Uniform Hazard Spectra Research Articles

유사 성능 기기의 지진에 대한 거동특성 분석

Damage to nuclear power plants results in loss of life and environmental disasters. Safety-related equipment installed in the nuclear power plants are controlled by electrical control equipment. Safety-related equipment are required to have seismic performance. In this study, the seismic performance of the motor control center (MCC), an electrical control equipment, is tested. Therefore, a shaking table test was conducted using the design response spectrum and uniform hazard spectra. The equipment installed in the nuclear power plants are supplied by various manufacturers. Because the seismic performance of MCC varies depending on the design and manufacturing method, the shaking table test is conducted for two different MCCs with the same function. The dynamic characteristics and shapes of two different types of MCCs were compared, and the seismic performance according to the shape was examined. Additionally, the difference in seismic performance according to the structural shape was examined through seismic fragility tests.

A new framework for ground motion selection for structural seismic assessment

Performance-based structural seismic assessment generally requires the selection of ground motions wherein the target acceleration response spectrum (Sa) should be matched in a desired period range. The pros and cons of the commonly used uniform hazard spectrum (UHS), conditional mean spectrum (CMS) and conditional spectrum (CS) are thoroughly examined. Aiming at combining the merits and overcoming the challenges of these methods, a new framework is proposed to construct a series of conditional target spectra for selecting ground motions. The response spectrum is expressed as the product of peak ground acceleration (PGA) and the normalized spectra (β = Sa/PGA) which can be smoothed by four spectral shape parameters. The proposed framework provides the expected median and covariance of the spectral shape parameters conditioned on specified design values of PGA and PGV/PGA (peak ground velocity/PGA). Efficient step-by-step procedures are developed for the application of the proposed alternative approach, and a sample application is illustrated and compared with the CS approach. It is demonstrated that the proposed methodology is capable of incorporating conventional probabilistic seismic hazard analysis (PSHA), considering spectral shape and spectral variability of realistic ground motions while avoiding the choice of the conditioning period. Site class effects can also be incorporated into the proposed approach.

Risk-targeted hazard for seismic design in New Zealand considering individual and societal risk targets

The promotion of risk-targeted hazard as the basis of seismic design internationally has increased over the past decade. Risk-targeted hazard estimates are derived through convolution of hazard curves with representative fragility functions and provide a means to target uniform risk across a region. Using risk targets also enables performance objectives of building codes that are consistent with other life risks and can include societal input into the expected performance of structures. Current design procedures using a uniform-hazard spectrum are unable to provide equal risk across the country due to variation in the shapes of hazard curves in different locations and uncertainty in structural response. In this article a framework is proposed, which can be used with national seismic hazard models, to produce risk-targeted hazard spectra for seismic design. This study extends the risk-targeted hazard framework through full quantification of epistemic uncertainty in seismic hazard and includes multiple risk targets for individual and societal risk at the building and city scales. The article describes a framework that can be used to adjust current seismic design force levels in New Zealand standards to target uniform seismic risk for buildings considering both the performance of individual buildings and the performance of cities.

A non-extensive approach to probabilistic seismic hazard analysis

Abstract. We modify the probabilistic seismic hazard analysis (PSHA) formulation by replacing the Gutenberg–Richter power law with the SCP (Sotolongo-Costa and Posadas) non-extensive model for earthquake size distribution and call it NEPSHA. The SCP claimed to model the regional seismicity better than the classical models. The proposed method (NEPSHA) is implemented in the Tehran region, and the results are compared with the classic PSHA method. The hazard curves show that NEPSHA gives a higher hazard, especially in the range of practical return periods. The uniform hazard spectra of NEPSHA provide more spectral accelerations, especially for the medium-height buildings, which are the most common urban structures.

Parametric study on the collapse probability of modern reinforced concrete frames with infills

This study aims to compare the collapse probability of reinforced concrete infilled frames designed in accordance with modern Indian seismic codes. Multiple versions of bare and infilled (both fully and partially) frames are considered. A total of 33 frames are divided into eight sets based on the parameter being varied, namely, the quality and thickness of infills, design code, aspect ratio, and number of stories. The collapse probability is evaluated through multiple stripe analysis with eight stripes corresponding to different return periods. A set of 40 ground motions, consistent with the site-specific uniform hazard spectra corresponding to each return period, is selected. Results suggest that the frames designed for pre-2016 Indian codes perform considerably worse than those designed for post-2016 Indian codes, which are on par with other national codes. The effect of collapse probability of addition of infills to bare frames is found to be negligible to significantly negative depending on factors such as adopted design code, relative strength of infill strut to columns, aspect ratio, and number of stories. Four-story frames perform worse than 8-story frames. Interestingly, the 8-story open first story frames, designed for modern codes, are not found to be significantly more vulnerable than their fully infilled counterparts.

Assessment of seismic acceleration demands on nonstructural components in moderately ductile RC frame buildings designed according to NBC 2015

This paper presents an extensive investigation of the seismic demands on acceleration-sensitive nonstructural components (NSCs) attached to reinforced concrete moment-resisting frame buildings with limited ductility and designed according to the National Building Code of Canada. A total of four benchmark structures with varying heights are considered in this study. Linear and nonlinear time-history analyses were performed using artificial ground motions compatible with Montreal site Class C uniform hazard spectra having 2% and 10% probabilities of being exceeded in 50 years. Seismic acceleration demands on NSCs were evaluated by assessing the horizontal height factor Ax, the component dynamic amplification factor Ar, and the component force factor Sp compared with various building code provisions. The overall acceleration demands of NSCs are discussed, and a corresponding optimized Sp profile is proposed to improve the estimation of the seismic force demands on NSCs at the rooftops of buildings located in moderate seismic zones.

Towards seismic performance quantification of viscous damped steel structure: Site-specific hazard analysis and response prediction

With an emphasis on predictable performance, the paramount importance of performance-based earthquake engineering (PBEE) in quantifying earthquake risks and facilitating the better-informed design of the built environment to achieve earthquake resilience has been widely acknowledged. And the uses of damping systems, i.e., structures incorporated with supplemental damping devices, have been recognized as an effective structural development to increase the resilience of structures to earthquakes. This paper presents a uniform hazard spectrum (UHS) based site-specific ground motion selection procedure, to implement the PBEE framework to relate earthquake hazard to structural performance for structures with variable dynamic properties. This ground motion selection procedure accounts for the effect of variable structural dynamic properties on hazard characterization for structures with damping systems. A paradigm building site, on which the site-specific hazard from the probabilistic seismic hazard analysis (PSHA) is consistent with the risk-targeted hazard from the design maps of ASCE/SEI 7-10, is selected for the implementation of the procedure. Three suites of 40 ground motions representing various hazard intensities at the building site are selected for time history dynamic analysis of a prototype structure with nonlinear viscous damping system. Evaluated is the probability distribution of major engineering demand parameters (EDPs) including story drift, residual story drift, floor velocity, and floor acceleration. The results demonstrated that the major EDPs of a building structure with supplemental nonlinear viscous dampers at a paradigm site can be estimated using the lognormal probability distribution, and the proposed UHS-based site-specific ground motion selection procedure is critical to the performance assessment of structures with variable dynamic properties in the PBEE framework.

Seismic hazard analysis of Silchar city located in North East India

North East India is a highly seismically active zone of India due to the collision tectonics between the Indian and the Eurasian plate in the north and subduction tectonics along the Indo-Myanmar range (IMR) in the east. It is utmost important to investigate the seismicity of this region along with uniform hazard spectra (UHS). In this paper, seismic hazard study has been performed at the Silchar district headquarters of Assam state, located in North East India. The earthquake data have been collected from various agencies from 1762 to 2020 and the fault data has been collected from seismotectonic atlas (SEISAT). Earthquake and fault data are superimposed on the Northeast India map using MapInfo software. The fault zone is defined based upon the earthquake’s density near the fault. Seismicity parameters and maximum magnitude have been estimated using Gutenberg-Richter (G-R) relationship. The best-fitted ground motion equation is considered. The outcome is represented in terms of controlling source, seismic hazard curve and UHS for Silchar city with varying ground motions. Finally, the results are presented as PGA based upon DSHA and PSHA methods. The effect of Importance Factor (I) and Response reduction factor (R) have been investigated by comparing UHS with IS: 1893-2016 code response spectra.

Application of Probabilistic Seismic Hazard Assessment to Understand the Earthquake Hazard in Attock City, Pakistan: A Step towards Linking Hazards and Sustainability

Within the last three decades, twelve major earthquakes (Mw > 6.0) have jolted Pakistan and contributed to a heavy death toll and an economic loss of billions of dollars, which is immense for any underdeveloped country. Despite the generalized description of seismic hazards in various regions of Pakistan, densely populated cities still require a detailed and integrated vulnerability analysis to overcome the impact of a significant earthquake. This study aims to integrate seismic hazard assessment schemes to understand the vulnerability of Attock city against an earthquake. It initially evaluates the threat from an earthquake due to tectonic activity in the region, splits the region (about 200 km radius) into six seismic zones and uses area source parameters. The ground motion prediction equations compatible with the study area’s seismotectonic environment are also used in this study. Peak horizontal ground acceleration (PGA) and 5% damped spectral acceleration are critical features of ground motions. The site classification is carried within Attock city, indicating the presence of SB (foundation condition with Vs30 = 760 m/s), SC (foundation condition with Vs30 = 400 m/s) and SD (foundation condition with Vs30 = 300 m/s). The peak ground accelerations for a return period of 475 years at the SB, SC and SD sites are estimated as 0.23 g, 0.28 g and 0.30 g, respectively. Uniform hazard spectra are obtained for each site classification at three return periods (475, 975 and 2475 years). Another possible threat can be the local site conditions of the study area, as Attock city exists on the unlithified sediments of upper Pleistocene to Recent alluvial deposits. That is why microtremor recordings are conducted at 20 sites within Attock city to understand the fundamental frequency (f0), horizontal to vertical spectral amplitude (A0) and Kg parameter, a seismic vulnerability index. The values of f0 are found between 0.6 and 9 Hz and A0 is observed between 2.1 and 5 Hz, whereas the Kg is estimated between 0.24 and 20 Hz. Despite evidence of the seismic vulnerability of Attock city, the current building designs and infrastructure development are not synchronized with the uniform hazard response spectra and the soil amplification, thus enhancing the exposure of the study area to disaster during a major earthquake. This study will be instrumental in pre-disaster mitigation strategies for urban planners and policymakers.

Best Matching Scenario Earthquake: an alternative paradigm to disaggregation of PSHA

Disaggregation of Probabilistic Seismic Hazard Assessment (PSHA) is commonly employed to estimate the most likely magnitude and distance values (Mw - R) contributing to the hazard at the site of study. Nevertheless, however defined, probabilistic disaggregation techniques are affected by several ambiguities. Among the others, they require the selection of a specific branch of the logic tree used to perform the PSHA, and of a specific structural period, which may be of limited use in some engineering applications, such as for structures with strong participation of higher modes and highly affected by non-linear response, as well as site response analyses and most of the geotechnical engineering applications. In order to overcome some of these drawbacks, this paper introduces a new possibility aiming at estimating a Best Matching Scenario Earthquake (BMSE) defined as the earthquake, specified in terms of (Mw - R), the median spectral ordinates of which best approach a target uniform hazard spectrum (UHS) for the range of structural periods of interest. A more general form of the BMSE can be introduced, i.e., in terms of (Mw – R –) instead of (Mw – R – =0), but the paper shows that excellent results can be obtained in most cases along the Italian territory, by limiting the search to =0. In this article, after highlighting the assumptions behind the method, a few examples of the application of the BMSE with reference to the current seismic hazard model of Italy are shown. Our results show the potential effectiveness of the best matching method that gives in most cases scenarios which are consistent with historical observations. Besides normative implications, such as the selection of input motions for non-linear time history analyses, the BMSE may provide a rationale support for using seismic ground shaking scenarios in a probabilistic framework.

Probabilistic seismic hazard assessment for Western Mexico

An updated probabilistic seismic hazard analysis in terms of peak ground acceleration and spectral acceleration values, for B, B/C, and C NEHRP site classes, for a 5% and 10% probability of exceedance in 50 years, has been conducted for Western Mexico. To achieve this assessment, a unified and updated declustered earthquake catalog (1787–2019), as well as an updated focal mechanism database (1963–2016), was compiled, prepared, and processed specifically for this work. Two alternative source models (area sources and a spatially smoothed seismicity model) were considered in the assessment, within a logic tree scheme. A characteristic earthquake model has been also implemented for some of the defined sources. The designed logic tree has additionally included other parameters: the possible uncertainty related to the Gutenberg-Richter b-value way of estimation, the maximum expected magnitude value, as well as two-alternative GMPEs selected for the subduction seismic sources. The obtained ground-motion results have been presented as seismic hazard isoacceleration maps, as well as uniform hazard spectra and hazard curves for 15 selected cities, for the three considered site conditions. A comparison between the obtained hazard values and the current seismic design regulations, as well as previous studies, has been also performed, and a new design response spectrum has been proposed. Finally, some regression fitting relationships between the obtained ground-motion values have been achieved. Concerning the most significant results, it is worth noting that the southwestern coast (in the perimeter of the Middle America Trench of the Mexican Subduction Zone) presents the highest hazard values. For instance, the cities of Apatzingán, Autlán, Colima, Lázaro Cárdenas, and Manzanillo exhibit the largest observed PGA values among all the studied cities. Our approach and results are supported by the fact that the recent September 19, 2022, Mw 7.6 earthquake has been located in this highest seismic hazard area.

The 2022 Seismic Hazard Model for Costa Rica

AB4STRACTCosta Rica is located at the boundary of four tectonic plates where the regularity of destructive earthquakes highlights the necessity of seismic hazard estimations. This study contains the most recent Probabilistic Seismic Hazard Assessment (PSHA) for Costa Rica, calculated with the largest and the most updated earthquake database from both—the Earthquake Engineering Laboratory and the National Seismological Network of the University of Costa Rica. For the PSHA, we updated the seismicity parameters for the upper plate, subduction interplate, and intraslab tectonic domains, characterized the upper-plate zones by percentages of fault types, and used weighted ground-motion models for each of the tectonic domains. The resulted maps of peak ground acceleration (PGA) at return periods of 475 yr (PGA-475) and 2475 yr, as well as the spectral accelerations, show geographic trends that allow for the division of the country in four seismic hazard levels: (1) extremely high for the Nicoya, Osa, and Burica peninsulas, situated directly above the subduction interplate, where the PGA-475 could be 0.55–1.20g; (2) very high for most of the Guanacaste Province, where the PGA-475 may be 0.55–0.70g; (3) high for most of the country (∼41%) with PGA-475 values of 0.40–0.55g, including Central Costa Rica and the capital city of San Jose; and (4) moderate for the Talamanca Cordillera and Northern Costa Rica, with PGA-475 up to 0.40g. These ground-motion values are 0.1–0.6g higher than the previous PSHA for the Pacific peninsulas, Guanacaste, and the southeastern Caribbean. Further, hazard curves, uniform hazard spectra, and a hazard disaggregation indicate that the seismic hazard is lower but more complex in San Jose than in Liberia—the largest city in Guanacaste.

A Caribbean and Central America Seismic Hazard Model for Sovereign Parametric Insurance Coverage

ABSTRACTA fully probabilistic seismic hazard model with a single domain and sufficiently accurate resolution level for national analyses has been developed, for the Caribbean and Central America, to support the design of parametric earthquake policies offered by the Caribbean Catastrophe Risk Financing Facility to sovereign countries. This model provides updated earthquake hazard and risk information for 34 countries in the region, allowing to obtain detailed seismic hazard results at any location within the area of analysis. Besides a detailed zonation of subduction and crustal sources, updated seismicity and tectonic data have been utilized. Different seismicity, geometric, and rupture models have been adopted using state-of-the-art methodologies and tools. The main output of this model is a stochastic event-set, which is the hazard representation utilized not only to compute a long-term overview of the seismic hazard in the region but also to probabilistically estimate earthquake modeled losses at a national level that are used as triggers for underpinning parametric earthquake insurance. Although not of direct interest for parametric insurance purposes, this model also yields hazard maps and uniform hazard spectra for different return periods. For the operational purposes of the parametric insurance coverage, a postevent tool was developed to calculate, in near-real time, the ground-motion intensities associated with any earthquake within the region under study. This is done automatically by choosing a rupture plane from the moment tensor solution based on predefined rules while maintaining full consistency with all the assumptions made in the probabilistic seismic hazard analysis. The predicted ground motions in the region and the vulnerability functions developed for the exposed assets are then used to estimate whether the economic losses for each affected country exceed the trigger value stated in the country-specific insurance policy conditions and whether a payout should be issued, which is typically disbursed within 14 days.

Probabilistic and Scenario-Based Seismic Hazard Assessment on the Western Gulf of Corinth (Central Greece)

The Gulf of Corinth (Central Greece) is one of the most rapidly extending rifts worldwide, with its western part being the most seismically active, hosting numerous strong (M ≥ 6.0) earthquakes that have caused significant damage. The main objective of this study was the evaluation of seismic hazard through a probabilistic and stochastic methodology. The implementation of three seismotectonic models in the form of area source zones via a logic tree framework revealed the expected level of peak ground acceleration and velocity for return periods of 475 and 950 years. Moreover, PGA values were obtained through the stochastic simulation of strong ground motion by adopting worst-case seismic scenarios of potential earthquake occurrences for known active faults in the area. Site-specific analysis of the most populated urban areas (Patras, Aigion, Nafpaktos) was performed by constructing uniform hazard spectra in terms of spectral acceleration. The relative contribution of each selected fault segment to the seismic hazard characterizing each site was evaluated through response spectra obtained for the adopted scenarios. Almost all parts of the study area were found to exceed the reference value proposed by the current Greek National Building Code; however, the three urban areas are covered by the Eurocode 8 regulations.

Variations in uniform hazard spectra and disaggregated scenarios during earthquake sequences

Seismic hazard varies greatly during an earthquake sequence. Understanding this variation can be useful to end-users, such as emergency managers, as it would enable them to make more informed decisions about potential risk reduction measures. This article presents examples of how two commonly-used products of probabilistic seismic hazard assessments: uniform hazard spectra and disaggregated earthquake scenarios, vary during two severe seismic sequences in western Greece. These calculations are made using a recent time-dependent seismic hazard model based on a Bayesian ETAS approach. The examples show that time-dependent uniform hazard spectra for short return periods (1 and 10 years) are significantly higher than standard time-independent spectra but that uniform hazard spectra for the commonly-used return periods of 475 and 2475 years are similar to those from time-independent assessments. The time-dependent spectra generally converge within a couple of days to the time-independent spectra. The examples also show that the dominant earthquake scenarios evidenced by the disaggregation for the time-dependent assessment can show significant differences from the time-independent scenarios. This is particularly true when the earthquake sequence is distant from the location of interest as the aftershocks contribute greatly to the overall hazard. To show these changes more clearly this article introduces a new graphical representation of the disaggregated results: contour maps showing the magnitude or distance of the dominant earthquake scenario with axes of the structural period and response spectral acceleration.

Selecting Ground Motions for Flexible Structures Based on Multiple Conditional Mean Spectra: A Case Study on Cable-Stayed Bridges

ABSTRACT For flexible structures, this study proposes multiple conditional mean spectrum (MCMS) to account for the scattered structural modes and the piecewise conditional mean spectrum (PCMS) for the higher mode effect. A flexible cable-stayed bridge is used for a case study. Several scenario earthquakes are identified responsible for the maximum responses of the bridge. The envelope of structural responses under different scenario earthquakes would be used as the design seismic effect. Adopting uniform hazard spectrum to selection ground motions would for most cases amplify the structural seismic responses, especially for tower bending moment and shear force.

Amplification analysis at Nador city and surrounding area (NE of Morocco)

This study presents a ground response assessment performed at Nador city (Morocco) and surrounding areas. We carried out the one-dimensional equivalent linear approach using the known geotechnical parameters to derive amplification analysis at studied sites, using the shear wave velocity (VS(30)) derived from boreholes. Seismic hazard assessment is performed using two gridded seismicity models and four ground motions attenuation equations, in order to address the epistemic uncertainty in a logic-tree framework. Seismic hazard maps depicting the mean estimated hazard values and including the uncertainty are presented in terms of spectral acceleration at 0.0, 0.1 and 0.2 s, with 10% and 5% probabilities of exceedance in 50 years, and for B, B/C and C soil conditions, as defined by the National Earthquake Hazards Reduction Program classification and for local site conditions as well. Typical values of Peak Ground Acceleration (PGA) equal to 0.18g±0.03 and 0.26g±0.05 are obtained at the Nador city for 475 and 975 return period respectively and for B/C soil conditions. The deaggregation analysis for both returns periods (475 and 975 years) in terms of magnitude-distance (named 2D deaggregation) and in azimuth (longitude and latitude named 3D deaggregation) allows us to derive the contribution of the different couple of magnitude-distance (M,D) and the different sources, the controlling earthquake are also determined at the five considered cities, for both return periods. From the site-specific seismic hazard analysis performed at five main cities located in the studied area, the seismic hazard curves and the uniform hazard spectra (UHS) are obtained with their uncertainties for the considered return periods and soil conditions, as well as the design spectra proposed by Malhotra (2005). Real accelerogram time-history is used to generate compatible acceleration time-history with the design spectra proposed by Malhotra (2005) and the EC-8 type 2 (Eurocode 8.2004), for both return periods and for B/C soil conditions. The lower root mean square error value (RMSE) is archived by using the Malhotra design spectra for the considered return periods. The results display a clear amplification in silty and clayey soils, greater than in silt-gravely soils, as well as no significant amplification in volcanic soils. Moreover, the resulting fundamental frequency is in the range of 1.4–2.8 Hz, and the maximum amplification factor observed in Nador is about 3.5. These values are consistent with the results derived from ambient noise, seismic records using HVSR technique and with the values given by Moroccan Seismic Code in the 2000 and 2011 regulations.

Seismic hazard map of India and neighbouring regions

This article presents probabilistic seismic hazard analysis (PSHA) of India and adjoined region, carried out to develop a new national seismic hazard map for India. The hazard map is developed using fault oriented spatially smoothed seismicity approach. A catalog of earthquakes has been compiled for the region (Latitude 50N − 400 N and Longitude 650E − 1000E) from 2600BCE to 2019CE to estimate the seismicity parameters. Eighteen suitable ground motion prediction equations in the logic tree framework are used for the four major geological regions of the country. The hazard is estimated at rock sites (B-C boundary type) conditions in terms of peak ground acceleration (PGA), short-period (0.2 s), and long-period (1s) spectral acceleration maps and uniform hazard spectra, with 2% and 10% probabilities of exceedance in 50 years. Higher hazard values are observed in the Hindukush-Pamir regions and Northeast India, whereas central India and the southern peninsular regions are less prone to seismic threat. The proposed maps find their application in the seismic design of structures, risk assessment, and as an input for updating the existing code provisions.

Seismic zoning maps of the National Capital Region (NCR) of India

AbstractWe present seismic zoning maps of the National Capital Region (NCR) of India in terms of uniform hazard spectrum (UHS) amplitudes using three different attenuation equations to model the contributions of local earthquakes around the region and distant earthquakes in northwestern Himalaya (NWH) and Hindu Kush Subduction (HKS) sources. Our results show that the shape of design spectra must depend on the nature of contributing earthquake sources, their relative activity, the potential to create large magnitudes, and their geographical placement relative to the site selected for hazard computation. Our results also show that specifying strong motion amplitudes for use in engineering design by means of peak ground acceleration and a standard spectral shape may result in significant underestimation of intermediate and long‐period ground motions critical for buildings higher than four to five stories and hence should not be used. We present zoning maps for nine possible combinations of three different local geological site conditions (sediments, intermediate sites, and basement rock sites) and three different soil site conditions (“rock,” stiff soil sites, and deep soil sites). One of these maps can be readily used to extract the UHS amplitudes of the pseudo relative velocity spectrum (PSV) at any site in the NCR by specifying the local geological site‐condition parameters from a high‐resolution surface geological map of the region and determining the local soil site conditions by experiments limited to the site alone. The paper also provides a geological map of NCR on a scale of 1:1 million for a rough idea of the local geological parameters for a site of interest.

A novel framework for seismic risk assessment of structures

AbstractSeismic risk assessment of new and existing structures has a long history, which is based on the development of specific methods for seismic hazard and vulnerability analyses. Performance‐based earthquake engineering is the natural evolution and integration of all these aspects that is nowadays implemented in several precodes. However, some critical aspects are not yet completely clarified. As a matter of fact, the vulnerability analysis suffers the so‐called record‐to‐record variability, which is usually controlled through different parameters used as seismic intensity measures (IMs), whose significance is still under discussion. Moreover, this variability represents an additional source of uncertainty, which is added to the dispersion derived from the ground motion prediction equation (GMPE), leading to an overconservatism. For these reasons, the present paper aims to propose a novel framework for the seismic risk assessment of structures, which is found on the idea to control the response variability in evaluating seismic hazard curves without taking into account the randomness (ε) of the GMPE, which instead is transferred to fragility curves. These latter are built by using groups of accelerograms, whose median and 84% fractile spectra fit well, for different return periods, the uniform hazard spectra for ε= 0 and 1, respectively. For this purpose, a new search algorithm for selecting natural records is formulated. The proposed method offers considerable advantages as it is no longer necessary to refer to a specific IM, and allows to select pairs of spectrum‐compatible natural records, which is enable to solve the problem of the seismic assessment of three‐dimensional structures. The procedure is then applied to assess seismic risk of a typical reinforced concrete frame, whose results demonstrate the robustness of the method and the practical independence from the record set used.