Damage and loss probability assessment of reinforced concrete building due to Yogyakarta earthquake scenario using pushover and hazus analysis (case study: student center building, faculty of social science, UNY)

Yogyakarta is one of the cities in Indonesia prone to earthquake. The Student Center Building, Faculty of Social Science (FSS), Universitas Negeri Yogyakarta (UNY) was built in 2012 using the earthquake code of 2002. Seismic vulnerability assessment of the building due to the newest seismic code is essential to do as one of earthquake disaster mitigation efforts. The study aimed to determine the seismic building performance level based on ATC-40 criteria, damage probability matrix using HAZUS method, and damage loss economic value of the building based on FEMA-1999. The research begins by modelling building structures in 3D followed by performing pushover analysis then calculate the probability of building damage based on HAZUS method. The results show that the building is included in the Immediate Occupancy (IO) performance level which means that if an earthquake happens, only a little structural damage occurs. The total vulnerability value based on the HAZUS method analysis reached 48.12%, in which the probability of damage for the slight, moderate, extensive and complete level reached 22.59%, 21.60%, 3.71% and 0.23% respectively. In addition, the damage economic value obtained is 4.692% means that the building has a small damage economic probability due to the 2006 Yogyakarta earthquake scenario.


Introduction
Yogyakarta as one of the largest educational destination cities in Indonesia encourages educational institutions to develop and build some new educational facilities buildings from elementary schools to universities to support teaching and learning activities. New buildings, especially those located in earthquake-prone areas such as Yogyakarta must be designed with the latest earthquake resistance design code. The Student Center Building, Faculty of Social Science (FSS), Universitas Negeri Yogyakarta (UNY) completed in 2014 was designed using the seismic design code of 2002 (SNI 1726(SNI -2002. The earthquake hazard map of Indonesia in 2010 used in [1] shows an increase of peak ground acceleration (PGA) value from [2] earthquake map for Yogyakarta area (from 0.15g to range of 0.2-0.25g). Damage and loss probability study of the building due to the earthquake is essential as one of the disaster mitigation efforts.
Reference [3] states that the safety of building depends not only on the level of strength but also on the level of deformation and measurable energy on the performance of the structure. Pushover analysis method is an appropriate method to conduct the evaluation of building performance. The previous research related to pushover analysis was done in [4] and [5], but the results do not explain the building damage probability in each level of damage. While research on Damage and Loss Assessment has been done by [6] using the form of RC Frame Buildings. They used the different method to calculate the building damage and loss using FEMA P-58 which was written in [6].
This study is a combination and development of previous research [7], [8], and [9] using pushover analysis with performance level determination based on ATC-40 followed by building damage analysis using HAZUS method to get the probability value of building damage, then measure the damage loss estimation using FEMA-1999. The study aimed to obtain the performance level, damage probability matrix, and damage loss economic value of the building.

Research method
The method used in this research is pushover analysis or commonly known as nonlinear static analysis to determine the building performance level, followed by damage probability analysis using HAZUS method and calculation of damage loss based on FEMA-1999. The initial step of this study is modeling the structure in 3-dimensional structure, performing load, pushover, and damage probability analysis, followed by calculating the damage losses due to earthquake using Yogyakarta earthquake scenario year 2006. The detailed description of research stages is presented in subchapters below.

Data and structural modeling
The building under study is a 3-story reinforced concrete building that functions as the center of student activities located in Yogyakarta. The typical floor plan of the building is shown in Fig. 1. The quality of the material used is as follows: a. Quality of concrete: K300 (eq.to 25 MPa) b. Quality of steel: -deformed bar with the diameter of >10 mm, fy value of 320 MPa.
-plain round bar with the diameter of <10mm, fy value of 240 MPa. The structure is modelled in the 3-dimensional open frame structure according to As-Built Drawing data as shown in Fig. 2. The structural data based on the document of structure, as follows: a. Plates: 120 mm (floors, stairs, and borders); 100 mm (rooftop) b. Beams and Columns: each type of beam and column size is presented in Table 1.

Pushover analysis
Pushover analysis is a nonlinear static analysis in which the design earthquake load effect on the building structure is considered as static loads that are given at the centre of mass in each storey, which value is gradually increased to exceed the loading that causes the first plastic hinge in the structure of the building, then with a gradual increase of load resulted in the building undergoing major post-elastic form changes until it reached the plastic condition [14]. Capacity Spectrum Method (CSM) is the method which is most commonly used to compare the capacity and demand on pushover analysis. The CSM is basically focused on the reduction of elastic spectrum intersecting the capacity curve at the spectrum coordinate to obtain performance point.
Pushover analysis obtains a capacity curve that illustrates the relationship between base shear (V) and the roof displacement ( Δ ). The capacity curve is then converted into ADRs (Acceleration Displacement Response Spectrum) into a capacity spectrum curve that describes the relationship between Spectral Acceleration (Sa) and Spectral Displacement (Sd) which is shown in Fig. 3. while the spectrum response in the ADRS format is shown in Fig. 4.

Structural performance level
The determination of the performance level is based on the Applied Technology Council (ATC) 40. Reference [11] provides the limits of the roof drift ratio evaluated on the performance points in order to evaluate the performance of ductile structures with nonlinear static analysis. The parameters used are maximum total drift and maximum inelastic drift. The limit of roof drift ratio according to ATC-40 is presented in Table 2. The value of drift ratio can be calculated using the equation as follows: Where:

Vulnerability analysis on HAZUS method
The Hazard-US (HAZUS) method is a method developed by the National Institute of Building Sciences (NIBS) [12] for risk assessment due to various disasters. This method was issued by The Federal Emergency Management Agency (FEMA) in 1997 to assess the losses caused by the earthquake in the US. The HAZUS earthquake method is one of the analysis model used to determine the level of building vulnerability in an area due to earthquake disaster.
The HAZUS method produces vulnerability curves with a high degree of accuracy that can be applied in analyzing the building vulnerability level. The conditional probability of being in, or exceeding, a particular damage state, ds, given the spectral displacement, S d , is defined by the function: where:

Building damage economic value
According to [13], the economic value of the damage is the value of building losses due to earthquakes that include structural and non-structural damage. The value of the loss is measured by the percentage value of the building value itself. The relationship between the level of damage, the percentage of damage, and the economic value of the damage follow the research developed in the USA (FEMA, 1999) as presented in Table 3.
Where: The result of lateral earthquake load distribution in each story is presented in Table 4.

Pushover analysis result
Pushover analysis result shows the capacity curve of x and y-direction. The relationship between the base shear (ton) with displacement (m) is presented in Fig. 5. Based on the result, the x-direction capacity curve is higher and tend to be upright than the y-direction capacity curve. The slope of the capacity curve shows the stiffness of the structure building. The Curve that tends to be upright shows that the building has a higher stiffness value than the sloping capacity curve.    5. shows that the x-axis direction structure is more rigid than the y-axis direction because the larger column dimension is in the x-axis direction so that the column inertia moment is greater toward the x-axis than the y-axis.

Performance level
The intersection result of the capacity curve and spectrum demand is obtained performance point value. Performance point values, effective natural vibration time and effective viscous damping based on the ATC-40 method in x and y-direction is presented in Figure 6 and Fig. 7.  Performance levels can be determined based on the drift ratio criteria obtained when the performance point is reached. The calculation of drift ratio on x and y direction of pushover analysis using the equation (1) Based on the calculation of the drift ratio, both maximum total drift and maximum inelastic drift ratio show that the performance level of the structure building is Immediate Occupancy, which means that only very limited structural damage has occurred, so the building is safe and can be directly used.
Based on the two figures, the base shear value result of pushover-x is greater than pushover-y, on the contrary, the displacement obtained from the performance point of pushover-y is slightly higher than pushover-x. The displacement value of the pushover analysis results will affect both the maximum total drift ratio and the maximum inelastic drift ratio.
The larger the displacement value on the performance point, the greater the maximum total drift ratio will be.The recapitulation of performance point parameters values based on the analysis is presented in Table 5.

Seismic design level
Seismic design level of each building can be determined from the result of a field survey and structural design data used. Based on the previous research [9], the regulations applied to the seismic design of each building type are different from one another, starting from the Pre-code, Low-code, Moderate-code, to Highcodes. The assumption of code classification for engineering buildings in this research which is developed from the previous research and the development of seismic building design code is presented in Fig. 8. :design of buildings using SNI 03-2847-2012 and thereafter Based on that classification, the building belongs to moderate-code seismic design level category.

Peak spectral displacement
Peak Spectral Displacement value is obtained from the intersection of the response curve and capacity curve graph. The response curve is made based on the attenuation equation that corresponds to the Yogyakarta earthquake characteristics occurred on May 27, 2006, i.e. Joyner -Boore -Fumal attenuation, 1997. While the capacity curve obtained from the analysis of pushover that has been converted in ADRS format.
The attenuation equation is chosen because it alludes to the earthquake mechanism of the shallow crustal earthquake -strike-slip which is earthquake mechanism of Yogyakarta. In addition, Yogyakarta earthquake magnitude is included in the large magnitude quake range required at the attenuation (5,5M-7,5M). The The value of inelastic spectral displacement is obtained by the following empirical formulas and the result is presented in Table 6. S d = 9,8 * S a * T 2 (7) where: S a = Spectral acceleration (g) S d = Spectral displacement (in) T = Time period (s) Table 6 shows the conversion of pushover analysis in the ADRS format based on the results of the S a and S d values due to the attenuation (equation (6) and (7)). In addition, the results of the intersections are presented in Fig. 9.  Fig. 9. shows that based on the intersection of capacity and response curve of the building, the spectral displacement value obtained is 0.863 in.

Results of damage probability analysis
The S d value obtained (0.863 in) used to find the parameter of fragility curve. The building type is included on C1L model (low reinforced concrete buildings). The cumulative probability results are presented in Table 7, while the fragility curve is shown in Fig. 10. Fig. 10. shows that the Student Center building, Yogyakarta State University designed with Moderatecode has a fairly sloping shape of its fragility curve. There is no damage at the first 0.1 in spectral displacement and just after the value of S d above 0.1 in, the probability of damage began to appear. Furthermore, the probability of damage in each damage level is shown in Table 8, while its matrix of damage probability is presented in Fig. 11. Fig. 11. shows the probability value of the building damage reviewed due to Yogyakarta earthquake scenario. The probability of slight damage level is 22.59%, the moderate level is 21.60%, the extensive level is 3.71%, and the complete level is 0.23%, so the total vulnerability value is 48.12%. It is evidence that the probability of building damage is not more than 50%, and in the event of damage, the greatest possible damage is a slight level of damage.

Damage loss analysis (FEMA, 1999)
The value of the economic loss is calculated based on the probability analysis of the building damage for each level of damage. Calculation of Mean Damage Probability (MDP) of the Student Center building is calculated based on equation (4) and the following results are obtained: MDP = (0,0023x1) + (0,0371x0,5) + (0,2160x0,1 ) + (0,2259x0,02) = 4,6921

Conclusions
The conclusions obtained from this study are as follows: 1. The performance level of the building based on pushover analysis is at Immediate Occupancy level, either at x-axis direction or in the y-axis direction. This means that if an earthquake occurs, the structure is able to withstand the earthquake. There is no significant damage to the structure, so it can be directly used without having to be repaired. 2. The vulnerability of buildings at each level of building damage obtained due to earthquakes for the level of slight, moderate, extensive, and complete are 22.59%, 21.60%, 3.71%, and 0.23% respectively. This means that total building vulnerability is 48.12% and the possibility of no-damage is 51.88%.
3. The estimated value of economic losses based on the method of FEMA-1999 approach obtained by MDP value of 4,692. This means that the building has a small damage economic loss probability of 4.692% due to the 2006 Yogyakarta earthquake scenario.