Abstract

Maintenance of chemical process plants is a complex topic. Data associated to its components is saved in many different files and data formats. This can be made easier by referencing the heterogeneous data through an Information Model. Via this model the information from all different data formats regarding a specific component can be queried. To provide the Model with the needed information, all data has to be assigned to its corresponding components before it can be stored. This is covered by an automated mapping algorithm based on name comparison. However, this algorithm cannot achieve a complete matching of all data objects. To fill this gap, in this thesis a prototype of a graphical application for manual matching is developed. It enables the user to assign data objects, which are from different files but provide information about the same component, to each other. This way, all data objects that were not assigned automatically are covered. An evaluation with industry experts affirmed the use of the resulting application. No comparable tool is being used in the industry yet. Thus, the developed prototype com- plements the automated matching and supports its applicability. Together, the programs present a new approach to providing a data model with the needed information and improve the way data of process plants is acquired.

Results/Implementation/Project Description

Conclusion

[ PDF (optional) ] 

[ Slides Kickoff/Final (optional)]