Knowledge modeling for performance measurement of built environment

Buildings undergo frequent changes, but very little is known about the effect of these changes on building’s performance. As a case study, an underground station in the Netherlands is considered. In May 2000 a questionnaire regarding the perception of public safety and comfort at Rijswijk underground train station was developed and distributed. In total 246 responses were used for knowledge modeling. Knowledge modeling was done using a neuro-fuzzy system, and sensitivity analysis was used to elicit knowledge from the knowledge base. Before July 2006 some design changes followed. In July 2006 the same questionnaire was handed out to measure the effect of the changes. In this paper, first a brief description of the station and of the studied aspects is provided. Thereafter, the results from the year 2000 and 2006 will be compared and the effect of the changes on the overall perception of public safety and comfort will be highlighted.