UBC Faculty Research and Publications

Comparison of Seismic Performance & Recovery Metrics for a 1970s vs Modern Tall Steel Moment Frame Building Molina Hutt, C.; Rossetto, Tiziana; Almufti, I.; Deierlein, Gregory G. (Gregory Gerard), 1959- 2018-06-29

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


52383-Hutt_C_et_al_Comparison_seismic_performance.pdf [ 1.18MB ]
JSON: 52383-1.0379176.json
JSON-LD: 52383-1.0379176-ld.json
RDF/XML (Pretty): 52383-1.0379176-rdf.xml
RDF/JSON: 52383-1.0379176-rdf.json
Turtle: 52383-1.0379176-turtle.txt
N-Triples: 52383-1.0379176-rdf-ntriples.txt
Original Record: 52383-1.0379176-source.json
Full Text

Full Text

Eleventh U.S. National Conference on Earthquake EngineeringIntegrating Science, Engineering & PolicyJune 25-29, 2018Los Angeles, CaliforniaComparison of Seismic Performance & Recovery Metricsfor a 1970s vs Modern Tall Steel Moment Frame BuildingC. Molina Hutt1, T. Rossetto2, I. Almufti3, and G.G. Deierlein4ABSTRACTThis study benchmarks the performance of older existing tall steel moment resisting frame (MRF)buildings designed following historic code-prescriptive requirements (1973 Uniform BuildingCode) against modern design standards (2012 International Building Code). The comparison isbased on risk-based assessments of alternative designs of a 50-story archetype office building,located at a site in San Francisco, CA. The following metrics are compared: (i) mean annual rate ofcollapse, λc (ii) average annual loss (AAL), and (iii) average annual downtime (AAD). The meanannual frequency of collapse of the the 1973 archetype building is 28 times greater than theequivalent 2012 building (28·10-4 versus 1·10-4), or approximately 13% versus 0.5% probability ofcollapse in 50 years. The expected AAL is 65% higher for the 1973 than the 2012 building (0.66%versus 0.40% of building replacement cost); and the AAD to re-occupancy is 72% greater for the1973 than the 2012 building (8.1 vs 4.7 days). The AAD to functional recovery for the 1973 buildingis twice that of the 2012 building (10.4 vs 5.0 days). An evaluation of the results at variousearthquake ground motion shaking intensities suggests that existing 1970s tall steel moment framesare far from complying with modern design requirements in terms of both collapse safety underextreme ground motions and damage control in small to moderate magnitude earthquakes.Furthermore, while modern building code requirements provide acceptable seismic collapse safety,they do not ensure a level of damage control to assure a swift recovery after a damaging earthquake.IntroductionA major concern in earthquake disaster resilience is the large risks posed by existing buildings,which may have built-in deficiencies that are not permitted by current building codes. For sometypes of buildings, such as unreinforced masonry structures, the risks are so large, that mandatorylaws have been enacted to assess and retrofit the buildings. However, in other cases, such as with1Assistant Professor, Dept. of Civil Engineering, University of British Columbia, Vancouver, BC Canada V6T 1Z4carlos.molinahutt@civil.ubc.ca2Professor, Dept. of Civil, Env. and Geomatic Engineering, University College London, London, WC1E 6BT, UK3Associate, Advanced Technology and Research, Arup, 560 Mission Street, San Francisco, CA 94105, USA4Professor, Dept. of Civil and Environmental Engineering, Stanford University, Stanford, CA 94305, USA.Molina Hutt C., Rossetto T., Almufti I. and Deierlein G. G. Comparison of Seismic Performance & RecoveryMetrics for 1970s vs Modern Tall Steel Moment Frame Building Proceedings of the 11th National Conference inEarthquake Engineering, Earthquake Engineering Research Institute, Los Angeles, CA. 2018.non-ductile concrete buildings or older tall steel buildings, the risks and mitigation strategies arenot as straightforward. The goal of this study is to benchmark the performance of older seismicallyvulnerable tall steel moment resisting frame (MRF) buildings, which constitute a significantportion of tall buildings constructed between 1960 and 1990 [1, 2] in San Francisco, Los Angelesand other high seismic regions of the western US, against modern designs. The comparison isfocused on risk metrics that can help inform policy and decision making.This study benchmarks the performance of older existing tall steel MRF buildings designedfollowing historic code-prescriptive requirements (UBC 1973 [3]), against modern designstandards (IBC 2012 [4]) by conducting a comparative risk-based assessment of alternative designsof a 50-story archetype office building, located at a site in San Francisco, CA. The followingmetrics are compared: (i) annual rates of collapse, λc; (ii) average annual losses (AAL); (iii)average annual downtime (AAD).MethodologyThe archetype buildings are developed based on a database of the existing tall building stock inSan Francisco. Numerical models of the structure are developed to assess structural performanceusing nonlinear dynamic analyses. Performance is evaluated under sets of earthquake groundmotions, selected and scaled to various intensities. Expected losses are determined from structuralresponse following the FEMA P-58 methodology [5]. Downtime estimates follow the Resilience-based Earthquake Design Initiative’s (REDi) downtime assessment methodology [6].The occupancy of the archetype buildings is assumed to be commercial office. Typicalstory heights and beam spans are 3.8 m and 8.5 m, respectively. In the 1970s, it was customary tohave moment connections in all beam-to-column intersections, as illustrated in Figure 1b, but dueto practical and economic factors, this practice evolved to perimeter frame structures in which onlytwo frame lines in each direction are moment-resisting, as illustrated in Figure 1c. The resultingsection sizes for a typical frame in each archetype building are shown in Figure 1a. The framesconsist of wide flange beams, welded beam-to-column connections, and built-up box columns(denoted R in Figure 1a) in the 1973 building and built-up I sections (denoted I in Figure 1a) inthe 2012 building. The total building dead load is approximately 800,000 kN. The design windand seismic base shears are equal to 1.80% and 1.96% of total building dead load, respectively,per UBC 1973 requirements, versus 4.26% and 3.74% per IBC 2012. For a detailed description ofthe design method and assumptions refer to [2].There are many requirements in modern design standards, which are not present in the1973 designs, which drastically improve seismic performance, e.g. response spectrum analysismethod as opposed to equivalent lateral force, minimum base shear requirements, p-delta effects,strong column weak beam provisions, capacity design principles, etc. In addition, the fractureprone-moment connections, which resulted from a combination of low toughness weld materials,large built-in initial flaws, and high stain demands are expected to negatively affect the seismicperformance of the older steel frames. The switch in the weld process that led to welds with lowtoughness, as evidenced by fractures observed in the 1994 Northridge earthquake, took place inthe mid-1960s [7]. Therefore, it is assumed that that fracture-prone pre-Northridge momentconnections are common in designs from the 1970s.(a)UBC 1973 IBC 2012(b) (c)(d) (e)(f) (g)(h) (i)Figure 1: Archetype steel MRF building elevation (a), plan (b, c) and sample hysteretic responseof key structural components: beams (d, e), columns (f, g) and panel zones (h, i).The nonlinear dynamic analyses are performed using finite element models capable ofcapturing the response of all structural elements that significantly contribute to the strength andstiffness of the system. A 2D numerical model of a representative frame, as illustrated in Figure1a, is developed in LS-DYNA [8]. The first (T1), second (T2) and third (T3) modes of the-2,500-2,000-1,500-1,000-50005001,0001,5002,0002,500-0.075 -0.05 -0.025 0 0.025 0.05 0.075Moment(kN-m)Rotation (rad)ASCE 41-13 BackboneNumerical SimulationMy = 1890 kN-mθp = 0.018 radθpc = 0.158 radΛ = 0.869 radW33x130θfracture = 0.008 rad-3,000-2,500-2,000-1,500-1,000-50005001,0001,5002,0002,500-0.075 -0.05 -0.025 0 0.025 0.05 0.075Moment(kN-m)Rotation (rad)Experimental DataNumerical SimulationMy = 1890 kN-mθp = 0.018 radθpc = 0.158 radΛ = 0.869 radSpecimen: P9704W33x130-500-400-300-200-1000100200300400500-0.15 -0.10 -0.05 0.00 0.05 0.10 0.15Moment(kN-m)Rotation (rad)Experimental DataNumerical SimulationMy = 240 kN-mθp = 0.021 radθpc = 0.147 radΛ = 0.743 radPn = 3,365 kNALR = 0.3Specimen: S1703200x200x17x17(mm)-500-400-300-200-1000100200300400500-0.15 -0.10 -0.05 0.00 0.05 0.10 0.15Moment(kN-m)Rotation (rad)Experimental DataNumerical SimulationMy = 240 kN-mθp = 0.031 radθpc = 0.436 radΛ = 2.598 radPn = 3,365 kNALR = 0.1Specimen: S1701200x200x17x17(mm)-2,000-1,500-1,000-50005001,0001,5002,000-0.010 -0.005 0.000 0.005 0.010Moment(kN-m)Rotation (rad)Experimental DataNumerical SimulationColumn:H300x300x16x25(mm)Beam:H600x200x11x17(mm)Specimen:TS9900b-TCLRB1γy = 0.0012 radγp = 0.0048 radMy = 850 kN-mMp = 1050 kN-m -7,500-5,000-2,50002,5005,0007,500-0.015 -0.010 -0.005 0.000 0.005 0.010 0.015Moment(kN-m)Rotation (rad)Experimental DataNumerical SimulationColumn:W36x230Beam:W35x150Specimen:ATLSS0413-SPEC1-Eγy = 0.002 radγp = 0.008 radMy = 3800 kN-mMp = 4800 kN-mrepresentative 2D frame are 5.5, 2.1 and 1.3 seconds, respectively, for the 1973 archetype buildingversus 5.1, 2.0 and 1.2 seconds, respectively, for the 2012 archetype building.Key structural elements include beams, columns and panel zones. Beams are modeled aslumped plasticity beam elements following recommendations in [9], which propose modelingparameters based on a large database of experimental tests. To account for fracture in the momentconnections, a plastic rotation threshold at which fracture is set to occur in the connections isintroduced according to ASCE 41 recommendations [10]. The impact of the plastic rotationthreshold at fracture is illustrated in the hysteretic responses for a pre- versus post-Northridgesample moment connection, as shown in Figure 1d versus 1e. Columns are modeled as lumpedplasticity beam elements with yield surfaces capable of capturing interactions between bendingmoment and axial force following the recommendations in [11], which account for different ratesof degradation in the moment-rotation response as a function of axial load-to-capacity ratios(ALR). Sample moment-rotation responses of columns with an ALR of 0.3, representative of the1973 building, and applied ALR of 0.1, representative of the 2012 archetype, are illustrated inFigures 1f and 1g, respectively. Panel zones are modeled using the Krawinkler model as outlinedin [12], which incorporates an assembly of rigid links and rotational springs to represent the finitedimensions and flexibility of the panel zone. A sample shear force-deformation response of a panelzone is illustrated in Figures 1h and 1i. The analytical component models are verified againstavailable test data, as seen in Figures 1e to 1i.(a) (b)Figure 2: Seismic hazard curve illustrating preliminary assessment points (a), target spectra andassociated ground motion suite for a sample earthquake ground motion intensity level (b).The performance of the archetype buildings is evaluated through a Multiple Stripe Analysis(MSA) approach, where assessments are performed at a series of ground motion shaking intensitiesspanning from high to low probabilities of occurrence. The upper and lower bound intensity levelsconsidered result in a range of damage from negligible to complete loss. Preliminary bounds forassessment are applied to the seismic hazard curve at a representative site in downtown SanFrancisco, as shown in Figure 2a, with soil properties consistent with site class D [1]. Whennecessary, additional assessment points are introduced to obtain the desired range of damage. Thehazard curve is obtained using the USGS hazard curve calculation tool [13]. A seismic hazard0.0000.0010.0100.1001.0000.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45λ(e)Annualfrequencyofexceedance,AFEEarthquake intensity, eSpectral acceleration, SA(T=5 sec) (g)Hazard Curve T=5 secIntensity Level BoundsAssessment PointsΔλ1e1e2e3e4e5 e6 e7 e84Δe1 Δe2 Δe3 Δe4 Δe5 Δe6 Δe7 Δe80.010.101.0010.000.01e4AFE = 0.00141SA (T = 5 sec) = 0.18gReturn Period = 712 yearsM = 7.70R(km) = 15.60ε  = 0.850.010.1 1.0 10.0Period (sec)curve for a 5 second period is selected as it is close to the fundamental period of the archetypebuildings and is the longest period for which USGS provides seismic hazard data. To select groundmotion records for the dynamic analyses, a conditional spectrum (CS) [14], conditioned at a 5second period, is selected as the target spectrum for each intensity level, using seismic hazard de-aggregation results. Suites of 20 ground motions are selected to collectively match the entiredistribution of the CS, as illustrated in Figure 2b. The selected ground motions are input at thefixed base supports of the frame analysis models. A critical damping ratio of 2.5% is assumed inthe analysis [15]. For further details of the model and ground motion selection procedure, refer to[2].Owners, insurers and financial institutions often use quantitative statements of probablebuilding repair cost expressed as a percentage of building replacement value. This metric is usedin this work, where future repair costs are converted to present dollars. Repair costs are expressedas a percentage of building replacement cost, which is estimated based on the gross square footageat a rate of $3,550/m2 ($330/ft2) [1]. A building performance model is developed to evaluate theloss and downtime metrics of interest. At each earthquake ground motion intensity levelconsidered in the MSA, 1000 loss and associated downtime estimates are calculated. Results fromnonlinear dynamic analysis are used as input demands to the building performance model, whichcontains structural and non-structural components at each story level for all components in thebuilding that are susceptible to earthquake damage. Structural component quantities are based onthe structural design of the archetype buildings. Non-structural component quantities are estimatedbased on typical quantities found in buildings of similar occupancy by use of the NormativeQuantity Estimation Tool [5]. The components employed in the building performance model forthe archetype buildings are assumed to be the same for the IBC 2012 and UBC 1973 designs, butfragilities are adjusted to account for modern seismic design requirements of structural and non-structural components. This adjustment is possible because the fragility library developed for theFEMA P58 project includes variations of the same component adjusted for different seismic designcategories. The library also includes variations of a same component, e.g. beam-to-columnconnection, to account for important changes in design and construction practice, i.e. pre- and post-Northridge moment connection detailing.Each of these structural and non-structural building components has a component fragilityfunction, i.e. a statistical distribution that indicates the conditional probability of incurring damageat a given value of demand. Each damage state has an associated consequence function, fromwhich the repair cost and repair time associated with the level of damage in the component isestimated. SP3 [16], which adopts the FEMA P58 methodology, is used to conduct the lossassessment. When estimating losses, the impact of excessive residual story drifts is considered bymeans of a typical building repair fragility [5], to account for scenarios in which the building isdamaged beyond repair. If irreparable, repair costs are taken as the building replacement values.A similar approach is followed to consider collapse contribution to losses by means of a collapsefragility, developed from the nonlinear dynamic analysis results.While seismic loss estimates associated with direct economic losses enable discussionswith building owners and investors about expected building performance, they do not provide aquantitative measure of resilience. In addition to direct economic losses, there is significantvulnerability to indirect economic losses due to downtime, defined as the time required to achievea recovery state after an earthquake. Downtime to re-occupancy and functional recovery areconsidered in this work. Re-occupancy occurs when the building is deemed safe enough to be usedfor shelter; and functional recovery occurs when the building regains its primary function, i.e. it isoperational. These estimates follow the REDi downtime assessment methodology [6], whichidentifies the extent of damage and criticality of building components that may hinder achieving arecovery state. It provides a logical approach for labor allocation and repair sequencing includingstructural, interior, exterior, mechanical, electrical, elevator and stair repairs on a floor per floorbasis. Furthermore, the methodology includes delay estimates associated with impeding factors,which may impede the initiation of repairs such as post-earthquake inspection, engineeringmobilization for review or re-design, financing, contractor mobilization, permitting andprocurement of long lead items. Lastly, utility disruptions are also considered when estimatingdowntime for functional recovery. For a more detailed description of the building performancemodel refer to [2].Collapse RiskAssessing the collapse risk of a structure entails combining information related to the behavior ofthe structure with seismic hazard data at the site. The response of the structure is characterized bya collapse fragility, which uses nonlinear dynamic analysis results to describe the increasingprobability of collapse as a function of the ground motion shaking intensity, typically the spectralacceleration (SA) at the fundamental period of the structure. Following the MSA approach,nonlinear dynamic analyses are performed at various intensity levels (‘stripes’) and hazardconsistent ground motions are selected at each intensity level. At the higher earthquake groundmotion shaking intensities, the fraction of ground motions that cause collapse are recorded andused to obtain the collapse fragility for the building, as illustrated in Figure 3.(a) (b)Figure 3: Multiple stripe analysis results (a) for collapse fragility derivation (b) of the 1973 and2012 archetype buildings.The resulting collapse fragility of the 1973 building has an estimated median of 0.13g and adispersion of 0.37, while that of the 2012 building has an estimated median of 0.54g and adispersion of 0.28, as illustrated in Figure 3a. The mean annual frequency of collapse, λc, isobtained by integrating the collapse fragility with seismic hazard curve, i.e., mean annual0. 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0ProbabilityofCollapseSpectral Acceleration, SA(T=5sec) (g)IBC 2012 - Fitted FragilityUBC 1973 - Fitted FragilityIBC 2012 - Fraction of Analyses causing CollapseUBC 1973 - Fraction of Analyses causing Collapse = 0.54 β = 0.28 = 0.13 β = 0.370. 0.03 0.05 0.08 0.10 0.13 0.15SpectralAcceleration,SA(T=5sec)(g)Peak Interstorey Drift Ratio, IDR (-)UBC 1973 - Dynamic Analysis ResultsIBC 2012 - Dynamic Analysis ResultsIBC 2012 - Estimated Median Collapse FragilityUBC 1973 - Estimated Median Collapse FragilityResults shown beyond 0.10 IDRrepresent collapse realizations.frequency of exceeding the ground motion shaking intensity [17]. The calculated, λc is equal to28·10-4 for the 1973 building, versus 1·10-4 for the 2012 building. The results for the modernbuilding are consistent with the λc calculated by [18] for modern, code-conforming concretemoment frames, which is estimated to range from 0.7·10-4 to 7·10-4. Comparing the two sets ofresults, the annualized collapse risk of the 1973 building is 28 times greater than that of the 2012archetype building and about 14 times larger than the maximum risk target of 1% in 50 years(equivalent to λc equal to about 2·10-4), which is implied in the risk-targeted MCER design valuemaps.Loss and Downtime RiskThe results of the MSA can also be used to develop tools and metrics related to the expectedeconomic losses and downtime associated with the performance of the building, such as theannualized loss, i.e. AAL, or annualized downtime, i.e. AAD. These parameters can be inferredfrom the loss and downtime function, respectively, and can be calculated as outlined in [2]. Thesemetrics provide information in for risk management and recovery planning. For instance, the AALcan easily relate to annual insurance payments, and the AAD can be a useful tool for estimatingbusiness disruption and other indirect losses associated with building closures due to seismicdamage.Figures 4a and 4b illustrate the probabilities of observing (i) repairable damage, (ii)residual IDRs rendering the building irreparable or (iii) collapse at each earthquake ground motionintensity for both archetype buildings. These results indicate that collapse risk is the greatestcontributor to the expected losses in the 1973 building. In contrast, residual story drifts renderingthe building irreparable are the greatest contributor to the expected loss in the 2012 building. Thelikelihood of observing these outcomes needs to be considered when computing the AAL and theAAD. In the event of collapse or residual deformations rendering the building irreparable, lossesand downtime are essentially equal to the building replacement cost and time. The buildingreplacement cost is denoted by a loss ratio of 1. The building replacement time is estimated as1000 days based on 300 days for demolition and re-design, and 700 days for reconstruction(approximately two weeks per story).The AAL for the 1973 building is estimated at 0.66% of building replacement cost, whereasthe AAL for the 2012 building is estimated at 0.40%. The contribution of repairable damage,irreparable damage and collapse to the AAL is illustrated in Figures 4c and 4d. These data areconsistent with previous observations of collapse and irreparable damage controlling the overallloss in the 1973 and the 2012 buildings, respectively. As a reference point, [20] evaluated theperformance of a set of modern concrete-framed 20-story buildings to have AALs on the order of0.4 to 0.7% of building replacement cost. These results are consistent with the results for the 201250-story archetype considered in this study, particularly when considering that normalized seismiclosses tend to be lower in tall buildings due to significant damage concentrated in a few storiesrather than distributed throughout the building height [2].Loss Contributions AAL AADReoccupancy AADFunctional RecoveryUBC1973(a)0.66%(c)8.1 days(e)10.4 days(g)IBC2012(b)0.40%(d)4.7 days(f)5.0 days(h)Figure 4: Contribution of repairable damage, irreparable damage from residual drifts and collapse to overall loss, average annuallosses, and average annual downtime to re-occupancy and functional recovery for UBC 1973 and IBC 2012 archetype buildings.Repairable Damage Non-Repairable Damage Collapse0% 20% 40% 60% 80% 100% to overall losses (-)SpectralAcceleration,SA(T=5sec)(g)ExpectedLossRatio,Total0.960.910.820.730.580.230.110.0526.10%6.51%67.39%39.34%5.40%55.26%52.68%4.21%43.11%0% 20% 40% 60% 80% 100% to overall losses (-)SpectralAcceleration,SA(T=5sec)(g)ExpectedLossRatio,Total0.930.720.530.450. estimated AADs to re-occupancy for the 1973 building is 8.1 days versus 4.7 days forthe 2012 building. The estimated AAD to functional recovery for the 1973 building is 10.4 daysversus 5.0 days for the 2012 building. The contribution of repairable damage, irreparable damageand collapse to the AAL is illustrated in Figures 4e to 4h, coinciding with previous observationsof collapse and irreparable damage controlling the loss in the 1973 and the 2012 buildings,respectively.While annualized loss and downtime metrics are useful for risk management and recoveryplanning, they are not as intuitive as an intensity based assessment. For reference, the expectedlosses and downtime under the intensity level closest to the 10% in 50 year hazard (SA at a 5-second period equal to 0.14g in Figures 4a and 4b) are briefly discussed. This intensity level isselected as it is consistent with DBE shaking and with that used by building rating systems. At theselected intensity level, the 1973 building has an expected loss ratio of 0.73, and expecteddowntime of 778 and 790 days to re-occupancy and functional recovery, respectively. Theseestimates are based on a 31% probability of observing repairable damage, 12% of irreparabledamage and 57% probability of collapse. In contrast, the 2012 building has an expected loss ratioof 0.16, and expected downtime of 227 and 268 days to re-occupancy and functional recovery,respectively. These estimates are based on a 91% probability of observing repairable damage, 9%of irreparable damage and 0% probability of collapse.ConclusionsThe following metrics are compared for a 1970s versus a modern 50-story steel MRF archetypebuilding in San Francisco, CA: (i) mean annual rate of collapse, λc (ii) average annual loss (AAL),and (iii) average annual downtime (AAD). The mean annual frequency of collapse of the the 1973archetype building is 28 times greater than the equivalent 2012 archetype building (28·10-4 versus1·10-4), or approximately 13% versus 0.5% probability of collapse in 50 years. The expected AALis 65% higher for the 1973 than the 2012 50-story archetype building (0.66% versus 0.40% ofbuilding replacement cost). The AAD to re-occupancy for the 1973 archetype building is 72%greater than that of the 2012 archetype building (8.1 vs 4.7 days). The AAD to functional recoveryfor the 1973 archetype building is twice that of the 2012 archetype building (10.4 vs 5.0 days). Anevaluation of the results at different earthquake ground motion shaking intensities suggests thatexisting 1970s tall steel moment frames are considerably more vulnerable to both collapse anddamage than modern code-conforming buildings. Furthermore, while modern building coderequirements provide acceptable seismic collapse safety, they do not ensure a level of damagecontrol to enable a swift recovery after a damaging earthquake.AcknowledgmentsThe authors would like to thank Haselton and Baker Risk Group for providing access to SP3, aswell as technical support in the development of the metrics presented in this study.References1. Molina Hutt C., Almufti I., Willford M., Deierlein G. (2015). “Seismic loss and downtime assessment of existingtall steel-framed buildings and strategies for increased resilience.” Journal of Structural Engineering, 142(8):C4015005.2. Molina Hutt C. (2017). “Risk-based seismic performance assessment of existing tall steel framed buildings.”Ph.D. Dissertation, University College London (UCL), London, UK.3. UBC (1973). “Uniform building code 1973 edition.” UBC 73, International Conference of Building Officials,Whittier, CA, USA.4. IBC (2012). “2012 International Building Code.” IBC 2012, International Code Council, Country Club Hills, IL,USA.5. FEMA (2012). “Seismic performance assessment of buildings.” FEMA P-58 prepared by the Applied TechnologyCouncil for the Federal Emergency Management Agency, Washington, D.C., USA.6. Almufti I. and Willford M. (2013). “Resilience-based Earthquake Design Initiative (REDi) for the NextGeneration of Buildings.” Arup, San Francisco, CA, USA.7. FEMA (2000). “State of the Art Report on Past Performance of Steel Moment-Frame Buildings in Earthquakes.”FEMA 355-E, Federal Emergency Management Agency, Washington, D.C., USA.8. LSTC (2011). [Computer software]. LS-DYNA, Livermore Software Technology Corporation, Livermore, CA,USA.9. Lignos D. and Krawinkler H. (2011). “Deterioration Modeling of Steel Components in Support of CollapsePrediction of Steel Moment Frames under Earthquake Loading”. Journal of Structural Engineering, 131(11):1291-1302.10. ASCE (2013). “ASCE-41: Seismic Evaluation and Retrofit of Existing Buildings.” ASCE/Structural EngineeringInstitute (SEI) 41-13, Reston, VA, USA.11. Lignos D. and Krawinkler H. (2010). “A steel database for component deteriorration of tubular hollow squaresteel columns under varying axial load for collapse assessment of steel structures under earthquakes”. JointConference Proc., 7th Internation Conference on Urban Earthquake Engineering & 5th International Conferenceon Earthquake Engineering, Tokyo Institute of Technology, Tokyo, Japan.12. PEER (2010a). “Tall buildings initiative: guidelines for performance-based seismic design of tall buildings.”PEER Report 2010/05, Pacific Earthquake Engineering Research Center, University of California, Berkeley, CA,USA.13. USGS (2008). “United States Geological Survey hazard curve calculation tool.” [Online]. Available at:geohazards.usgs.gov/hazardtool/ [Last accessed: September 2016].14. Lin T. and Baker J. (2015). “Conditional Spectra.” Encyclopedia of Earthquake Engineering, Edited by BeerM., Kougioumtzoglou, I., Patelli E. and Au S., 461-472.15. PEER (2010b). “Modeling and acceptance criteria for seismic design and analysis of tall buildings.” PEER Report2010/111 also published as PEER/ATC-72-1, Pacific Earthquake Engineering Research Center, University ofCalifornia, Berkeley, CA, USA.16. SP3 (2017). [Computer software]. Seismic Performance Prediction Program, Haselton Baker Risk Group, Chico,CA, USA.17. Eads L. (2013). “Seismic collapse risk assessment of buildings: effects of intensity measure selection andcomputational approach.” John A. Blume Earthquake Engineering Center, Technical Report 184, Stanford, CA,USA.18. Haselton C., Liel A., Deierlein G., Dean B. and Chou J. (2011). “Seismic Collapse Safety of Reinforced ConcreteBuildings: I. Assessment of Ductile Moment Frames,” Journal of Structural Engineering, 137(4): 481-491.19. Krawinkler H. and Deierlein G. (2014). “Challenges Towards Achieving Earthquake Resilience ThroughPerformance-Based Earthquake Engineering.” In: Performance-Based Seismic Engineering: Vision for anEarthquake Resilient Society, Edited by Matej Fischinger, Chapter 1.20. Ramirez M., Liel A., Mitrani-Reiser J., Haselton C., Spear A., Steiner J., Deierlein G. and Miranda E. (2012).“Expected earthquake damage and repair costs in reinforced concrete frame buildings.” Earthquake Engineering& Structural Dynamics, 41 (11): 1455-1475.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items