Observations of performance of structures during past earthquakes demonstrate that buildings with asymmetric floor plan tend to suffer greater damage because it is impractical to avoid such buildings with irregular distribution of stiffness and strength in plan and such buildings are especially vulnerable to earthquake damage, there is urgent need to develop an improved understanding of the earthquake response of these buildings. The objectives of this research project are to: (1) develop an improved understanding of the inelastic response of buildings undergoing torsional motions during earthquakes; (11) develop simplified analysis procedures suitable for preliminary design of such new buildings and for the design of retrofit schemes for existing buildings; and (III) propose--for consideration by code-writing entities like the SEAOC Seismology Committee--improved code provisions for accidental and natural torsion. The central feature of the research is to utilize recorded motions of well instrumented buildings during recent earthquakes to evaluate and calibrate analysis and design procedures in order to ensure that these procedures provide reliable estimates of torsional effects. The comprehensive but simple analysis and design developed should be of direct benefit to the profession. They would be useful for checking the preliminary design of new buildings, for evaluating alternatives to retrofit existing buildings with unacceptably large asymmetry in their plans, and for improving building codes.
