Earthquakes are devastating natural disasters affecting human safety and causing high degrees of damage to infrastructures (Bryant, 1993; Fiedrich et al., 2000; Marcuson et al., 2003), and most cities and regions throughout the world have experienced devastating seismic events (Guo-hua et al., 2009). Since social, economic, and engineering consequences of earthquake disasters are of significant research interest, case studies, improved approaches, and statistical procedures have been established to investigate the effects of these disasters (Akason et al., 2006). The prevention of seismic risks mainly necessitate understanding of engineering solutions (Tansel, 1995). During 2010, on January 12th, Leogane, Haiti (M_w = 7.0), on February 27th, Maule, Chile (M_w = 8.8), and April 14th, Qinghai, China (M_w = 7.1) earthquakes occurred. Among these disasters, the Haiti earthquake caused more than 230,000 deaths making it the 6th deadliest earthquake. The Chilean earthquake was the seventh strongest earthquake since 1900.

Recently, China experienced major earthquakes: The 2014 Ludian earthquake struck Ludian County, Yunnan, (M_w = 6.1) on August 3rd 2014. On April 20th 2013, Ya’an earthquake (M_w = 6.9), on April 14th 2010, Qinghai earthquake (M_w = 7.1), on July 9th 2009, Yunnan earthquake (M_w = 6.0), and on May 12th 2008, Sichuan earthquake (M_w = 7.9) struck China (Ou and Li, 2010). Among these disasters, the most devastating, the Sichuan earthquake, resulted in the collapse of 5 million buildings and damage to more than 21 million. Human loss was also high with 360,000 injured, over 60,000 dead, and about 14 million people displaced. South-west China lies in an area that is seismically active. In 1970, 15,000 people dead in Yunnan earthquake with a magnitude 7.7. Figure 1 locates historical earthquakes (M_w > 6) within circles on the map (Gu et al., 2015).

Among the recent disasters in China, the most devastating was the Sichuan earthquake which produced a magnitude of 7.9 and occurred 92 km northwest of the Chinese Sichuan province on May 12th 2008. The Sichuan earthquake occurred on the margin of the Tibetan Plateau; in contrast to the April 14th 2010 event, the Qinghai earthquake occurred in the plateau’s interior. The Sichuan province has one of the highest populations in China with 100-million people and a heavily industrialized region (Yongshuang et al., 2007; Huang, 2008; Zifa, 2008; Wu and Olson, 2009). This earthquake was the most devastating since the 1976 Tangshan earthquake, at where 240,000 people died. In the Sichuan earthquake, 5 million buildings collapsed, and 20 million buildings were damaged. China’s Information Office of the State Council reported that over 60,000 people died and 360,000 injured. The total economic loss was estimated more than 80 billion US dollars (ABS, 2008; EERI (Earthquake Engineering Research Institute), 2008; Miyamoto, 2008; Cui et al., 2011; Perlstein and Ortolano, 2015). The earthquake history in China appears in Table 1, and Figure 2 gives the population distribution and the seismically active regions.

During the Sichuan earthquake, different types of reinforced concrete (RC) residential buildings were severely damaged or collapsed. Damaged/collapsed buildings represented construction during different time periods. Buildings constructed strictly according to design codes (Gbj11-1989, 1989; GB 50011-2001, 2001; GB 50011-2002, 2002) and with good workmanship performed well during the earthquake. However, a great number of damaged or collapsed residential buildings represented construction with poor workmanship, substandard materials, and inadequate reinforcement (Burchfiel et al., 2008; Chen and Somerville, 2008).

This study develops a detailed structural evaluation for residential low- to mid-rise buildings in China for a structural risk assessment and mitigation. A detailed probabilistic risk analyses were conducted with data from 900 different earthquakes that occurred between 1973 and 2016 to give an assessment approach. The earthquake data allow plotting occurrence probabilities of earthquakes by using the Poisson model and the magnitudes of the seismic events, and the investigation considers collapse and damage to existing low- to mid-rise residential buildings that dominate the building inventory in seismically active regions in China. Among the results of the investigation, relations and links between safety and risk are defined. The objective of the study is to evaluate the low- to mid-rise residential buildings in China in terms of earthquake resistance. In the study, typical RC residential buildings were designed to define the seismic vulnerability. The sample models represent variety of existing RC residential buildings. Non-linear seismic analyses of RC residential buildings were completed.

In last decade, majority of the existing buildings have been traded with low- to mid-rise RC buildings including schools, hospitals, governmental offices, hotels, and financial and business centers. The majority of such buildings were exiguous designed and are similar to each other in all over the China. Majority of the existing residential buildings are suspected too vulnerable to strong earthquakes (ABS, 2008; EERI (Earthquake Engineering Research Institute), 2008; OCHA, 2008; USGS, 2010). Figure 3 gives an idea about example of the existing buildings. Low-rise residential buildings have the majority that got damage/collapsed. Severe damage at beam-column connections also occurred. Various types of damages have been observed in the earthquake, and these damages are given in Figure 4. As seen in Figure 4, damaged and collapsed buildings in the region are weak buildings against seismic loads. Therefore, buildings got severe damages.

Masonry walls are common as infill walls for both interiors and exteriors (Park et al., 1985). Most of the RC buildings have shop openings on the ground floor; each shop occupies one bay in the direction parallel with the street with the shop’s front open to the street and separated from the other shops by partition walls in the direction perpendicular to the axis of the street. This design makes buildings particularly vulnerable to soft-story collapse in a direction parallel with the street. The damage could also result from a lack of consideration in the design phase for the change in relative stiffness provided by infill walls between the ground and the upper floors. Constructional material is the ranging from natural stones to bricks and blocks. Designing these structures for satisfactory structural behavior is a demanding task. Large displacements and damage are sustained by the ground floor column, where, relative to the stiffer upper floors, most of the deformation demands from the earthquake occur. Figure 5 shows a typical RC building with a soft story at the ground level. Soft stories are considered as weak to seismic loads, and many of the damages are observed in soft stories in the previous earthquakes experienced.

Risk is defined with the probability of occurrence of an event or as the magnitude of the consequences in a period (Solomon et al., 1983). Probabilistic seismic risk is the probability of occurrence of a destructive seismic event in a location in a period (Glaser et al., 2007). When investigating the earthquake occurrences in a location, an assumption for probabilistic assessment is a linear relationship between magnitude and frequency that may be obtained as a function of magnitude (Belazougui et al., 2004; Boylu, 2005; Glaser et al., 2007; Lia et al., 2014). Recent earthquake data were taken into account to determine the probabilistic seismic hazard between 1973 and 2016. Establishing return periods and probabilities of earthquake occurrences in the regions that have high seismic risk is important endeavors to construct high strengthen buildings. Researchers use various statistical methods to determine seismic risk and to estimate the probabilities of earthquake occurrence, but the Poisson model is the most common method mostly to expose potential seismic risks due to its accepted accurate results in terms of probability of failure. Among the data for Poisson modeling is magnitude which clarifies the effects of an earthquake for determining seismic risk.

This study applies the Poisson model estimation and earthquakes between 1973 and 2016 that provides data for the methodology. In total, data from 900 earthquakes were taken into account. In Table 2, seismic risk values for selected earthquakes are given. In Table 3, probability values using the Poisson model for different periods are given. Figure 6 defines the cumulative distribution and total distribution diagrams. As seen in Figure 6, magnitudes in between 4 and 6 have the highest frequency, while magnitudes in between 7 and 9 have the lowest frequency. In the cumulative perspective, magnitudes between 4 and 6 constitute 60% of the all seismic activity. Detailed figures for probability definition appear in Tables 2 and 3.

Figure 7, sketched with the Poisson model, represents the probability surface which gives earthquake occurrence probabilities for different time periods in continuous transactions. Probability surface was defined with probability of failure, Modified Mercalli Index, and years. The earthquake occurrence probability curves in the assessment process for 20, 30, 40, 50, 60, and 80 years are given in this study.

Based on this assumption, for the earthquake occurrences, Poisson distribution is determined for given case. The cumulative frequency distribution is the probability of occurrence of N or fewer earthquake considerations in a period, t is given in Eq. 1 (Feller, 1968): (1)F(N,t)=p(k;λt)=∑λ=0N(λt)kk!e−λt where λ is the frequency.

In particular, the probability of number of earthquakes in a period, t, is: (2)P(0,λt)=e−λt.

Times between earthquake occurrences show a negative distribution. Thus, the probability of a time period given between two earthquake considerations in (t, t + dt) will be calculated as: (3)P(t)=−λe−λt.dt.

The probability of earthquake occurrences, F(t), can be expressed as 1 − p (0; λt): (4)F(t)=1−eλt

In given case, for magnitude of M₁, the probabilities of failure (P_f) in t years can be directly calculated with as given in Eq. 5 (Tabban and Gencoglu, 1975; Tuksal, 1976): (5)Pf(M≥M1)=1−e−n(M≥M1)t where n(M) is the annual numbers of magnitude of M₁ and greater earthquakes. The return periods for the considered earthquakes in the analysis can be obtained as given in Eq. 6: (6)Q=1n(M≥M1)

This study presents a detailed seismic safety and risk assessment for low- to mid-rise residential buildings in China for structural risk control and integrated health monitoring and assesses the reasons structural damage and collapse occurred in the recent earthquakes. In the probabilistic seismic risk assessment part, risk was considered in terms of probabilistic evaluation. Earthquake occurrence probabilities by using the Poisson model were sketched. Various earthquake magnitudes were considered for the analyses. Earthquake performance of typical low- to mid-rise RC residential buildings was investigated using non-linear approaches. Definition of structural behavior of existing buildings is very important for mitigating the consequences of a future earthquake. New buildings depend on many different parameters which are, reasonably, mostly construction and architectural requirements. Soft stories are one of reasons for damage in earthquakes. Consequently, avoiding the soft stories in buildings would be an important step of assuring safer buildings against to earthquakes.

KK is the only author in this article. He carried out the research for this research.

The author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.