To overcome the problem of outlier data in the regression analysis for numerical-based damage spectra, the C4.5 decision tree learning algorithm is used to predict damage in reinforced concrete buildings in future earthquake scenarios. Reinforced concrete buildings are modelled as single-degree-of-freedom systems and various time-history nonlinear analyses are performed to create a dataset of damage indices.Subsequently, two decision trees are trained using the qualitative interpretations of those indices. The first decision tree determines whether damage occurs in an RC building. Consequently, the second decision tree predicts the severity of damage as repairable, beyond repair, or collapse.

A thorough four-step performance-based seismic evaluation for a six-story unreinforced masonry building is conducted. Incremental dynamic analysis is carried out using the applied element method to take advantage of its ability to simulate progressive collapse of the masonry structure including out-of-plane failure of the walls. The distribution of the structural responses and inters-tory drifts from the incremental dynamic analysis curves are used to develop both spectral-based (Sa) and displacement-based (interstory drift) fragility curves at three structural performance levels. The curves resulting from three-dimensional (3-D) analyses using unidirectional ground motions are combined using the weakest link theory to propose combined fragility curves. Finally, the mean annual frequencies of exceeding the three performance levels are calculated using the spectral acceleration values at four probability levels 2%, 5%, 10%, and 40% in 50 years. The method is shown to be useful for seismic vulnerability evaluations in regions where little observed damage data exists.

Ambient vibration measurements and 3-D nonlinear time-history numerical modeling are used to assess the retrofitting measures conducted in a 6-story unreinforced masonry building (URM) built in the end of the 19 th century in Switzerland. Retrofitting measures were taken in order to improve the soundproofing and possibly the seismic performance of the building. Reinforced concrete (RC) footings were added under the walls and horizontal steel beams were added to link the walls together with a RC slab at each floor, though the wooden beams were left in place. Several ambient vibration recordings were performed before, during and after the retrofitting work in order to monitor the evolution of the dynamic behavior of the structure. Moreover, numerical models representing the state of the building before and after the retrofit work have been developed to perform nonlinear dynamic analyses using various ground motion records. The change in the modal vibration frequencies, mode shapes, and failure mechanism are presented and discussed in further details. According to ambient vibration measurements, the performed retrofitting resulted in an increase of about 25% of the fundamental frequency. From the results of both the numerical modeling and the ambient vibration measurements, it is confirmed that the in-plane behavior of the slabs evolved from non-rigid floors with in-plane deformation to rigid floors with diaphragm effects. The ambient vibration measurements show that the new stiff slabs could lead to torsion behavior in the building as the result of the diaphragm effect and to higher seismic demand. However, the numerical models show that the displacement capacity of the building increases as a result of those new stiff slabs. Consequently, higher deformation capacity, indicated by the inter-story drift values, on average, are observed for all the damage grades in the post-retrofit state of the building. Finally, the overall seismic safety was only slightly improved.

In the context of the seismic vulnerability evaluation of buildings, the score assignment method can be used as the first step of a multiphase procedure aimed at identifying hazardous buildings that must then be analysed in greater detail. Because the existing Canadian rapid visual screening procedure has not been updated since 1992, a new procedure is proposed based on a set of vulnerability indices for different cities in the province of Quebec. A seismicity level (low, moderate, or high) is attributed to each city using the spectral acceleration response values included in the 2005 edition of National building code of Canada (NBCC) and the criteria proposed in FEMA 310 Handbook for the seismic evaluation of buildings - a prestandard. The structural vulnerability indices (SVIs) are calculated using the recently improved nonlinear static analysis procedure in FEMA 440, Improvement of nonlinear static seismic analysis procedures for each seismicity level. The NBCC 2005 reference soil class C is considered in the calculation of the SVIs, and index modifiers for the building height, irregularities, and design and construction year. The application of these indices to the estimation of the probable damage distribution in building inventories is discussed at the end.

Fling-step and forward directivity are the major consequences of near-fault ground motions as they can impose unexpected seismic demands on structures located in the vicinity of the fault. The pernicious effect of forward directivity on the seismic behavior of structures has been studied widely. However, not much research has been conducted to investigate the influence of fling-step that is related to a large co-seismic displacement. Moreover, the inconsistent results reported in the literature create a scientific challenge about the effect of fling-step on the seismic behavior of long-period structures. In this paper, the effect of fling-step is studied by comparing the displacement ductility demand in various single degree of freedom systems with different natural frequencies and strength reduction factors, subjected to long-period ground motions (generated and as-recorded) with and without fling-step. Subsequently, two 11and 20-story reinforced concrete buildings are considered and the effect of removing the fling-step on their maximum inter-story drifts is studied. The results indicate that the ratio of the fundamental period of the structure to the fling-pulse period plays an important role and the demands imposed on those systems without fling-step may increase or decrease based on the ground motions type and structural characteristics. Also, a similar trend in the displacement ductility demand was observed in this condition.

Abstract:As an approach to the problem of seismic vulnerability evaluation of existing buildings using the predicted vulnerability method, numerical models can be applied to define fragility curves of typical buildings which represent building classes. These curves can be then combined with the seismic hazard to calculate the seismic risk for a building class (or individual buildings). For some buildings types, mainly the unreinforced masonry structures, such fragility analysis is complicated and time consuming if a Finite Element-based method is used. The FEM model has to represent the structural geometry and relationships between different structural elements through element connectivity. Moreover, the FEM can face major challenges to represent large displacements and separations for progressive collapse simulations. Therefore, the Applied Element Method which combines the advantages of FEM with that of the Discrete Element Method in terms of accurately modelling a deformable continuum of discrete materials is used in this paper to perform the fragility analysis for unreinforced masonry buildings. To this end, a series of nonlinear dynamic analyses using the AEM has been performed for two unreinforced masonry buildings (a 6-storey stone masonry and a 4-storey brick masonry) using more than 50 ground motion records. Both in-plane and out-of-plane failure have been considered in the damage analysis. The distribution of the structural responses and inter-storey drifts are used to develop spectral-based fragility curves for the five European Macroseismic Scale damage grades.

No abstract

At White City in west London, the Hammersmith and City line runs on a high-level brick arch viaduct that crosses over both Wood Lane and the eastbound Central line. As part of the White City development, a section of this viaduct was removed and replaced with a new twin-track highly skewed bridge. This comprised a halfthrough bridge formed from three principal steel girders with connecting cross-beams and concrete infill. The deck was built off line, then slid into position during a four-day possession of the Hammersmith and City line, requiring part of the existing viaduct to be demolished; the Central line below remained open throughout. The piled substructure was constructed within the constraints of the existing viaduct arches. The existing early 1930s steel bridge deck over Wood Lane was strengthened in advance to provide new bearing positions. During the possession it was truncated by 2 m and supported on new piers.