamental dimensions reliability, accessibility, and relevance. It specifies which data that is relevant for the information system and application, structures all information needed to describe characterisation models transparently, and provide one common language and communication format. The toxicologists and modellers in the project are responsible for deciding which characterisation models and toxic substances that are relevant to include, and what precision that is needed for the application. The data quality requirements applied in the documentation and review process is to make correct interpretations of the data documentation format, to address all agreed data documentation fields in the format, and to produce understandable information. Review reports are written in the review for all data received by the information system developers. The reports comments on the data quality requirements described above, the substance property specifications, i.e. that correct substance property specifications are applied, and the import file format, i.e. that the excel template is correctly formatted. All corrections are made by the data suppliers and the updated data is sent to the information system administration for a new review, in an iterative process. Conclusions of the work so far with the data quality dimensions and requirements in the OMNIITOX project are that they are comprehensible and sufficient. The review process is time consuming, but improves considerably after education and practice. Key words: data quality, OMNIITOX, data review, information system |
