Inconsistencies in Codified Procedures for Seismic Design of Masonry Buildings

Student: Paolo Morandi
Supervisors: Dr Guido Magenes

ABSTRACT

The recent development of a new Italian seismic code, thought as a transition towards the final adoption of Eurocodes, has been an occasion to reconsider the current criteria for seismic design of masonry buildings and their consequences on practice.

Inconsistencies in codified procedures for seismic design of masonry buildings were found I relation to the definition of the design seismic action, to the geometrical and construction requirements to the conditions for the applications of the structural analyses on masonry buildings.

Moreover, to verify the given provisions for seismic analysis and safety checks, several linear and non-linear analyses were carried out on different configurations of URM (unreinforced masonry) and RM (reinforced masonry) buildings. Inconsistencies in codified procedures of linear elastic analyses with the use of q-factor in comparison with the results of experimental tests, past-earthquake damages on buildings and non-linear analyses were found.

The behavior factor q has been redefined as a function of the results of the non linear static analyses through the introduction of the overstrength ratio (OSR). This new redefinition of the behavior factor q has allowed a substantial improvement in the safety verifications carried out with linear elastic methods, especially in comparison with the results of the non-linear analyses and of the experimental tests.

Nevertheless, the choice of a specific value of OSR, even for the same homogeneous typology of masonry buildings, does not overcome completely the intrinsic problems of the linear methods of analysis. Considering a homogeneous class of buildings whose OSR are nevertheless very scattered, the choice of a single conservative value, be it a "sufficiently conservative" percentile, has the consequence that in the wide majority of the cases, in which the OSR is much higher, the design seismic action will be much higher than it should.

A possible solution to these issues is the application of a redistribution of the internal forces after the linear analysis. If this procedure is carried out in a suitable way it can be very effective, provided the limits to redistribution are not as strict as currently required by EC8.

Finally, starting from the results of the non-linear static analyses on the structural configurations proposed, some conditions for the definition of the "simple building" and for the minimum areas of the masonry wall sections were revised, new requirements on the construction details were added and better defined to those already included in the OPCM3274. Moreover, some conditions for the applicability of the structural analyses and for the verifications were subjected to revision.

The revisions and the integrations proposed were included in the more recent version of the new Italian seismic structural code (OPCM 3431).

You may download a digital version of this MSc dissertation here.