SEISMIC PERFORMANCE BASED EVALUATION OF

NONSTRUCTURAL COMPONENTS IN TALL BUILDINGS

Nwe Ni Sein Toe1, Naveed Anwar2, Thaung Htut Aung3 and Fawad Ahmed Najam4

ABSTRACT

With the development of seismic design of building, the performance of structural components and
systems have been focused for load resisting systems, however, nonstructural components are mostly
overlooked in structural design analysis. During earthquakes, when the harmonization does not occur
between structural and nonstructural parts, nonstructural components which are not considered under
seismic resisting requirements are easily more fragile and damaged than structural components. Thus,
seismic performance of nonstructural components in buildings is becoming important and evaluated
well among engineers and researchers. The issue is viewed as main concern of seismic performance
based evaluation since the earthquake happened in San Fernando in 1971. The objective of this study
is to assess performance of nonstructural components within tall building locating in Bangkok and
Manila respectively under Service Level Earthquake (SLE), Design Based Earthquake (DBE) and
Maximum Considered Earthquake (MCE). The seismic performance of nonstructural components can
be measured by component damage and this damageability can be translated to losses associating with
direct economic losses regarding with direct economic cost, direct social loss in term of serious
injuries and casualties, and indirect losses relating with downtime and business interruption loss. Thus,
seismic performance based evaluation of nonstructural components under various earthquake levels
can produce a well preparedness for nonstructural building components and contents to cope with
damage and losses caused by upcoming earthquakes. This study provides that losses resulting from
nonstructural damage are higher due to maximum considered earthquake (MCE) than other earthquake
levels.

Keywords: Performance based evaluation1; nonstructural components2; SLE3; DBE4; MCE4.

1. INTRODUCTION

Earthquakes naturally cause damage to structures and losses to environment such as economic losses
and loss of lives in many countries. A recent seismic hazard assessment provides that Bangkok closed
to active fault zones and locating on soft soil layer can suffer damage and losses due to long distance
earthquakes. In Manila, the west valley fault transverses the city and poses a life threating to the city
and damaging to structures existing near the fault line. The past earthquake in San Fernando in 1971
presented that nonstructural components experienced damage resulting in economic losses and
casualties. The role of nonstructural components was overlooked in previous generation but in recent
time, it has been considered as an issue of damage and losses for high rise buildings. In the next
generation of performance based earthquake engineering, seismic performance of building
components is measured in term of losses such as direct economic loss, direct social loss, downtime
and business interruption loss.
Seismic performance assessment of building components can be done by various methods to estimate
earthquake losses. Since 1973, Czarnecki created the first approach of investigating the damage
analysis and loss analysis and the relationships of component vulnerabilities. The PEER’s Loss
Estimation methodology provided by Keith [Link] (2013) is to estimate repair cost for damage
components, loss of lives due to damage and required time to repair damaged components. The FEMA
P-58 seismic performance assessment framework assess seismic performance of individual building
component. The performance is measured by component damage and earthquake losses associated to
human loss, repair cost and repair time. Artudorn, K., (2016) defined the component-based loss
estimation method to consider damage and losses of each building component type in term of direct
economic loss, direct social loss regarding serious injury or casualty and downtime. Almufti and
Willford (2013) provided the Resilience-based Earthquake Design Initiative (REDi™) Rating System
to achieve downtime for building component repairs into specific recovery state. This study provides
that losses resulting from nonstructural damage are higher due to maximum considered earthquake
(MCE) than other earthquake levels.

2. PERFORMANCE ASSESSMENT METHODOLOGY

This study predicts earthquake losses of non-structural components in 37-story building in Bangkok
and 50-story building in Manila. There are five steps to calculate earthquake losses. Then performance
of nonstructural components is calculated in term of losses when the case study buildings are
subjected to earthquakes.

2.1 Data collection

At first, the methodology is to collect building data of case study buildings in Bangkok and Manila.
This data includes drawing plans: architectural drawings, interior drawings and structural drawings,
building configuration, locations of non-structural components within buildings, replacement cost of
each component type and repair time of individual component repair and replacement and labor
availability and rental cost for occupancy. Collection of these database is able to estimate economic
loss, casualties and downtime.

2.2 Selection of earthquake scenario

The second step is to select earthquake scenarios of case study buildings. In 37-story building in
Bangkok, existing in moderate seismic zone, appropriate three pairs of ground motion are used to
check the performance of the building against the service level earthquake, design-based earthquake
level and maximum considered earthquake level with probability of 50% exceedance in 30 years,
probability of 10% exceedance in 50 years and 2% probability of exceedance in 50 years respectively.
For 50-story building, in Manila locating in high seismic zone, the structural system is designed by
using seven pairs of ground motion. These ground motions are based on service level earthquake
scenario, design-based earthquake scenario and maximum considered earthquake scenario.

2.3 Review of structural analysis

The third step is to review the structural analysis and building responses due to earthquakes. The
structural analysis program is not shown in detail and this program is done by structural engineers. For
37-story building in Bangkok, the structural performance is investigated by using the PERFORM 3D
program. The model includes inelastic properties for elements and components that are anticipated to
be loaded beyond their elastic limits. When the case study building is attacked by input ground
motions, the structural responses are evaluated as drift, acceleration and velocity. Furthermore, for 50-
story building in Manila, the structural performance is checked by using the nonlinear time history
analysis. The building responses such as drift, acceleration and velocity, are checked by subjecting
nonlinear model into seven ground motions of maximum considered earthquake level. In order to
evaluate the seismic performance of 50-story building under service level earthquake and design-based
earthquake, the response spectrum analysis will be used.

2.4 Damage assessment

Nonstructural damage calculation is a primary part of performance assessment of nonstructural

components. The damage states of each component type can be estimated by assigning a single
demand parameter into fragility function provided by FEMA P-58 fragility specification. Additionally,
in each damage state, the number of damaged component can be estimated by multiplying the number
of that component in a performance group and the probability of each damage state. The number of
damaged component in damage state can be computed as follow.
2
 = x
(2.1)

Where,
= the number of damaged component in damage state
= the number of damaged component in a performance group
= the probability of damaged component being in damage state

2.5 performance assessment

The last step is to calculate seismic performance of nonstructural components by using FEMA P-58
consequence database. Then nonstructural component damage is translated to losses such as direct
economic loss, direct social loss, downtime and business interruption loss.

2.5.1 Direct economic loss

The direct economic cost of individual nonstructural component is computed by multiplying the
number of component being in each damage state, repair cost ratio and replacement cost of each
component type. Additionally, cost ratio can be defined as ratio of repair cost in a considered damage
state to replacement cost in maximum damage state. Replacement cost is the cost to repair or replace
damage for nonstructural component damage. In this study, replacement cost for each component is
obtained from construction contractors or estimators who are familiar with the repair work from
Bangkok and Manila.

2.5.2 Direct social loss

This study estimates the number of serious injury and casualty by using FEMA P-58 consequence
functions. In this study, the residential population is estimated by distributing number of occupant on
each floor area and considering when earthquakes hit the buildings at times of peak population in
daytime and nighttime. The population models follow default population distribution provided by
FEMA and HAZUS.

2.5.3 Downtime

In this study, downtime is computed by using five methods. Firstly, the Resilience-based Earthquake
Design Initiative (REDi™, 2013) Rating System is used as the REDi™ guideline-1 to estimate
downtime for case study buildings. The REDi™ provides that estimated downtime depends on not
only repair time of nonstructural damage but also delay time due to utility disruption and impeding
factors. Under this guideline, each type of nonstructural repair is in sequence in each floor along the
buildings and recommended number of worker is allocated on each floor. Furthermore, other
downtime methods are modified by the REDi™ guideline-1 and contractors’ recommendations. In
addition to that, repair sequences, required number of labor for replacement of each component type
and working hour are adjusted by contractors.

2.5.4 Business interruption loss

On nonstructural repair duration, the buildings are inoperable and unserviceable for occupants to run
business. This impact leads to the business interruption for residents so business interruption loss is
estimated since downtime is calculated. This study computes the business interruption cost by
multiplying downtime with rental cost for occupancy and breakdown area of case study buildings.

3. RESULTS AND DISCUSSION

This study aims to observe nonstructural damage, direct economic loss, direct social loss, downtime
3
and business interruption loss by applying the component-based loss estimation method when the
SLE, DBE and MCE subject to 37-story building in Bangkok regarded as moderate seismic zone and
50-story building in Manila regarded as high seismic zone. The outcomes will be discussed in five
sections.

3.1 Nonstructural damage analysis

When the case study buildings in Bangkok and Manila are subjected to SLE, DBE and MCE,
nonstructural damage is calculated. This study compares damage ratio from each type of architectural
components by using actual quantities of components from architectural drawings and default data
from FEMA P-58 normative quantity estimation. Figure 1 and figure 2 show the comparison of actual
damage ratio and default damage ratio of each architectural component in 37-story building in
Bangkok and 50-story building in Manila due to MCE.

Actual damage ratio due to maximum considered earthquake

quantity in each damage

100%

100%

99%
percentage of total

69%
150%
64%

61%
60%

54%
46%
35%
26%

20%
17%

100%
13%

11%
8%

6%

6%

4%

1%
0%

0%

0%
0%
0%

0%
0%
0%

0%
0%

0%
0%
state

50%
0%
Interior Interior wall wall glazing-X glazing-Y Ceilings Stair
wall-X wall-Y finishing- X finishing- Y

None damage DS1 DS2 DS3

Default damage ratio due to maximum considered earthquake

quantity in eacg damage

100%

99%
100%
69%
percentage of total

150%
61%
59%

57%
54%

43%
35%
24%

22%

20%

100%
16%

13%

11%
6%

6%

4%

1%
0%

0%

0%
0%
0%

0%
0%
0%

0%
0%

0%
0%
50%
state

0%
Interior Interior wall wall glazing-X glazing-Y Ceilings Stair
wall-X wall-Y finishing- X finishing- Y

None damage DS1 DS2 DS3

Figure 1. The comparison of damage ratios from actual quantity and default quantity of each
architectural component within 37-story building in Bangkok

Actual damage ratio in each damage state due to maximum considered earthquake
quantity in each damage

99%
98%
92%
percentage of total

84%

150%
55%

45%
43%

43%
35%
32%

30%
27%

25%
19%

17%

100%
16%

13%
11%

7%
3%
2%

1%
1%

1%
0%

0%

0%

0%

0%

0%
0%

0%
state

50%
0%
Interior Interior wall wall glazing-X glazing-Y Ceilings Stair
wall-X wall-Y finishing- X finishing- Y

None damage DS1 DS2 DS3

Default damage ratio in each damage state due to maximum considered earthquake
quantity in each damage

99%
98%
89%
percentage of total

79%

150%
51%

45%
43%
38%
33%

30%
29%

27%

25%
22%

20%

17%

100%
13%

13%
11%
7%

7%

1%

1%
1%

1%
0%

0%

0%

0%

0%
0%

0%

50%
state

0%
Interior Interior wall wall glazing-X glazing-Y Ceilings Stair
wall-X wall-Y finishing- X finishing- Y

None damage DS1 DS2 DS3

Figure 2. The comparison of damage ratios from actual quantity and default quantity of each
architectural component within 50-story building in Manila

4
It can be seen that actual damage resulting from architectural components are higher than default
damage in considered earthquake. For example, actual quantities of interior wall are more damaged
than default quantities because actual damage ratio of interior wall corresponding to none damage
state is higher than default damage ratio of that component in same damage state.
Due to earthquakes, nonstructural components are damaged by drift, acceleration and velocity and
then damage of these damaged components are evaluated in the study. Figure 3 and figure 4 describe
the nonstructural damage on each floor of 37-story building in Bangkok and 50-story building in
Manila subjected to MCE.

Drift-sensitive component Acceleration Sensitive Component Velocity-sensitive component

damage under MCE Damage under MCE damage under MCE
40 40 40
36 36 36
32 32 32
28 28 28
NO: of story

No; of story
24 24 24
20 20
NO: of story

20
16 16 16
12 12 12
8 8 8
4 4 4
0 0
0 -4
-4 -4
0% 5% 10% 15% 20% 0% 50% 100%
0% 50% 100%
percentage of total quantity on each percentage of total quantity on
percentage of total quantity on
floor each floor
each floor
Architectural damage due to drift- Architectural damage due to acceleration-
SRSS-X+30%Y content damage due to velocity-
SRSS-XY
SRSS-XY
Architectural damage due to drift- Mechanical and electrical damage due to
SRSS-Y+30%X acceleration-SRSS-XY
Architectural damage due to drift- Content damage due to acceleration-SRSS-
SRSS-XY XY

Figure 3. Nonstructural damage of 37-story building in Bangkok due to MCE

Drift-sensitive component Acceleration-sensitive component Velocity-sensitive component

damage under MCE damage under MCE damage under MCE
50 50
50 45 45
45 40 40
40 35 35
No; of story

No: of story

35 30 30
NO: of story

30 25 25
25 20 20
20 15 15
15 10 10
10 5 5
5 0 0
0 -5 -5
-5 0% 50% 100% 150% 0% 50% 100% 150%
0% 50% 100% 150%
percentage of total quantity on each
percentage of total quantity on floor percentage of total component on
each floor each floor
Architectural component damage due to
Architectural damage due to drift- acceleration-SRSS-XY Content damage due to velocity-
SRSS-X+30%Y SRSS-XY
Mechanical and electrical component
Architectural damage due to drift-
damage due to acceleration-XY
SRSS-Y+30%X
Architectural damage due to drift- Content damage due to acceleration-
SRSS-XY SRSS-XY

Figure 4. Nonstructural damage of 50-story building in Manila due to MCE

It is observed that due to MCE, in Bangkok, moderate seismic zones, drift-sensitive component
damage is high in many floor levels but acceleration sensitive component damage are not significant.
On the other hand, damage results resulting from acceleration sensitive components and velocity-
sensitive components are significant in many floor levels of 50-story building in Manila locating in
high seismic zone after subjecting to MCE. Moreover, overall damage of each nonstructural
component type at three earthquake levels is calculated for 37-story building in Bangkok locating in
5
moderate seismic zone and 50-story building in Manila locating in high seismic zone.
Overall nonstructural damage

Percentage of total
10.0% 7.6%
5.8%

component
4.7%
5.0% 3.1% 2.7%4.0% 2.8%
0% 0% 0.1% 0.4%1.8%
0.0%
Total Damage Architectural damage Mechanical and Content damage
Electrical damage

Service level earthquake Design based earthquake Maximum considered earthquake

Overall nonstructural damage in term of sensitivity

Percentage of total
component

10.0% 7.6%
5.8% 4.7%
5.0% 3.1% 2.7% 4% 1.6%
0% 0.8% 0.4%1.0%1.3%
0.0%
Total Damage Drift Sensitivity Acceleration Velocity Sensitivity
Sensitivity

Service level earthquake Design based earthquake Maximum considered earthquake

Figure 5. Overall nonstructural damage of 37-story building in Bangkok

According to figure 5, damage from architectural components is the most influence when 37-story
building in Bangkok is experienced in three earthquake levels such as SLE, DBE and MCE.
Additionally, most of architectural components are sensitive in drift so drift sensitive component
damage is also significant due to earthquakes.

Overall nonstructural damage

Percentage of total

36%
40.0%
component

30.0% 25% 23%

20.0% 15%
6.3% 9% 11%
10.0% 3.5% 2.8%
0% 1% 2%
0.0%
Total Damage Architectural damage Mechanical and Content damage
Electrical damage
Service level earthquake Design based earthquake Maximum considered earthquake

Overall nonstructural damage in term of sensitivity

Percentage of total

40.0% 36%
25% 24%
component

30.0%
20.0% 15%
6.3% 8% 9%
10.0% 3.4% 2.6% 0.3% 2% 3%
0.0%
Total Damage Drift Sensitivity Acceleration Velocity Sensitivity
Sensitivity
Service level earthquake Design based earthquake Maximum considered earthquake
Figure 6. Overall nonstructural damage of 50-story building in Manila

In figure 6, it is found that major nonstructural damage is caused by high acceleration demand when
50-story building is under maximum considered earthquake level and design based earthquake level.
Overall, the highest percentage of nonstructural damage occurs when case study buildings in moderate
and high seismic zone respectively are subjected to maximum considered earthquake.

3.2 Direct economic loss

Direct economic cost is estimated to repair nonstructural damage of the case study buildings after
subjecting to earthquakes.

6
 Table 1 Direct economic loss of nonstructural components for 37-story building in Bangkok

Repair Cost (Million-baht)

Architectural Mechanical and
Earthquake Total repair Building contents
components electrical components
SLE 5 4 0 1
DBE 14.4 5.9 0.1 8.4
MCE 23.8 7.4 0.4 16.1
Drift-sensitive Acceleration-sensitive Velocity-sensitive
Earthquake Total repair
components components components
SLE 5 4 0 1
DBE 14.4 5.9 6.6 1.9
MCE 23.8 7.3 14 2.5

The table 1 compares direct economic loss resulting from nonstructural damage of 37-story building
due to SLE, DBE and MCE. The repair cost of drift-sensitive components is the largest amount of cost
under service level earthquake. However, due to maximum considered earthquake, the repair cost from
acceleration sensitive components is governed. Most of acceleration sensitive components are from
building contents.

Table 2 Direct economic loss of nonstructural components for 50-story building in Manila

Repair cost (million-peso)

Architectural Mechanical and
Earthquake Total repair Building contents
components electrical components
SLE 16.9 4.7 0.3 11.9
DBE 122 51 7 64
MCE 217.5 103.4 17.9 96.2
Drift-sensitive Acceleration- Velocity-sensitive
Earthquake Total repair
components sensitive components components
SLE 16.9 2.7 13.6 0.6
DBE 122 7 110 5
MCE 217.5 8.9 202.7 5.9

The table 2 shows direct economic loss resulting from nonstructural damage of 50-story building due
to SLE, DBE and MCE. It is found that direct economic cost is governed by acceleration sensitive
components.

3.3 Direct social loss

The damage of nonstructural components cause serious injury to occupants because in non-collapse
case, total number of population in casualty-affected area will not become casualties. Thus, direct
social loss is related to the number of serious injury and casualties from nonstructural damage.

Table 3. Direct social loss due to actual damage of architectural components for 37-story building in
moderate seismic zone

FEMA-Daytime HAZUS-Daytime Nighttime

Earthquake Injury Casualty Injury Casualty Injury Casualty
(people) (people) (people) (people) (people) (people)
SLE 0 0 0 0 0 0
DBE 1 0 1 0 1 0
7
 MCE 1 0 1 0 1 0

Table 4. Direct social loss due to default damage of architectural components for 37-story building in
moderate seismic zone

FEMA-Daytime HAZUS-Daytime Nighttime

Earthquake Injury Casualty Injury Casualty Injury Casualty
(people) (people) (people) (people) (people) (people)
SLE 0 0 0 0 0 0
DBE 1 0 1 0 1 0
MCE 1 0 1 0 1 0

Table 3 describes direct social loss resulted from actual damage of architectural components and table
4 represents direct social loss resulted from default damage of architectural components for 37-story
building in moderate seismic zone due to earthquakes. As the result, there is no difference between
social loss due to default architectural damage and actual architectural damage because these actual
and default damage of injury affected components are likely to be same under earthquakes.

Table 5. Direct social loss due to actual damage of architectural components for 50-story building in
high seismic zone

FEMA-Daytime HAZUS-Daytime Nighttime

Earthquake Injury Casualty Injury Casualty Injury Casualty
(people) (people) (people) (people) (people) (people)
SLE 1 0 1 0 1 0
DBE 2 0 1 0 3 0
MCE 7 0 4 0 13 0

Table 6. Direct social loss due to actual damage of architectural components for 50-story building in
high seismic zone

FEMA-Daytime HAZUS-Daytime Nighttime

Earthquake Injury Casualty Injury Casualty Injury Casualty
(people) (people) (people) (people) (people) (people)
SLE 1 0 1 0 1 0
DBE 2 0 1 0 3 0
MCE 7 0 4 0 13 0

Table 5 describes direct social loss resulted from actual damage of architectural components and table
6 represents direct social loss resulted from default damage of architectural components for 50-story
building in high seismic zone. As the result, there is no difference between social loss due to default
architectural damage and actual architectural damage because these actual and default damage of
injury affected components are likely to be same under earthquakes. The number of serious injury at
peak population time, nighttime is higher than other time distribution population models.

Table 7. Direct social loss due to nonstructural damage for 37-story building in moderate seismic zone

FEMA-Daytime HAZUS-Daytime Nighttime

Earthquake Injury Casualty Injury Casualty Injury Casualty
(people) (people) (people) (people) (people) (people)
SLE 0 0 0 0 0 0
DBE 1 0 1 0 1 0
MCE 1 0 1 0 1 0
8
 Table 8. Direct social loss due to nonstructural damage for 50-story building in high seismic zone

FEMA-Daytime HAZUS-Daytime Nighttime

Earthquake Injury Casualty Injury Casualty Injury Casualty
(people) (people) (people) (people) (people) (people)
SLE 1 0 1 0 2 0
DBE 5 1 3 1 9 1
MCE 11 1 7 1 22 1

Table 7 represents the number of serious injury and casualty after 37-story building in Bangkok is
attacked by earthquakes. When the building is under maximum considered earthquake, direct social
loss is still low, representing with 1 number of serious injury and no loss of lives for each considered
population model. In moderate seismic zone, nonstructural damage is governed by drift-sensitive
components. In addition to that, percentage of serious injury, percentage of casualty and casualty-
affected planar area which are provided by FEMA-consequence database for drift-sensitive
components are low. Table 8 shows the direct social loss of 50-story building in Manila due to
different earthquakes. Direct social loss is high in 50-story building in Manila. In high seismic zone,
major damage is caused by acceleration sensitive components. Additionally, according to FEMA-
consequence database, percentage of serious injury rate and percentage of casualty rate in casualty
affected area for acceleration sensitive component damage is higher than consequence database for
drift-sensitive component damage. Furthermore, when earthquakes come at peak occupancy time such
as nighttime, the number of injury is estimated as the largest values of human loss.

3.4 Downtime

This repair time assessment is estimated for 37-story building in Bangkok and 50-story building in
Manila. Downtime assessment is required to recover nonstructural damage into full recovery
conditions.

Table 5. Downtime for 37-story building in Bangkok (left), downtime for 50-story building in Manila
(right)

Table-5 describes estimated downtime for 37-story building in Bangkok and 50-story building in
Manila by using five methods. These methods are developed by the REDi™ guideline and contractors’
recommendations. It is found that downtime by using the REDi™ guideline-4 is the shortest downtime
among other downtime so this repair method can decrease high business interruption loss due to long
downtime. Figure 7, figure 8 and figure 9, describe the examples of repair schedule by using five

9
methods for 50-story building in Manila under MCE.

Figure 7. Example of downtime by the REDi™ guideline-1-when 50-story building in Manila is

subjected to MCE

Figure 8. Example of downtime by the REDi™ guideline-2 (Left) and example of downtime by the
REDi™ guideline-3 (Right) when 50-story building in Manila is subjected to MCE

Figure 9. Example of downtime by the REDi™ guideline-4 (Left) and example of downtime by the
REDi™ guideline-5 (Right) when 50-story building in Manila is subjected to MCE

It is found that downtime by using the Resilience-based Earthquake Design Initiative Rating System
(REDi™ guideline-1) is the longest downtime since nonstructural damage are repaired in sequences in
10
each floor and REDi™ recommended number of worker is allocated in each floor. Other downtime
methods are developed by the REDi™ guideline-1 and contractors’ recommendations. The REDi™
guideline-2 reduces long downtime because it uses maximum number of worker provided by the
REDi™ to repair nonstructural damage simultaneously within one repair floor set. Due to worker
unavailability after earthquake, the REDi™ guideline-3 reduces half of maximum number of worker
recommended by the REDi™. Then this downtime leads to longer downtime than previous method.
The downtime by the REDi™ guideline-4 is the shortest downtime because it uses recommended
maximum number workers and adds over time work. The REDi™ guideline-5 reduces downtime
because it repairs nonstructural damage on each floor simultaneously in a same repair set and reduce
half of maximum number of worker recommended by REDi™ guideline and adds extra working hours.
It can be concluded that downtime estimated by the REDi™ guideline-4 is the shortest downtime to
achieve nonstructural damage of case study buildings into recovery state although this guideline has
uncertain factors such as number of worker and over time work which are recommended by
contractors.

3.5 Business interruption loss

When the case study buildings are under repair duration, these buildings are not operable and
functional. This repair duration leads to interrupt the running business. Therefore, the business
interruption cost is calculated after estimating downtime of nonstructural repairs.

Table 6. Business interruption loss due to downtime for 37-story building in Bangkok

Business Interruption Loss SLE DBE MCE

Business interruption loss by using REDi™ guideline-1
Business interruption cost (million-baht) 102 130 155
Business interruption loss by using REDi™ guideline-2
Business interruption cost (million-baht) 49 51 57
Business interruption loss by using REDi™ guideline-3
Business interruption cost (million-baht) 55 62 93
Business interruption loss by using REDi™ guideline-4
Business interruption cost (million-baht) 45 49 53
Business interruption loss by using REDi™ guideline-5
Business interruption cost (million-baht) 49 57 76

Table 7. Business interruption loss due to downtime for 50-story building in Manila

Business Interruption Loss SLE DBE MCE

Business interruption loss by using REDi™ guideline-1
Business interruption cost (million-peso) 277 1,087 2,124
Business interruption loss by using REDi™ guideline-2
Business interruption cost (million-peso) 119 212 346
Business interruption loss by using REDi™ guideline-3
Business interruption cost (million-peso) 133 346 642
Business interruption loss by using REDi™ guideline-4
Business interruption cost (million-peso) 119 178 247
Business interruption loss by using REDi™ guideline-5
Business interruption cost (million-peso) 128 247 445

Table 6 and table 7 present the outcomes of business interruption loss due to downtime for 37-story
building in Bangkok and 50-story building in Manila due to different earthquake levels. According to
these tables, for case study buildings, business interruption cost by using the REDi™ guideline-4 is
lower than cost from other downtimes because downtime under this guideline is the shortest. It is
observed that business interruption loss is significantly higher than direct economic loss that cost to
11
repair or replace nonstructural damage because business interruption loss depends on repair time for
nonstructural damage and time for delays which interrupt buildings’ operability and availability to run
business.

4. CONCLUSIONS
It is concluded that damage of drift sensitive components is major damage in moderate seismic zone
and in high seismic zone, significant damage is governed by acceleration sensitive components. The
rate of serious injury due to nonstructural damage is low with 0 to 1 person in 37-story building in
Bangkok but high with 1 to 22 people in 50-story building in Manila. In addition, the rate of casualty
is rather low which is from 0 to 1 person for each considered buildings. Direct economic cost from
nonstructural damage ranges from 1% to 3% of its replacement cost in 37-story building in Bangkok
and it ranges from 2% to 24% of nonstructural component replacement cost in 50-story building in
Manila. Downtime for 37-story building in Bangkok is from 2.6 to 4 months and for 50-story building
in Manila is from 2.6 to 9 months. Business interruption loss is significantly higher than direct
economic loss due to nonstructural damage for each case study building.

5. REFERENCES

Almufti, I., Willford, M. (2013). Resilience-based Earthquake Design Initiative (REDi™) for the Next
Generation of Buildings. Arup, San Francisco, CA.

Artudorn. K,. (2016), Estimating Losses and Downtime in High-Rise Buildings in Bangkok due to Long-
Distance Earthquakes Using a Component-Based Methodology. MEng Thesis, Asian Institute of Technology,
directed by Prof. Warnitchai, P,.

A. S. Whittaker and T. T. Soong, (2003). An Overview of Nonstructural Components Research at Three US

Earthquake Engineering Research Centers. Prepared by Applied Technology Council.

Farokhnia, K,. (2013). Nonstructural Vulnerability Functions for Building Categories. Ph.D. Thesis, University
of Colorado, directed by Prof. Porter. K. A,.

FEMA P-58, (2012). Seismic Performance Assessment of Buildings Methodology, prepared by Applied
Technology Council for the Federal Emergency Management Agency, report No. P-58-1, Washington, D.C.

FEMA P-58, (2012). Seismic Performance Assessment of Buildings Implementation Guide, prepared by Applied
Technology Council for the Federal Emergency Management Agency, report No. P-58-2, Washington, D.C.

HAZUS-MH-MR3, (2003). Multi-hazard Loss Estimation Methodology developed by Department of Homeland

Security Emergency Preparedness and Response Directorate for the Federal Emergency Management Agency
Mitigation Division, Washington, D.C.

HAZUS-MH-MR4, (2003). Multi-hazard Loss Estimation Methodology developed by Department of Homeland

Security Emergency Preparedness and Response Directorate for the Federal Emergency Management Agency
Mitigation Division, Washington, D.C.

Porter, K.A., (2003) An Overview of PEER’s Performance-Based Earthquake Engineering Methodology.

Proceedings of Ninth International Conference on Applications of Statistics and Probability in Civil
Engineering. San Francisco, California.

Saeki, T., Tsubokawa, H., Midorikawa, S., (1968). Seismic Damage Evaluation of Household Property by Using
GIS, Geographic Information System proceedings of 12 th World Conference on Earthquake Engineering, WCEE
(2000).

Taghavi, S., Miranda, E., (2003). Response Assessment of Nonstructural Building Elements. PEER Report
2003/05., Berkeley.

Whittaker, A. S., Huang, Y.N., Hamburger, R.O., (2007). Building-Specific Seismic Loss Assessment for Next-
generation performance based earthquake engineering.

12