Use of Pattern Recognition and Fuzzy Sets in Seismic Risk Analysis

The main topics of this research are: development of methods for determining vulnerability curves of structures under seismic loading, formulation of loss estimation procedures for specific structures and regions, use of pattern recognition and fuzzy set theory in the evaluation of seismic intensity and damage forecasting.

After a detailed review of currently available damage assessment models, vulnerability relationships are derived for both single and groups of structures. The derivation includes consideration of uncertainties of loading (demand) of structural resistance (capacity) and modeling. Quantifiable and nonquantifiable factors (such as construction quality and architectural details) which influence the performance of the structures are considered with probability and fuzzy set models, respectively.

Two methods of identifying earthquake intensity are presented and compared. The first method is based on the theory of pattern recognition where a discriminative function is developed based on Bayes' criterion. The second method applies the theory of fuzzy sets. These methods are used for the purpose of intensity classification after an earthquake and also for checking past classifications. Damage severity versus intensity relations consistent with the actual damage data and intensity classification are derived.