The quality of monitoring data in civil engineering works

Abstract

This paper presents the research and discusses the findings concerning the identification and definition of quality dimensions related to the monitoring data obtained in the safety control of major civil engineering works, namely dams and bridges. The analysis of the behavior and the safety evaluation of these works essentially depend on the quality of the data collected by the sensors, named the observation data, besides the quality of the models and, last but not least, the knowledge and experience of the experts involved. In addition, LNEC1 stores and preserves the observation data concerning major civil engineering works, which represents an important research source for LNEC and Universities and, thus, shapes itself as an e-Science scenario. Differently from other e-Science environment authors, we have not restricted our work to objective dimensions, whose measures can be calculated automatically. It appears, from the experts’ opinions, that some subjective dimensions can largely enrich the quality information about archived data at the expense of some additional work in data curation. In the development of this research we used an online version of the Delphi method with the Q-Sort technique, complemented with upstream interviews and a downstream meeting with the experts. We found out, and sorted by the importance level assigned by the experts, ten dimensions for the quality of observation data, which are error-of-observation, coherence, relevancy, interpretability, timeliness, completeness, accessibility, appropriate amount of data, access security and preservation.