applied

sciences
Article
Real-Time City-Scale Time-History Analysis and Its
Application in Resilience-Oriented Earthquake
Emergency Responses
Xinzheng Lu 1, * , Qingle Cheng 2 , Zhen Xu 3 , Yongjia Xu 2 and Chujin Sun 2
1 Key Laboratory of Civil Engineering Safety and Durability of China Education Ministry, Department of
Civil Engineering, Tsinghua University, Beijing 100084, China
2 Beijing Engineering Research Center of Steel and Concrete Composite Structures, Tsinghua University,
Beijing 100084, China
3 School of Civil and Environmental Engineering, University of Science and Technology Beijing,
Beijing 100083, China
* Correspondence: [email protected]; Tel.: +86-10-6279-5364




Received: 6 July 2019; Accepted: 21 August 2019; Published: 24 August 2019


Abstract: The resilience of cities has received worldwide attention. An accurate and rapid assessment
of seismic damage, economic loss, and post-event repair time can provide an important reference
for emergency rescue and post-earthquake recovery. Based on city-scale nonlinear time-history
analysis (THA) and regional seismic loss prediction, a real-time city-scale time-history analysis
method is proposed in this work. In this method, the actual ground motion records obtained
from seismic stations are input into the building models of the earthquake-stricken area, and the
nonlinear time-history analysis of these models is subsequently performed using a high-performance
computing platform. The seismic damage to the buildings in the target region subjected to this
earthquake is evaluated according to the analysis results. The economic loss and repair time of the
earthquake-stricken areas are calculated using the engineering demand parameters obtained from
the time-history analysis. A program named, “Real-time Earthquake Damage Assessment using
City-scale Time-history analysis” (“RED-ACT” for short) was developed to automatically implement
the above workflow. The method proposed in this work has been applied in many earthquake events,
and provides a useful reference for scientific decision making for earthquake disaster relief, which is
of great significance to enhancing the resilience of earthquake-stricken areas.

Keywords: city-scale nonlinear time-history analysis; real-time city-scale time-history analysis;

resilience-oriented earthquake emergency response; multiple-degree-of-freedom (MDOF) model;
ground motion records

1. Introduction
Earthquakes cause severe damage and economic loss to urban areas, and the resilience of cities
has received worldwide attention. An accurate and rapid assessment of seismic damage, economic loss,
and post-event repair time can provide an important reference for emergency rescue and post-earthquake
recovery. Therefore, it is of great importance to the enhancement of community resilience.
The experience of several major earthquakes in recent years indicates that the assessment of
building damage in an earthquake-stricken area needs to be improved further. After an earthquake,
communication in the disaster area is delayed; the disaster site is usually chaotic, and there are not
enough professionals to evaluate building safety in a short time. Furthermore, rumors and fake
information on the internet may interfere with accurate seismic damage assessment. Therefore, it is
necessary to propose a scientific, objective, and timely method for earthquake loss assessment.

Appl. Sci. 2019, 9, 3497; doi:10.3390/app9173497 www.mdpi.com/journal/applsci

Appl. Sci. 2019, 9, 3497 2 of 16

To date, available near-real-time loss estimation tools can be divided into two categories based
on the size of the area they cover: (1) Global systems and (2) local systems [1]. Global near-real-time
earthquake loss estimation tools primarily include the Prompt Assessment of Global Earthquakes for
Response (PAGER) [2], the Global Disaster Alert and Coordination System (GDACS) [3], and the World
Agency of Planetary Monitoring and Earthquake Risk Reduction (WAPMERR) [4]. Local systems
include the Earthquake Rapid Reporting System in Taiwan; USGS-ShakeCast, Istanbul’s earthquake
rapid response system; and the Rapid Response and Disaster Management System in Yokohama,
Japan [5]. These seismic loss estimation systems are generally comprised of three parts: The ground
motion intensity measure (IM), building inventory and fragility, and direct economic losses and
casualties. The ground motion IM can be obtained directly from the real-time monitoring data of
a seismic network or calculated using ground motion prediction equations (GMPE). The building
inventory data can be determined using either a detailed building database or macroscopic statistical
data. The seismic damage to buildings can be predicted using the damage probability matrix (DPM)
method [6], the capacity spectrum method [7], or artificial intelligence-based methods [8,9]. Economic
losses and casualties are generally calculated using empirical models.
However, the main problems existing in these systems are as follows: (a) The dynamic
characteristics of ground motion are not comprehensively considered; (b) the DPM method is difficult
to apply in areas where historical earthquake data are lacking or in quickly developing areas where
there are large differences between the inventories of current and historical buildings; (c) the capacity
spectrum method cannot easily represent the concentration of damage to different stories or the
time-domain properties of ground motions (e.g., the velocity impulse of ground motions); (d) the
earthquake loss prediction method relies on historical seismic damage data, and the repair time cannot
be provided in these systems.
Consequently, this work proposes a real-time city-scale time-history analysis method for
post-earthquake damage assessment. The actual ground motion records obtained from seismic
stations were input into the building models of the earthquake-stricken area, and the nonlinear
time-history analysis of these models was subsequently performed. The seismic damage, economic
loss, and repair time of the target region subjected to this earthquake were evaluated according to
the analysis result. A program, named “Real-time Earthquake Damage Assessment using City-scale
Time-history analysis” (or “RED-ACT” for short), was developed. The application and the advantages
of the proposed method were demonstrated through actual earthquake events.

2. Real-Time City-Scale Nonlinear Time-History Analysis

2.1. Framework
The proposed framework to conduct the real-time city-scale time-history analysis and loss
assessment is illustrated in Figure 1. The corresponding procedures are as follows:

(1) Obtaining the real-time ground motion records from the seismic stations;
(2) Establishing the building inventory database for the target region;
(3) Conducting the city-scale nonlinear time-history analysis to predict the seismic damage of the
target region; and
(4) Performing the regional seismic loss prediction to assess the seismic economic loss and repair
time of the target region.
Appl. Sci. 2019, 9, 3497 3 of 16
Appl. Sci. 2019, 9, x FOR PEER REVIEW 3 of 16

Figure 1. Framework
Figure 1. Frameworkfor
forreal-time
real-timecity-scale time-historyanalysis
city-scale time-history analysis and
and loss
loss assessment.
assessment.

2.2. (1)
Real-Time Recorded
Obtaining Ground ground
the real-time Motionsmotion records from the seismic stations;
(2)
TheEstablishing the building
ground motion recordsinventory database
can fully for the
describe thetarget region;
features of the ground motions with no
information loss. The densely distributed seismic stations andpredict
(3) Conducting the city-scale nonlinear time-history analysis to the seismic damage
communication networkofmake
the it
possible target region; and
to obtain the real-time ground motion records. After an earthquake, the ground motion
(4) Performing the regional seismic loss prediction to assess the seismic economic loss and repair
record near the epicenter can be quickly obtained through the seismic stations and communication
time of the target region.
network, and information, such as the station’s latitude, longitude, and recording time, can be collected
simultaneously.
2.2. Real-Time With the Ground
Recorded development
Motionsof monitoring and data-transforming technology, the densely
distributed strong motion network will cover more regions, and the ground motion data will be easier
The ground motion records can fully describe the features of the ground motions with no
to access in a timely manner after an earthquake.
information loss. The densely distributed seismic stations and communication network make it
2.3. possible
Building to obtain the real-time ground motion records. After an earthquake, the ground motion
Inventory Database
record near the epicenter can be quickly obtained through the seismic stations and communication
Based onand
network, theinformation,
Sixth Nationalsuch Population Census
as the station’s [10], longitude,
latitude, this workandconstructs a virtual
recording building
time, can be
inventory database
collected of cities With
simultaneously. in thethe
mainland of China.
development Specifically,
of monitoring according to the technology,
and data-transforming Sixth National
the densely
Population distributed
Census, strong motion
the number network
of buildings inwill
the cover
targetmore
regionregions, and the
classified byground motionofdata
the number stories,
will be type,
structural easier and
to access
yearin a timely
built manner
can be after an
obtained. earthquake.
Note that these data are classified by one attribute.
For example, the statistical data of the buildings for a typical region, Haidian District, are presented
2.3. Building Inventory Database
in Table 1. Therefore, the buildings are further divided into 33 categories according to the number
of stories,Based on thetype,
structural Sixthand
National Population
year built, such asCensus [10], thisstructure
the masonry work constructs a virtual
built before 1990building
with single
inventory database of cities in the mainland of China. Specifically, according
story, and the masonry structure built before 1990 with two or three stories. The proportions to the Sixth Nationalof the
Population
33 building Census,
types can bethedetermined
number of buildings in the
by solving thetarget regionequations
indefinite classified by thedescribe
that number this
of stories,
problem.
structural type, and year built can be obtained. Note that these data are classified by one attribute.
Then a virtual city with 1000 buildings that reflects the statistical characteristics of local buildings
For example, the statistical data of the buildings for a typical region, Haidian District, are presented
can be established. The building inventory database of each region can be established to serve the
in Table 1. Therefore, the buildings are further divided into 33 categories according to the number of
subsequent seismic damage prediction. Note that if the statistical data of each building can be obtained
stories, structural type, and year built, such as the masonry structure built before 1990 with single
for the target
story, andregion, then these
the masonry databuilt
structure can be directly
before 1990used
with to
twoestablish
or threethe analysis
stories. The model. In addition
proportions of the to
the cities in the mainland of China, other building inventory databases for other regions
33 building types can be determined by solving the indefinite equations that describe this problem. (e.g., Japan
and Then
the United States)
a virtual are under
city with construction.
1000 buildings As a result,
that reflects the proposed
the statistical method
characteristics ofcan
localbebuildings
further applied
can
to different regionsThe
be established. if the groundinventory
building motions and building
database inventory
of each region arecanavailable.
be established to serve the
subsequent seismic damage prediction. Note that if the statistical data of each building can be
obtained for the target region, then these data can be directly used to establish the analysis model. In
addition to the cities in the mainland of China, other building inventory databases for other regions
 Appl. Sci. 2019, 9, x FOR PEER REVIEW 4 of 16

(e.g., Japan and the United States) are under construction. As a result, the proposed method can be
Appl. Sci. 2019, 9, 3497 4 of 16
further applied to different regions if the ground motions and building inventory are available.

Table 1. Statistical data of the buildings for Haidian District, Beijing.

Table 1. Statistical data of the buildings for Haidian District, Beijing.
Number of stories 1 2–3 4–6 7–9 >9
Number of stories 1 2–3 4–6 7–9 >9
Proportions 20.6% 10.6% 36.3% 4.9% 27.6%
Proportions 20.6% 10.6% 36.3% 4.9% 27.6%
Year built <1990 1990–1999 >1999
Year built <1990 1990–1999 >1999
Proportions 30.3% 29.8% 39.9%
Proportions 30.3% 29.8% 39.9%
Steel and reinforced Other
Structuraltype
Structural type Steel and reinforced concrete Masonry
Masonry Wood
Wood Other structures
concrete structures
Proportions 47.9% 40.3% 11.6% 0.2%
Proportions 47.9% 40.3% 11.6% 0.2%

2.4.2.4.
City-Scale Nonlinear
City-Scale NonlinearTime-History
Time-HistoryAnalysis
Analysis
TheThecity-scale
city-scalenonlinear
nonlineartime-history
time-historyanalysis
analysis is used
used to toperform
performthe theseismic
seismicdamage
damage simulation
simulation
forfor
thethe
target
targetregion
region[11].
[11].InIngeneral,
general,the the buildings
buildings inin aa city
citycancanbebedivided
dividedintointotwo
two types:
types: Ordinary
Ordinary
multi-story
multi-story buildings and and
buildings ordinary tall buildings.
ordinary Multi-story
tall buildings. buildings
Multi-story often exhibit
buildings oftenshear deformation
exhibit shear
deformation
modes modes under
under earthquakes, earthquakes,
whereas whereas
tall buildings tall
will buildings
deform will deform modes.
in flexural-shear in flexural-shear
Thus, themodes.
multiple-
Thus, the multiple-degree-of-freedom
degree-of-freedom (MDOF) shear model(MDOF) will be shear model
used for the will be used for
multi-story the multi-story
buildings (Figure buildings
2a), and the
(Figure
MDOF 2a), and the MDOF
flexural-shear modelflexural-shear
will be applied model
to will
tall be applied (Figure
buildings to tall buildings
2b). In(Figure
the MDOF 2b). Inmodel,
the
MDOF model, the masses of the buildings are concentrated on their
the masses of the buildings are concentrated on their corresponding stories. Specifically, for the corresponding stories.
Specifically,
MDOF for the each
shear model, MDOF shear
story is model, each story
represented by a is represented
nonlinear by a as
spring, nonlinear
shown spring,
in Figure as shown
2a. Forinthe
Figure 2a. For the MDOF flexural-shear model, each story is discretized into
MDOF flexural-shear model, each story is discretized into a nonlinear shear spring and a nonlinear a nonlinear shear spring
and a spring,
flexural nonlinear flexural
which spring, which
are connected are connected
to each to each
other by rigid other
links, as by rigidin
shown links, as shown
Figure 2b. Theinnonlinear
Figure
2b. The nonlinear behavior of the structure is represented by the nonlinear inter-story force–
behavior of the structure is represented by the nonlinear inter-story force–displacement relationships
displacement relationships (Figure 2c). The tri-linear backbone curves recommended in the HAZUS
(Figure 2c). The tri-linear backbone curves recommended in the HAZUS report [7] are employed to
report [7] are employed to model the inter-story force–displacement relationships. Note that the
model the inter-story force–displacement relationships. Note that the parameter determination of the
parameter determination of the inter-story force–displacement relationships is critical for the
inter-story force–displacement relationships is critical for the rationality and accuracy of the simulation
rationality and accuracy of the simulation results, considering the limited available information for
results, considering the limited available information for buildings on a regional scale. The parameter
buildings on a regional scale. The parameter determination methods for buildings in China and the
determination
United Statesmethods
proposedforbybuildings
the authors in China and
[11] are the United
adopted in theStates proposed
city-scale by the
nonlinear authors [11]
time-history
areanalysis.
adoptedFor in buildings
the city-scale
in the other countries, the parameter determination procedure can countries,
nonlinear time-history analysis. For buildings in the other refer to
thethose
parameter determination procedure can refer to those two methods.
two methods.

Figure
Figure 2. 2.(a)
(a) Multiple-degree-of-freedom (MDOF)
Multiple-degree-of-freedom (MDOF) shear
shear model;
model; (b) MDOF
(b) MDOF flexural-shear
flexural-shear model;model;
and
and (c) trilinear backbone curve adopted in the MDOF
(c) trilinear backbone curve adopted in the MDOF model. model.

2.4.1. Parameter Determination Method for Buildings in China

The parameter determination method for buildings in China is based on the design codes and
statistics of extensive experimental and analytical results [12,13]. Because the buildings usually have a
regular layout along with the height, the stiffness and masses of different stories are assumed to be
uniform [14]. The elastic parameters of a building can be represented by the inter-story shear stiffness,
Appl. Sci. 2019, 9, 3497 5 of 16

k0 , and the mass, m, of each story. Equations (1) and (2) show the global stiffness [K] and mass matrices
[M] of a structure with a uniform stiffness and mass along the height [15].

−1

 2 
 −1 −1

2


 
 .. .. 
[K] = k0  −1 . .  = k [A]
 0 (1)
..
 
.
 

 2 −1 
−1 1


 
 1 
 

 1 

[M] = m
 1  = m[I ]

(2)
 .. 

 . 

1
 

The mass of each story, m, in Equation (2) can be determined based on the area of each story, A1 ,
and the mass per unit area, m1 (Equation (3)) [16]; m1 can be estimated according to the occupancy of
each story.
m = m 1 A1 (3)

The relationship among the stiffness, mass, and first vibration period, T1 , can be expressed using
Equation (4).
 [Φ1 ]T [I ][Φ1 ]  T
4π2 m  [Φ1 ] [I ][Φ1 ] 
   
2
k0 = mω1   =
 (4)
[Φ1 ]T [A][Φ1 ] T1 2  [Φ1 ]T [A][Φ1 ] 
 

where [Φ1 ] is the first mode vector. Given the stiffness matrix [K] and mass matrix [M], [Φ1 ] can be
computed using a generalized eigenvalue analysis. As shown in Equation (4), m and T1 are required
to obtain k0 . The vibration periods of different types of structures can be estimated using empirical
equations. For example, the fundamental period of a reinforced concrete (RC) frame can be calculated
using the empirical equation (Equation (5)) specified in the Chinese Code [17].
√
3
T1 = 0.25 + 0.00053H2 / B (5)

where H and B are the height and width of an RC frame, respectively. For other types of structures
(e.g., RC tall buildings, masonry structures, steel frames), the corresponding empirical vibration period
can also be adopted to determine the inter-story stiffness.
For engineering-designed structures (e.g., RC frames and reinforced masonry structures),
the design strength can be estimated according to the seismic design code. Therefore, an equivalent
lateral force analysis can be used to calculate the design shear force, V design, i , of each story, where i
is the story number [18,19]. Subsequently, according to related statistics of extensive experimental
and analytical results, the yield point, peak point, and softening point on the backbone curve can be
further obtained. For example, the yield point, peak point, and softening point of an RC frame can be
determined by Equations (6)–(8) as follows:

Vyield, i = Ω1 Vdesign, i (6)

Vpeak, i = Ω2 Vdesign, i (7)

Vultimate, i = Vpeak, i (8)

where V yield, i , V peak, i , and V ultimate i are the yield strength, peak strength, and ultimate strength,
respectively. Ω1 is the yield overstrength ratio of RC frames, which is determined according to the
Appl. Sci. 2019, 9, 3497 6 of 16

partial factor of steel reinforcement [20]. Ω2 is the peak overstrength ratio, which is determined by the
statistics of 155 pushover results of RC frames designed following the Chinese seismic design code.
The deformation parameters of RC frames, including the yield, peak, and ultimate deformations, can
be determined using the same procedure.
Non-engineered buildings (e.g., unreinforced masonry buildings and adobe buildings) lack a
design strength as the reference strength to establish the backbone curve in Figure 2c. Consequently,
the statistical strengths of different types of non-engineered buildings obtained from the literature are
used to establish the backbone curve. For example, Xiong et al. [13] proposed using the statistical
results of the peak strength per unit area of 1000 unreinforced masonry structures in China as the
reference strength. The other parameters on the backbone curve of unreinforced masonry buildings
are determined based on the statistical results of 97 unreinforced masonry wall experiments.

2.4.2. Parameter Determination for Backbone Curve Based on the HAZUS Data
Based on the HAZUS database, Lu et al. [15] also proposed a modeling approach by which
all parameters of the MDOF shear model can be determined from the basic building information
(i.e., number of stories, height, year built, structural type, floor area, and occupancy). Specifically,
the first vibration period of the building can be determined by the typical buildings presented in Tables
5.5 and 5.7 of FEMA [7]. The elastic parameters of a building, including the inter-story shear stiffness,
k0 , and the mass, m, are determined using Equations (1)–(4). Subsequently, the inter-story backbone
curve parameters of story i in Figure 3 are determined as follows:

4π2 m
k0,i = λ (9)
T1 2

Vy,i = SAy · α1 · m · g · N · Γi (10)

SAu − SAy SDy
ηi = · (11)
SDu − SDy SAy
SAu
βi = (12)
SAy
∆c,i = δCo · h (13)

where g is the acceleration of gravity; δCo is the inter-story drift ratio at the threshold of the complete
damage state, as suggested by HAZUS [7]; h is the story height; (SDy , SAy ) and (SDu, SAu ) are the
yield capacity point and ultimate capacity point, respectively, of the capacity curve suggested by
HAZUS [7], which are functions of the design intensity and year built; α1 is the mode factor suggested
by HAZUS [7]; and Γi is the ratio between the inter-story shear strength of the ith story, (V y,i ) and that
of the ground story (V y,1 ), which is calculated as follows:

V y,i
Γi = (14)
V y,1

A single-parameter pinching model (Figure 4) proposed by Steelman and Hajjar [21] is adopted
to represent the pinching behavior subjected to cyclic loads. Five damage states, ranging from none,
slight, and moderate, to extensive and complete damage, are considered, which are identical to the
damage states defined by HAZUS. The inter-story drift ratio is used as the threshold for each structural
damage state, and the values for different structural types are based on Table 5.9 of HAZUS [7].
slight,Appl.
and Sci.moderate,
2019, 9, x FORto extensive
PEER REVIEW and complete damage, are considered, which are identical to the
7 of 16
damage states defined by HAZUS. The inter-story drift ratio is used as the threshold for each
slight, and moderate, to extensive and complete damage, are considered, which are identical to the
structural damage state, and the values for different structural types are based on Table 5.9 of HAZUS
damage states defined by HAZUS. The inter-story drift ratio is used as the threshold for each
[7].
structural
Appl. Sci. damage state, and the values for different structural types are based on Table 5.9 of HAZUS7 of 16
2019, 9, 3497
[7].
V
V
βVy
βVy
ηk0
Vy ηk0
Vy

k0
k0

OO ΔcΔc Δ Δ
Figure
Figure 3.
Figure Backbone
3. 3. Backbonecurve
Backbone curve of
curve the
of the MDOF
theMDOF shear
MDOFshear
shear model.
model.
model.

Break Point Line Break Point

Break Point Line Break Point

Full Bilinear Envelope

Pinching Envelope
Full Bilinear Envelope
Pinching
Figure 4. Diagram of Envelope
the pinching envelope of the pinching model.
Figure 4. 4. Diagram of of the
the pinching
pinching envelope
envelope of of the
the pinching
pinching model.
model.
The classicalFigureRayleighDiagramdamping model is adopted in the analysis. The empirical values of the
damping
The ratio Rayleigh
classical are estimated according
damping modelto theis structural
adopted type ofanalysis.
in the buildingThe[15]. For example, theof the
The classical
damping ratio Rayleigh damping
of steel structures model
equals is
0.05. adopted
It should be in the
the
noted analysis.
that the The
empirical
empirical values
values
soil-structure-interaction of the
damping
damping ratio are
ratio are estimated according to the structural type of the building [15]. For example,
(SSI) effects andestimated according aren’t
structural eccentricity to theconsidered
structuralintype this of the building [15]. For example, the
work.
the damping
damping ratioratio of steel
of steel structures
structures equals
equals 0.05.It Itshould
0.05. shouldbebenoted noted thatthe the soil-structure-interaction
The reliability of the city-scale nonlinear time-history analysis thatmethod issoil-structure-interaction
further validated by
(SSI) effects
(SSI) comparingand
effects andthe structural
structural
simulation eccentricity
eccentricity
results with aren’t
aren’t considered
considered
earthquake in this work.
in this work. experimental results, and a
site investigations,
The
The reliability
largereliability
number of of of the
the city-scale
numerical nonlinear
results [11].
city-scale time-history
With outstanding
nonlinear time-history analysis method
computational
analysis method is further
efficiency,
is further validated
this method can by
validated by
comparing
be used the simulation
well for results
post-earthquake with earthquake
emergency site
responses. investigations,
There
comparing the simulation results with earthquake site investigations, experimental results, and aexists experimental
an inherent results,
uncertainty and
in a
the large
number
large of numerical
seismic
number performance
of numerical results [11]. [11].
of buildings,
results With outstanding
which
With is considered
outstanding computational
in this method
computational efficiency, thisthis
method
by incorporating
efficiency, method cancan
the be
used parametric
well forfor uncertainty
post-earthquake of the building
emergency backbone
responses.curve [22]. Consequently, the proposed method can
be used well post-earthquake emergency responses.There Thereexistsexistsanan inherent
inherent uncertainty
uncertainty in the
in the
notperformance
seismic only provide the of building response
buildings, which using
is the median
considered in value
this of the backbone
method by curve parameters
incorporating the but
parametric
seismic performance
also provide the
of buildings,
responses
which
with thecurve median
is value
considered in this deviation
± one standard
method by tomethod
incorporating
accountcan for not
the
the only
uncertainty
parametric of the
uncertaintybuilding of the backbone
building backbone [22]. Consequently,
curve the proposed
[22]. Consequently, the proposed method can
parametric uncertainty, which is crucial for scientific decision-making.
provide
not only the building
provide response
the building using the
response median
using value of
thebuildings
median the of
value backbone
thearea curvecurve
backbone parameters
parameters but also
The nonlinear time-history analysis of the in the target is implemented using the but
provide the
also provide responses
the responses with the median
with the value
theseismic
median ± one standard
value Subsequently, deviation
± one standard to account
deviation for the parametric
ground motions obtained from network. the time historiestoofaccount
the seismicfor the
uncertainty,
response
parametric which
uncertainty,is crucial
of each story which for
in everyscientific
building
is crucial decision-making.
for can be obtained.
scientific According to the engineering demand
decision-making.
The
The nonlinear
parameters
nonlinear(EDPs) time-history
and the damage
time-history analysis
analysis of the
criteria
of the buildings
[11], the damage
buildings in the
in the target
state area
of each
target area is implemented
implemented
building
is in the regionusing the
is the
using
ground
ground motions
determined, obtained
motions based
obtained from
on which,
from thethe seismic
theseismic network.
destructive
network. Subsequently,
powerSubsequently,
of the ground the the time
motion histories
timetohistories
the target of the
of area seismic
is
the seismic
response of
evaluated.each
To story
make fullin every
use of building
the real-time can be obtained.
earthquake ground
response of each story in every building can be obtained. According to the engineering demand According to
motions obtained the engineering
from the densely demand
distributed
parameters (EDPs)seismicandstations,
the damage the destructive
damage criteria[11], powers
[11], of
thedamage groundstate
damage motionseachobtained from indifferent
parameters (EDPs) and the criteria the state ofofeach building
building in thethe region
region is
seismic stations
is determined, based canonbewhich,
evaluated the by inputting the
destructive ground
power of themotions
ground one-by-one
motion into to thethe target
building area is
determined, based on which, the destructive power of the ground motion to the target area is
evaluated. To
evaluated. To make
make full
full use
use of
of the
the real-time
real-time earthquake ground motions
earthquake ground motions obtained
obtained from
from the
the densely
densely
distributed seismic stations, the destructive powers of ground motions obtained from different
distributed seismic stations, the destructive powers of ground motions obtained from different seismic
stations can
seismic be evaluated
stations by inputting
can be evaluated by the groundthe
inputting motions one-by-one
ground motions into the building
one-by-one into models of the
the building
target region. The distribution of building damage ratios under different station records can be given
subsequently, which provides an essential reference for post-earthquake rescue work. For example,
the destructive power of ground motions of the 08/13/2018 M5.0 Yunnan Tonghai earthquake can be
illustrated intuitively, as shown in Figure 5.
Appl. Sci.The
2019,human
9, x FOR sense of floor acceleration is highly important in the resilience assessment8 of
PEER REVIEW of 16
communities under moderate seismic actions. Based on the comfort criteria [23] and floor acceleration
models
computedof thebytarget region. The
the nonlinear distribution
time-history of building
analysis, damage
the human ratios
sense under different
of different ground station
motionsrecords
can
canbebeobtained.
given subsequently, whichofprovides
The distribution an essential reference
human uncomfortableness underfor the
post-earthquake
ground motionsrescue
of work.
the
For11/26/2018
Appl. example, the destructive
Sci. 2019, 9,M6.2
3497 Taiwan Strait power of ground
earthquake motions
is shown of the 6.08/13/2018
in Figure Although M5.0 Yunnanratio
the damage Tonghai
8of
of 16
buildings can
earthquake under bethis earthquake
illustrated is very small,
intuitively, the ratio
as shown of human
in Figure 5. uncomfortableness is still high.
The human sense of floor acceleration is highly important in the resilience assessment of
communities under moderate seismic actions. Based on the comfort criteria [23] and floor acceleration
computed by the nonlinear time-history analysis, the human sense of different ground motions can
be obtained. The distribution of human uncomfortableness under the ground motions of the
11/26/2018 M6.2 Taiwan Strait earthquake is shown in Figure 6. Although the damage ratio of
buildings under this earthquake is very small, the ratio of human uncomfortableness is still high.

Figure5.5.Destructive
Figure Destructivepower
power of
of ground
ground motions
motionsof
ofthe
the08/13/2018
08/13/2018M5.0
M5.0Yunnan
YunnanTonghai
Tonghaiearthquake.
earthquake.

The human sense of floor acceleration is highly important in the resilience assessment of
communities under moderate seismic actions. Based on the comfort criteria [23] and floor acceleration
computed by the nonlinear time-history analysis, the human sense of different ground motions can be
obtained. The distribution of human uncomfortableness under the ground motions of the 11/26/2018
M6.2 Taiwan Strait earthquake is shown in Figure 6. Although the damage ratio of buildings under
this earthquake is very small,
Figure 5. Destructive powerthe ratio ofmotions
of ground humanofuncomfortableness
the 08/13/2018 M5.0 is still high.
Yunnan Tonghai earthquake.

Figure 6. Distribution of human uncomfortableness under the ground motions of the 11/26/2018 M6.2
Taiwan Strait earthquake.
 2.5. Regional Seismic Loss Prediction and Resilience Assessment

2.5.1. Regional Seismic Loss Prediction Using Conventional Method

Based on the damage state of each building, the seismic loss can be predicted using the
conventional
Appl. loss prediction method. For example, according to the National Standard of China,
Sci. 2019, 9, 3497 9 of 16
“Post-Earthquake Field Works-Part 4: Assessment of Direct Loss” [24], the house damage loss Lh
and decoration damage loss Ld are calculated using Equations (15) and (16), respectively:
2.5. Regional Seismic Loss Prediction and Resilience Assessment
Lh = S × Dh × P (15)
2.5.1. Regional Seismic Loss Prediction Using Conventional Method
Based on the damage state L = γbuilding,
ofd each 1 × γ 2 ×the
(ξ seismic
× S ) ×loss η predicted
× (be
Ddcan × P ) using the conventional
(16)
loss prediction
where S is themethod.
buildingFor
areaexample, according
(m2); Dh and to the
Dd are the lossNational
ratios of Standard
the house of
andChina, “Post-Earthquake
decoration damage,
Field Works-Part 4: Assessment of Direct Loss” [24], the house damage loss L and decoration
given a damage state; P is the building replacement cost; γ 1 is the correction factor considering
h damage
loss Ld are calculated using Equations (15) and (16), respectively:
different economic conditions of different regions; γ 2 is the building function correction factor; ξ
is the proportion of buildings with mid-to-high-quality
Lh = S × Dh ×decoration;
P and η is the ratio of the building
(15)
decoration cost to the building construction cost. The values of γ 1 , γ 2 , ξ , η can be found in Table
L = γ1 × γ2 × (ξ × S) × Dd × (η × P) (16)
A.1–A.4 in GB/T 18208.4 [24]. d
where S is the building area (m2 ); Dh and Dd are the loss ratios of the house and decoration damage,
2.5.2aRegional
given damageResilience Assessment
state; P is the buildingUsing FEMA P-58 cost; γ is the correction factor considering
replacement 1
different economic conditions
Furthermore, based onofthe different
FEMAregions; γ2 is the(next-generation
P-58 method building function correction
seismic factor; ξ is
performance
theassessment
proportionmethod
of buildings with mid-to-high-quality decoration; and η is the ratio
of buildings) and the city-scale nonlinear time-history analysis, a practical of the building
decoration
approach cost to the building
for regional resilienceconstruction
assessment iscost. The [25]
proposed values of γa1 ,more
to give γ2 , ξ,detailed
η can be found
seismic inand
loss Table
repair-time
A.1–A.4 in GB/T prediction. The main process of the method is as follows: (a) Analyze the building
18208.4 [24].
response to determine the EDPs on each story of each building using the nonlinear MDOF models,
2.5.2.
andRegional Resilience
(b) calculate Assessment
the economic loss andUsing
repairFEMA
time ofP-58
components based on the building performance
models and the fragility
Furthermore, based ondata
theprovided
FEMA P-58 in the FEMA(next-generation
method P-58 document. The flowchart
seismic of the repair
performance cost
assessment
calculation is shown in Figure 7. Using this method, the damage states of components on different
method of buildings) and the city-scale nonlinear time-history analysis, a practical approach for regional
stories can be obtained, and the loss caused by the floor displacement, acceleration, and residual
resilience assessment is proposed [25] to give a more detailed seismic loss and repair-time prediction.
displacement can be considered.
The main process of the method is as follows: (a) Analyze the building response to determine the
The repair time is calculated according to the FEMA P-58 method. Specifically, the
EDPs on each story of each building using the nonlinear MDOF models, and (b) calculate the economic
corresponding repair time is determined according to the damage state of each component, and then
loss
theand repair
repair time
time of components
of components based
in each on are
story the summarized.
building performance
Using the models
parallel and thestrategy,
repair fragilitythe
data
provided
maximum repair time of each story is taken as the repair time of the building. Using high-efficiency 7.
in the FEMA P-58 document. The flowchart of the repair cost calculation is shown in Figure
Using this method,
computational the damage
codes, states of
this method cancomponents on different
quickly calculate stories can
the economic be obtained,
loss and repairand theof
time loss
caused by the floor displacement,
earthquake-stricken areas. acceleration, and residual displacement can be considered.

Figure 7. Calculation of repair cost for a building component using the FEMA-P58 methodology.
Figure 7. Calculation of repair cost for a building component using the FEMA-P58 methodology.
The repair time is calculated according to the FEMA P-58 method. Specifically, the corresponding
One of the critical challenges of using the FEMA-P58 method in a region is the assembly of
repair time is determined according to the damage state of each component, and then the repair time of
performance groups (PGs). The performance model of a building contains its basic information with
components in each story are summarized. Using the parallel repair strategy, the maximum repair time
of each story is taken as the repair time of the building. Using high-efficiency computational codes,
this method can quickly calculate the economic loss and repair time of earthquake-stricken areas.
One of the critical challenges of using the FEMA-P58 method in a region is the assembly of
performance groups (PGs). The performance model of a building contains its basic information with
both structural and nonstructural PGs. The types and quantities of building components can be
obtained using the following three methods: (a) Field survey data and building design drawings,
(b) building information models (BIM), and (c) the geographic information system (GIS) database.
Appl. Sci. 2019, 9, x FOR PEER REVIEW 10 of 16
Appl. Sci. 2019, 9, 3497 10 of 16
obtained using the following three methods: (a) Field survey data and building design drawings, (b)
building information models (BIM), and (c) the geographic information system (GIS) database.
(a) Field survey data and building design drawings
(a) Field survey data and building design drawings
The type
The type and
and quantity
quantity of
of each
eachstructural
structural PG
PGcan
canbe
beobtained
obtainedfrom
fromthe
thestructural
structuralandandarchitectural
architectural
drawings of the building. The nonstructural PGs can be determined using the field
drawings of the building. The nonstructural PGs can be determined using the field survey. Notesurvey. Note that
that
some effort
some effort is
is required
requiredtotocollect
collectthe
theinformation.
information.However,
However,the thedata
datacan
canbebe
implemented
implemented in in
parallel by
parallel
groups of people with basic knowledge of architectural and structural engineering.
by groups of people with basic knowledge of architectural and structural engineering.
(b)
(b) Building information
Building models
information models

The detailed
The detailed building
building datadata can
can be be automatically
automatically obtained
obtained from
from thethe BIM
BIM in
in which
which the the building
building
components have different levels of development (LODs). The determination
components have different levels of development (LODs). The determination of the component type of the component type
and the development of a component vulnerability function when the information
and the development of a component vulnerability function when the information is incomplete are is incomplete
are proposed
proposed to produce
to produce an acceptable
an acceptable loss loss prediction
prediction [26,27].
[26,27]. Specifically,
Specifically, the PGs
the PGs in FEMA
in FEMA P-58P-58
are
are organized as different classification trees. For example, the classification tree
organized as different classification trees. For example, the classification tree of a gypsum wall board of a gypsum wall
board (GWB)
(GWB) partition
partition is shownis shown in Figure
in Figure 8. According
8. According to thetoLODs
the LODs
of the ofBIMs,
the BIMs, all PGs
all PGs thatthat cannot
cannot be
be determined
determined as leaf
as leaf nodes
nodes in in
thethe classification
classification tree
tree duetotolack
due lackofofinformation
informationwill willbe be treated
treated as
as
“potential fragility classifications.” A Monte Carlo simulation is subsequently
“potential fragility classifications.” A Monte Carlo simulation is subsequently implemented implemented considering
all “potentialallfragility
considering classifications.”
“potential Consequently,
fragility classifications.” the loss can be
Consequently, thepredicted
loss can even when a even
be predicted very coarse
when
BIM is available. In addition, the modeling rules and information extraction for
a very coarse BIM is available. In addition, the modeling rules and information extraction for BIM areBIM are proposed to
obtain the component information [26,27].
proposed to obtain the component information [26,27].

Figure 8.
Figure Classificationtree
8. Classification tree of
of aa gypsum
gypsum wall
wall board
board (GWB)
(GWB) partition
partition component.
component.

(c) GIS
(c) GIS database
database

According to the building inventory of the GIS database, the structural component quantity is
According to the building inventory of the GIS database, the structural component quantity is
estimated based on the statistics from the available literature and design drawings. The nonstructural
estimated based on the statistics from the available literature and design drawings. The nonstructural
PG information can be identified according to the normative quantity information provided by
PG information can be identified according to the normative quantity information provided by
Appendix F of FEMA [28]. Subsequently, the structural and nonstructural PGs can be estimated.
Appendix F of FEMA [28]. Subsequently, the structural and nonstructural PGs can be estimated.
A new trend to efficiently and accurately establish the PGs of buildings on a regional scale uses the
A new trend to efficiently and accurately establish the PGs of buildings on a regional scale uses
City Information Model (CIM). CIM is defined as the integration of GIS and BIM [29]. The CIM of the
the City Information Model (CIM). CIM is defined as the integration of GIS and BIM [29]. The CIM
earthquake-stricken area can be pre-established, which will provide valuable data for the community
of the earthquake-stricken area can be pre-established, which will provide valuable data for the
resilience assessment on different scales.
community resilience assessment on different scales.
To demonstrate the resilience assessment method for a region, the seismic economic loss and
To demonstrate the resilience assessment method for a region, the seismic economic loss and
repair time of Tsinghua Campus [25] were calculated. The ground motion recorded at the Jiuzhaigou
repair time of Tsinghua Campus [25] were calculated. The ground motion recorded at the Jiuzhaigou
Baihe station was input to 619 buildings of Tsinghua Campus. The distribution of the median building
Baihe station was input to 619 buildings of Tsinghua Campus. The distribution of the median
loss ratios and repair/rebuilding times are shown in Figure 9. The total loss ratio is 0.576%, which is
building loss ratios and repair/rebuilding times are shown in Figure 9. The total loss ratio is 0.576%,
very small. The repair time of the campus is 15 days with parallel repair strategies. The results provide
which is very small. The repair time of the campus is 15 days with parallel repair strategies. The
a valuable reference for the resilience assessment of Tsinghua Campus.
results provide a valuable reference for the resilience assessment of Tsinghua Campus.
Appl. Sci. 2019, 9, 3497 11 of 16
Appl. Sci. 2019, 9, x FOR PEER REVIEW 11 of 16

(a) (b)
Figure9.9.Distribution
Figure Distribution of median
of median (a) building
(a) building loss ratios
loss ratios and (b)and (b) repair/rebuild
repair/rebuild time for Campus.
time for Tsinghua Tsinghua
Campus.
2.6. High-Performance Computing for Post-Earthquake Emergency Response
2.6 High-Performance
High-performance Computing for (HPC)
computing Post-Earthquake Emergency
is incorporated Response the real-time city-scale THA
to implement
for theHigh-performance
post-earthquake emergency
computingresponse.
(HPC) is The OpenMP to
incorporated library [30] is the
implement used to parallelize
real-time the code
city-scale THA
of the
for thecity-scale nonlinearemergency
post-earthquake time-history analysis.The
response. Only 136 s are
OpenMP required
library to used
[30] is complete the computation
to parallelize the code
of
ofathe
ground motion
city-scale recordtime-history
nonlinear (with an Intel Xeon E5
analysis. 2630136
Only @2.40
s areGHz CPU to
required and 64 GB of
complete RAM),
the which
computation
satisfies the requirement for the post-earthquake emergency response.
of a ground motion record (with an Intel Xeon E5 2630 @2.40 GHz CPU and 64 GB of RAM), which
Furthermore,
satisfies cloud computing
the requirement is introduced toemergency
for the post-earthquake perform coarse-grain
response. parallel analysis. Specifically,
a numberFurthermore, cloud computing is introduced to performoncoarse-grain
of virtual computers are quickly established (in minutes) the cloud computing
parallel platform
analysis.
(e.g., Aliyun Cloud
Specifically, a numberor Tencent Cloud)
of virtual after an earthquake.
computers are quickly Subsequently,
established (inthe analysesonof the
minutes) different
cloud
ground motion records are assigned to different virtual computers. Each virtual
computing platform (e.g., Aliyun Cloud or Tencent Cloud) after an earthquake. Subsequently, the computer has nearly
the same performance
analyses as the motion
of different ground local computing
records are environment. Thus, regardless
assigned to different of how many
virtual computers. Eachground
virtual
motions
computer arehas
to be computed,
nearly the samethe performance
time consumptionas theislocal
almost the same environment.
computing as that of a single ground
Thus, motion.
regardless of
Meanwhile, the computational cost is less than one US dollar for each virtual
how many ground motions are to be computed, the time consumption is almost the same as that of computer. Such a flexible
cloud computing
a single platformMeanwhile,
ground motion. makes the real-time analysis ofcost
the computational a number
is less of ground
than one USmotions
dollartechnologically
for each virtual
and economically
computer. Such afeasible.
flexible cloud computing platform makes the real-time analysis of a number of
ground motions technologically and economically feasible.
3. Applications in Earthquake Emergency Response
3. Applications
3.1. inApplications
Overview of the Earthquake Emergency Response

When anofearthquake
3.1 Overview occurs, the ground motion will be collected in a timely manner from the
the Applications
strong ground motion network. The real-time city-scale THA will be conducted for the target region,
When an earthquake occurs, the ground motion will be collected in a timely manner from the
and the analysis results will be fed back to the decision-makers and reported on the internet in a short
strong ground motion network. The real-time city-scale THA will be conducted for the target region,
time. A program is developed named as “Real-time Earthquake Damage Assessment using City-scale
and the analysis results will be fed back to the decision-makers and reported on the internet in a short
Time-history analysis” (“RED-ACT” for short) to automatically implement the above workflow. This
time. A program is developed named as “Real-time Earthquake Damage Assessment using City-scale
software is programmed in C++ and runs on the Windows operating system. To date, RED-ACT
Time-history analysis” (“RED-ACT” for short) to automatically implement the above workflow. This
system has been applied to several earthquakes in China and other countries around the world, as listed
software is programmed in C++ and runs on the Windows operating system. To date, RED-ACT
in Table 2. Typical applications will be introduced as follows.
system has been applied to several earthquakes in China and other countries around the world, as
listed in Table 2. Typical applications will be introduced as follows.

Table 2. Applications of the RED-ACT system.

ID Earthquake Name ID Earthquake Name

Appl. Sci. 2019, 9, 3497 12 of 16
Appl. Sci. 2019, 9, x FOR PEER REVIEW 12 of 16

1 12/08/2016 M6.2 Xinjiang Table 2. Applications

Hutubi earthquake of the
18 RED-ACT system.
11/26/2018 M6.2 Taiwan Strait earthquake
2 12/18/2016 M4.3 Shanxi Qingxu earthquake 19 12/08/2018 M4.5 Xinjiang Changji earthquake
ID Earthquake Name ID Earthquake Name
3 03/27/2017 M5.1 Yunnan Yangbi earthquake 20 12/16/2018 M5.7 Sichuan Yibin earthquake
4 1 12/08/2016 M6.2 Xinjiang Hutubi earthquake
08/08/2017 M7.0 Sichuan Jiuzhaigou earthquake 21 18 11/26/2018 M6.2 Taiwan Strait
12/20/2018 M5.2 Xinjiang earthquake
Kizilsu earthquake
2 12/18/2016 M4.3 Shanxi Qingxu earthquake 19 12/08/2018 M4.5 Xinjiang Changji earthquake
5 09/30/2017 M5.4 Sichuan Qingchuan earthquake 22 01/03/2019 M5.3 Sichuan Yibin earthquake
3 03/27/2017 M5.1 Yunnan Yangbi earthquake 20 12/16/2018 M5.7 Sichuan Yibin earthquake
64 02/06/2018 M6.5Sichuan
08/08/2017 M7.0 TaiwanJiuzhaigou
Hualien earthquake
earthquake 23
21 01/07/2019 M4.8Xinjiang
12/20/2018 M5.2 Xinjiang Jiashiearthquake
Kizilsu earthquake
75 02/12/2018 M4.3 Hebei Yongqing earthquake
09/30/2017 M5.4 Sichuan Qingchuan earthquake 24
22 04/16/2016 M7.3 Japan Kumamoto earthquake
01/03/2019 M5.3 Sichuan Yibin earthquake
8 6 02/06/2018 M6.5 Taiwan Hualien earthquake
05/28/2018 M5.7 Jilin Songyuan earthquake 23
25 01/07/2019 M4.8 Xinjiang Jiashi earthquake
08/24/2016 M6.2 Italy earthquake
97 02/12/2018M5.0
08/13/2018 M4.3Yunnan
Hebei Yongqing
Tonghaiearthquake
earthquake 24
26 04/16/2016
11/13/2016M7.3M8.0Japan
New Kumamoto earthquake
Zealand earthquake
8 05/28/2018 M5.7 Jilin Songyuan earthquake 25 08/24/2016 M6.2 Italy earthquake
10 08/14/2018 M5.0 Yunnan Tonghai earthquake 27 09/20/2017 M7.1 Mexico earthquake
9 08/13/2018 M5.0 Yunnan Tonghai earthquake 26 11/13/2016 M8.0 New Zealand earthquake
1110 09/04/2018 M5.5 Xinjiang Jiashi earthquake
08/14/2018 M5.0 Yunnan Tonghai earthquake 28
27 11/23/2017
09/20/2017 M7.8
M7.1 Iraq earthquake
Mexico earthquake
1211 09/08/2018 M5.9
09/04/2018 M5.5Yunnan
XinjiangMojiang earthquake
Jiashi earthquake 29
28 06/18/2018 M6.1M7.8
11/23/2017 Japan
IraqOsaka earthquake
earthquake
1312 09/12/2018 M5.3
09/08/2018 M5.9Shanxi
Yunnan Ningqiang earthquake 29
Mojiang earthquake 30 09/06/2018
06/18/2018M6.9
M6.1Japan
JapanHokkaido earthquake
Osaka earthquake
1413 09/12/2018 M5.3
10/16/2018 M5.4Shanxi Ningqiang
Xinjiang earthquake
Jinghe earthquake 30
31 09/06/2018 M5.4
10/26/2018 M6.9 Japan
JapanHokkaido
Hokkaidoearthquake
earthquake
1514 10/16/2018
10/31/2018 M5.4Sichuan
M5.1 XinjiangXichang
Jinghe earthquake
earthquake 31
32 10/26/2018 M5.4M7.0
12/01/2018 JapanAlaska
Hokkaido earthquake
earthquake
15 10/31/2018 M5.1 Sichuan Xichang earthquake 32 12/01/2018 M7.0 Alaska earthquake
16 11/04/2018 M5.1 Xinjiang Atushi earthquake 33 01/03/2019 M6.2 Japan Kumamoto earthquake
16 11/04/2018 M5.1 Xinjiang Atushi earthquake 33 01/03/2019 M6.2 Japan Kumamoto earthquake
1717 11/25/2018
11/25/2018M5.1
M5.1Xinjiang Boleearthquake
Xinjiang Bole earthquake

3.2 2017 M7.0 Jiuzhaigou Earthquake

3.2. 2017 M7.0 Jiuzhaigou Earthquake
The seismic damage assessment of the 2017 Jiuzhaigou earthquake is a typical application case
The seismic damage assessment of the 2017 Jiuzhaigou earthquake is a typical application case [31].
[31]. After the earthquake, several sets of ground motion records were obtained from the seismic
After the earthquake, several sets of ground motion records were obtained from the seismic network,
network, and the seismic damage prediction of the target region was completed in 2 h (including the
and the seismic damage prediction of the target region was completed in 2 h (including the time for the
time for the data checking and report editing/publishing) by using the method proposed. The
data checking and report editing/publishing) by using the method proposed. The predicted damage
predicted damage of a typical town and country in the Aba region under the ground motion of the
of a typical town and country in the Aba region under the ground motion of the Jiuzhaigou Baihe
Jiuzhaigou Baihe station is shown in Figure 10. The predicted results show that the buildings in the
station is shown in Figure 10. The predicted results show that the buildings in the disaster area may be
disaster area may be damaged to some extent, but the ratio of collapse is very small, which is
damaged to some extent, but the ratio of collapse is very small, which is consistent with the actual
consistent with the actual post-earthquake site investigations [32]. The results provide a useful
post-earthquake site investigations [32]. The results provide a useful reference for the earthquake
reference for the earthquake emergency response and scientific decision-making of earthquake
emergency response and scientific decision-making of earthquake disaster relief.
disaster relief.
100% 100%
Damage probability distribution
Damage probability distribution

75% 75%

50% 50%

25% 25%

0% 0%

None Slight Moderate Extensive Collapse None Slight Moderate Extensive Collapse
(a) (b)
Figure
Figure 10.
10. Seismic
Seismicresults
results of
of aa(a)
(a)typical
typicaltown
townand
and(b)
(b)country
countryin
inthe
theAba
Abaregion
regionsubjected
subjected the
the ground
ground
motion
motion from
from the Jiuzhaigou Baihe station.

3.3 2018 Mw 7.0 Anchorage Earthquake

The seismic damage assessment of the 2018 Mw 7.0 Anchorage earthquake is another typical
application case [33]. On 30 November 2018 (local time), a Mw 7.0 earthquake occurred in Alaska, the
United States. The epicenter was at 61.35 N, 150.06 W with a depth of 40 km [34]. Six ground motions
of the Anchorage earthquake event were recorded. The ground motions recorded at the 8047 station
Appl. Sci. 2019, 9, 3497 13 of 16

Appl. Sci. 2019, 9, x FOR PEER REVIEW 13 of 16

3.3. 2018 Mw 7.0 Anchorage Earthquake
The seismic damage assessment of the 2018 Mw earthquake is another typical
w 7.0 Anchorage earthquake

application case
application case[33].
[33].On
On3030November
November2018 2018 (local
(local time),
time), aMa wM7.0
w 7.0 earthquake
earthquake occurred
occurred in Alaska,
in Alaska, the
the United
United States.
States. The epicenter
The epicenter was atwas at N,
61.35 61.35 N, 150.06
150.06 W withWa with
depthaofdepth
40 kmof[34].
40 kmSix[34].
groundSixmotions
ground
motions
of theSci.
Appl. of the
9, x Anchorage
Anchorage
2019, earthquake
FOR earthquake
PEER REVIEW event wereevent were The
recorded. recorded.
groundThe ground
motions motions
recorded recorded
at the at ofthe
8047 station
13 16
8047 station
(61.189 (61.189W,
N, 149.802 N, 149.802
shown in W,Figure
shown 11)
in Figure 11) are
are typical typicalmotions
ground ground motions
[35]. The[35].
peakThe peak
ground
(61.189 accelerations
ground N, 149.802
accelerations (PGAs)W, (PGAs)
of shown in and
Figure
of horizontal
horizontal 11) are
and
vertical typical
vertical ground
components
components motions [35]. The were
of the 8047-ground
of the 8047-ground motion peak
motionground
were
807.162
accelerations
807.162 cm/s 2 (PGAs)
and cm/s
367.243 cm/s2 , respectively.
of2, horizontal and vertical components
The of shown
ground motions the 8047-ground
areinshown motion were 807.162
cm/s 2 and 367.243 respectively. The ground motions are Figurein11.
Figure 11.
cm/s2 and 367.243 cm/s2, respectively. The ground motions are shown in Figure 11.

Figure 11.
Figure 11. (a)
(a) Location
(a) Location of
Location of the
of the 8047-ground
8047-ground motion
8047-ground motion station,
motion station, and
station, and the
and the ground
the ground motion
motion recorded
motion recorded by
by the
the
8047station:
8047 station:(b)(b)EW
(b) EWdirection,
EW direction,(c)
direction, (c)NS
NSdirection,
direction,(d)
(d)UD
UDdirection.
direction.

Using the
Using the
Using ground
the ground motions
ground motions obtained
motions obtained from
obtained from the
the strong
from the strong motion
strong motion networks
motion networks and
networks and
and thethe
the city-scale
city-scale
city-scale
nonlinear time-history
nonlinear time-history
nonlinear analysis,
time-history analysis, the
analysis, the “RED-ACT”
the “RED-ACT” system
“RED-ACT” system predicted
system predicted the
predicted the damage
the damage ratio
damage ratio and
and human
ratio and human
human
uncomfortableness
uncomfortableness distribution of the buildings
buildings near different stations
stations in
in less
less than
than
uncomfortableness distribution of the buildings near different stations in less than 1 h, as shown in1
1 h,
h, as
as shown
shown in
in
Figures 12
Figures12
Figures and
12and 13.
and13. The
13.The post-earthquake
Thepost-earthquake investigation
post-earthquakeinvestigation showed
investigationshowed
showedthatthat this
thatthis Mw
thisMw
Mw7.0 7.0 earthquake
7.0earthquake produced
earthquakeproduced
produced
less-than-expected damage
less-than-expected damage
less-than-expected damage to to buildings,
to buildings, including
buildings, including businesses,
including businesses, homes,
businesses, homes,
homes, andand schools
and schools
schools inin
in downtown
downtown
downtown
Anchorage,
Anchorage,with
Anchorage, withmost
with most
mostdamage
damage
damagelimited to nonstructural
limited
limited to elements
to nonstructural
nonstructural and contents
elements
elements [34], which
and contents
and contents is consistent
[34],
[34], which isis
which
with the
consistentprediction
consistentwith
withthe given by
theprediction the
predictiongivenproposed
givenby bythe method.
theproposed
proposedmethod.
method.

(a)
(a)

Figure 12. Cont.

Appl. Sci. 2019, 9, 3497 14 of 16
Appl. Sci. 2019, 9, x FOR PEER REVIEW 14 of 16

Appl. Sci. 2019, 9, x FOR PEER REVIEW 14 of 16

(b)
(b)

Figure 12. 12.

12.
Figure Damage
Damage
Damageratio
ratio distribution
ratio distributionof
distribution of the
of the buildings
the buildings neardifferent
buildings near differentstations
stations
of of the
thethe 2018
20182018 Mww 7.0
Mw M
7.0
Anchorage earthquake: (a) Global
Anchorage earthquake: (a) Global view, view, and
view, and
and (b)(b) local
(b) local view.
local view.
view.

Figure
Figure 13. Distribution
13. Distribution of of humanuncomfortableness
human uncomfortableness under
underthe
theground
groundmotions
motionsof the 20182018
of the Mw M7.0w 7.0
Anchorage
Anchorage earthquake.
earthquake. of human uncomfortableness under the ground motions of the 2018 Mw 7.0
Figure 13. Distribution
Anchorage earthquake.
4. Conclusions
4. Conclusions
Based
4. Conclusions
Based on the
on the city-scale
city-scale nonlineartime-history
nonlinear time-history analysis
analysisand andthe regional
the seismic
regional lossloss
seismic prediction,
prediction,
a real-time city-scale time-history analysis method is proposed in this work. A program named “RED-
a real-time city-scale time-history analysis method is proposed in this work. A program named
Based was
ACT” on the city-scale
developed to nonlinear time-history
automatically implement analysis and theworkflow.
the described regional seismic loss prediction,
The reliability and
“RED-ACT” was developed to automatically implement the described workflow. The reliability and
a real-time city-scale
advantages of thetime-history analysis
proposed method in method
this workis were
proposed in this work.
demonstrated A program
through named “RED-
actual earthquake
advantages
events.
of thethe
proposed method in this work were demonstrated through actual earthquake events.
ACT” was Then
developed program was applied to
to automatically various earthquake
implement events. workflow.
the described The main conclusions are as and
The reliability
Thenfollows:
the program was applied to various earthquake events. The main conclusions are as follows:
advantages of the proposed method in this work were demonstrated through actual earthquake
events.
(1) The
(1) Then
The the program
uncertainty
uncertaintyproblemwasofapplied
problem groundto
ofground various
motion
motion earthquake
input
input isissolved events.
solvedproperly
properlyThe
with
with main
thetheconclusions
proposed
proposed are as
method
method
follows: based
based on on
thethe real-time
real-time groundmotion
ground motion obtained
obtained from
fromthe theseismic
seismicstations;
stations;
(2) (2)
TheThe amplitude,
amplitude, spectrum,
spectrum, and and duration
duration characteristics
characteristics of ground
of ground motions,
motions, as well as stiffness,
the
(1) The uncertainty problem of ground motion input is solved properly with as
thewell as the
proposed method
stiffness, strength, and deformation characteristics of different buildings are fully considered in
strength,
based on and deformation
the real-time characteristics
ground of different
motion obtained from buildings
the seismicare fully considered in this method,
stations;
this method, based on the nonlinear time-history analysis and MDOF models;
based
(2) (3)
The on the nonlinear
amplitude, spectrum,time-history analysis
andtime-history
duration and MDOF models;
characteristics
Using the real-time city-scale analysis and theofcorresponding
ground motions,report as well the
system, as the
(3) stiffness,
Using the
assessmentreal-time
strength,
of the city-scale
andearthquake’s time-history
deformationdestructive analysis
characteristics
power, and
of repair
different the corresponding
buildings
time, are fully
and economic report system,
considered
loss can be in
the obtained
this assessment
method, basedof on
shortly the earthquake’s
the
after nonlinear
an earthquake destructive
time-history
event, which power,
analysis repair
and
provides aMDOFtime,reference
useful and economic
models; loss can
for scientific
be obtained
(3) Using shortlyfor
the real-time
decision-making after an earthquake
city-scale
earthquake disaster event,
time-history which
analysis
relief. This and
workprovides
the
is asignificant
useful reference
corresponding
highly toreport for scientific
system,
enhancing the the
decision-making
resilienceof
assessment for earthquake
of earthquake-stricken disaster
areas.
the earthquake’s destructive relief.
power, repair time, and economic loss can the
This work is highly significant to enhancing be
resilience
obtained of earthquake-stricken
shortly after an areas.
earthquake event, which provides a useful reference
Author Contributions: Xinzheng Lu conceived the concept and guided the research, Qingle Cheng conducted for scientific
decision-making
the for earthquake
analytical work and disaster relief.
wrote the manuscript, ThisXu,work
and Zhen is highly
Yongjia Xu andsignificant
Chujin Sunto enhancing
reviewed the the
Author Contributions:
resilience
manuscript. X.L. conceived theareas.
of earthquake-stricken concept and guided the research, Q.C. conducted the analytical work
and wrote the manuscript, and Z.X., Y.X. and C.S. reviewed the manuscript.
Funding:
Author The authors
Contributions: are grateful
Xinzheng for the financial support received from
thethe National Key R&D Program
Funding: The authors are gratefulLu
forconceived the concept
the financial supportand guided
received from research, Qingle
the National Cheng
Key R&Dconducted
Program
(number
(number 2018YFC1504401).
2018YFC1504401).
the analytical work and wrote the manuscript, and Zhen Xu, Yongjia Xu and Chujin Sun reviewed the
manuscript.
Acknowledgments: The authors are grateful for the ground motion data provided by the China Earthquake
Networks Center, the Center for Engineering Strong Motion Data (CESMD), K-NET, and KiK-net.

Conflicts of Interest: : The authors declare no conflict of interest.

References
Appl. Sci. 2019, 9, 3497 15 of 16

Acknowledgments: The authors are grateful for the ground motion data provided by the China Earthquake
Networks Center, the Center for Engineering Strong Motion Data (CESMD), K-NET, and KiK-net.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Erdik, M.; Şeşetyan, K.; Demircioğlu, M.B.; Hancılar, U.; Zülfikar, C. Rapid earthquake loss assessment after
damaging earthquakes. Soil Dyn. Earthq. Eng. 2011, 31, 247–266. [CrossRef]
2. Wald, D.J.; Jaiswal, K.; Marano, K.D.; Bausch, D.B.; Hearne, M.G. PAGER—Rapid Assessment of an Earthquake’s
Impact; U.S. Geological Survey Fact Sheet 2010–3036; U.S. Geological Survey: Reston, VA, USA, 2010.
3. GDACS. Global Disaster Alert and Coordination System. Available online: http://www.gdacs.org (accessed
on 11 August 2019).
4. Trendafiloski, G.; Wyss, M.; Rosset, P. Loss estimation module in the second generation software QLARM.
In Human Casualties in Earthquakes; Springer: Dordrecht, The Netherlands, 2011; pp. 95–106.
5. Erdik, M.; Fahjan, Y. Damage scenarios and damage evaluation. In Assessing and Managing Earthquake Risk;
Springer: Dordrecht, The Netherlands, 2008; pp. 213–237.
6. Applied Technology Council (ATC). Earthquake Damage Evaluation Data for California; ATC-13; Applied
Technology Council: Redwood City, CA, USA, 1985.
7. Federal Emergency Management Agency (FEMA). Multi-hazard Loss Estimation Methodology-Earthquake Model
Technical Manual (HAZUS-MH 2.1); Final Report; Federal Emergency Management Agency: Washington, DC,
USA, 2012.
8. Morfidis, K.; Kostinakis, K. Approaches to the rapid seismic damage prediction of r/c buildings using artificial
neural networks. Eng. Struct. 2018, 165, 120–141. [CrossRef]
9. De Lautour, O.R.; Omenzetter, P. Prediction of seismic-induced structural damage using artificial neural
networks. Eng. Struct. 2009, 31, 600–606. [CrossRef]
10. Population Census Office under the State Council, Department of Population and Employment Statistics
of National Bureau of Statistics. Tabulation on the 2010 Population Census of the People’s Republic of China;
China Statistics Press: Beijing, China, 2012. (In Chinese)
11. Lu, X.Z.; Guan, H. Nonlinear MDOF Models for Earthquake Disaster Simulation of Urban Buildings.
In Earthquake Disaster Simulation of Civil Infrastructures: From Tall Buildings to Urban Areas; Springer: Singapore,
2017; pp. 257–301.
12. Xiong, C.; Lu, X.Z.; Guan, H.; Xu, Z. A nonlinear computational model for regional seismic simulation of tall
buildings. Bull. Earthq. Eng. 2016, 14, 1047–1069. [CrossRef]
13. Xiong, C.; Lu, X.Z.; Lin, X.C.; Xu, Z.; Ye, L.P. Parameter determination and damage assessment for THA-based
regional seismic damage prediction of multi-story buildings. J. Earthq. Eng. 2017, 21, 461–485. [CrossRef]
14. Hori, M. Introduction to Computational Earthquake Engineering; Imperial College Press: London, UK, 2006.
15. Lu, X.Z.; Han, B.; Hori, M.; Xiong, C.; Xu, Z. A coarse-grained parallel approach for seismic damage
simulations of urban areas based on refined models and GPU/CPU cooperative computing. Adv. Eng. Softw.
2014, 70, 90–103. [CrossRef]
16. Sobhaninejad, G.; Hori, M.; Kabeyasawa, T. Enhancing integrated earthquake simulation with high
performance computing. Adv. Eng. Softw. 2011, 42, 286–292. [CrossRef]
17. GB50009-2012. Load Code for Design of Building Structures (GB 50009-2012); Ministry of Housing and
Urban-Rural Development of the People’s Republic of China (MOHURD): Beijing, China, 2012.
18. American Society of Civil Engineers (ASCE). Minimum Design Loads for Buildings and Other Structures
(ASCE/SEI 7-10); American Society of Civil Engineers (ASCE): Reston, VA, USA, 2010.
19. GB50011-2010. Code for Seismic Design of Buildings (GB50011-2010); Ministry of Housing and Urban-Rural
Development of the People’s Republic of China (MOHURD): Beijing, China, 2010.
20. GB50010-2010. Code for Seismic Design of Concrete Structures (GB50010-2010); Ministry of Housing and
Urban-Rural Development of the People’s Republic of China (MOHURD): Beijing, China, 2010.
21. Steelman, J.S.; Hajjar, J.F. Influence of inelastic seismic response modeling on regional loss estimation.
Eng. Struct. 2009, 31, 2976–2987. [CrossRef]
Appl. Sci. 2019, 9, 3497 16 of 16

22. Lu, X.Z.; Tian, Y.; Guan, H.; Xiong, C. Parametric sensitivity study on regional seismic damage prediction
of reinforced masonry buildings based on time-history analysis. Bull. Earthq. Eng. 2017, 15, 4791–4820.
[CrossRef]
23. Simiu, E.; Scanlan, R.H. Wind Effects on Structures: Fundamentals and Applications to Design, 3rd ed.; John Wiley:
New York, NY, USA, 1996; p. 512.
24. GB/T18208. Post-Earthquake Field Works—Part 4: Assessment of Direct Loss (GB/T 18208.4-2011); Standardization
Administration of China: Beijing, China, 2011. (In Chinese)
25. Zeng, X.; Lu, X.Z.; Yang, T.Y.; Xu, Z. Application of the FEMA-P58 methodology for regional earthquake loss
prediction. Nat. Hazards 2016, 83, 177–192. [CrossRef]
26. Xu, Z.; Lu, X.Z.; Zeng, X.; Xu, Y.J.; Li, Y. Seismic loss assessment for buildings with various-LOD BIM data.
Adv. Eng. Inf. 2019, 39, 112–126. [CrossRef]
27. Zeng, X. Direct Seismic Loss Assessment and Fire Following Earthquake Simulation of Urban Buildings.
Ph.D. Thesis, Tsinghua University, Beijing, China, 2018. (In Chinese).
28. Federal Emergency Management Agency (FEMA). Seismic Performance Assessment of Buildings; Methodology;
Technical Report FEMA-P58; Federal Emergency Management Agency (FEMA): Washington, DC, USA, 2012;
Volume 1.
29. Xu, X.; Ding, L.Y.; Luo, H.B.; Ma, L. From building information modeling to city information modeling. J. Inf.
Technol. Constr. 2014, 19, 292–307.
30. Chandra, R.; Dagum, L.; Kohr, D.; Menon, R.; Maydan, D.; McDonald, J. Parallel Programming in OpenMP;
Morgan Kaufmann: San Diego, CA, USA„ 2001; pp. 8–14.
31. Lu, X.Z.; Gu, D.L.; Lin, X.C.; Cheng, Q.L.; Zhang, L.; Tian, Y.; Zeng, X. Seismic damage assessment of the
ground motion near the epicenter of the 7.0 earthquake in Jiuzhaigou, Sichuan. Stand. Eng. Constr. 2017,
68–73. (In Chinese)
32. Dai, J.W.; Sun, B.T.; Li, S.Y.; Tao, Z.R.; Ma, Q.; Zhang, L.X.; Lin, J.Q. Engineering Damage in Jiuzhaigou M 7.0
Earthquake; Seismological Press: Beijing, China, 2018; p. 18. (In Chinese)
33. Lu, X.Z. Damage Capacity of the Ground Motions of the Dec. 1, Alaska Earthquake; Research Group Report;
Tsinghua University: Beijing, China, 2018.
34. Structural Extreme Event Reconnaissance Network (StEER), Earthquake Engineering Research Institute
(EERI). Alaska Earthquake Preliminary Virtual Assessment Team (P-VAT) Joint Report; Earthquake Engineering
Research Institute: Oakland, CA, USA, 2018.
35. Center for Engineering Strong Motion Data (CESMD). Available online: https://www.strongmotioncenter.
org/cgi-bin/CESMD/iqrStationMap.pl?ID=us1000hyfh (accessed on 26 January 2019).

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).