Aurelio GHERSI, Department of Civil Engineering and Architecture, University of Catania

SEISMIC DESIGN OF IRREGULAR STRUCTURES FOR BUILDINGS: RESEARCH VERSUS CODE PROVISIONS
The influence of mass, stiffness and strength distribution in plan and elevation of seismic structures for buildings has been a topic of interest of the scientific research during the last fifty years. The basic criteria for definition and numerical evaluation of structural regularity were proposed and included in seismic codes in the last decades of XX century, basing on both research issues and practical considerations. Anyway, the main aspect considered by scientific literature of that period was the effectiveness of static analysis, common approach for seismic design, in predicting the results of modal analysis. Furthermore, in most cases the numerical models adopted during that period were extremely simplified and based essentially on the strong-beam weak-column model, which simplified the computational burden of numerical analysis. In the last decade of XX century and in the beginning of XXI century a large research effort was devoted to the definition of behaviour factor to be used in linear (modal) analysis for many structural typologies, but without particular attention to irregularity. Obviously, in these years the numerical models became more realistic and included the model of strong-column and weak-beam which now characterize the seismic design of framed structures. During the two decades of the XXI century, the topic of structural regularity has been analysed more widely, but the main goal become the influence of structural regularity in the comparison of the results obtained by different types of nonlinear static analysis to those provided by nonlinear dynamic analysis. Minor or no interest was devoted to a more appropriate definition of structural irregularity and to its correlation to the value to be used for the behaviour factor in linear analysis. Consequently, code provisions still use, for defining the value of behaviour factor, the criteria for regularity, in plan and in elevation, proposed in the past century for a totally different problem.