1 Introduction

Structural damages in buildings during an earthquake are accepted constructively by majority of seismic codes in the world, given that there is no human loss. Undeniably many such damages have occurred due to earthquakes in the past. Many new constructions were unaffected by the improvements of codes, but the earthquake safety of existing buildings is under question. In case of Indian subcontinent, it faces serious earthquake threat due to rapid growth of urban population, wherein nearly 60% of landmass in India is part of moderate-to-severe earthquake prone area. However, over 80% of population is living in this 60% of the land (Pradeep and Murty 2014). In last decade, India has witnessed earthquakes like that of 2001 Bhuj earthquake of M7.7, 2004 Sumatra earthquake of M9.3, 2005 Kashmir earthquake of M7.6, 2009 Andaman Islands of M7.5, and 2011 Sikkim earthquake of M6.8, etc. and an overall of 24 moderate to severe earthquakes were experienced, where around 4.25 lakh casualties and property loss was caused (Ref. Table 1). However, similar intensity earthquakes in the US, Japan, etc., did not lead to such an enormous loss of lives, as the structures in these countries are earthquake resistant (NDMA Guidelines 2007). During the 2001 Bhuj earthquake, numerous recently built Reinforced Concrete (RC) structures collapsed in Gandidham and Bhuj areas in Gujarat (EERI Earthquake Report 2001; Murty et al. 2002). In 2005 Kashmir earthquake, significant damage was attributed for stone masonry buildings, particularly random rubble type structures, which are poor seismic performance (EERI Earthquake Report, 2005). Most of the multi storey RC buildings were non-engineered and sustained considerable damage during the 2011 Sikkim earthquake (EERI Earthquake Report 2012). So, these failures project the urgent need to perform the seismic vulnerability assessment of buildings and suggest possible solutions to retrofit them. Since the detailed assessment of buildings is a complex and expensive task, it cannot be performed on all the buildings in an area. Past reconnaissance survey reports suggest that a simple assessment of existing buildings is necessarily required. Most of the methods follow a three level assessment procedure, namely:

Table 1 Major earthquake in India.
Full size table
  1. (a)

    Phase-I: Rapid visual screening (RVS).

  2. (b)

    Phase-II: Preliminary assessment.

  3. (c)

    Phase-III: Detailed evaluation.

The RVS methodology is referred to as a “sidewalk survey” in which an experienced screener visually examines a building to identify features such as the building type, seismic zone, soil conditions, horizontal and vertical irregularities, apparent quality in buildings and short column etc. that affect the seismic performance of the building. This side walk survey is carried out based on the checklists provided in a proforma for all five typologies of the buildings. Other important data regarding the building including the occupancy of the building and the presence of nonstructural falling hazards is also gathered during the screening. A performance score or RVS corresponding to these features is calculated for the building based on numerical values on the RVS form. The performance score is compared to a “cut-off” score to determine whether a building has potential vulnerability and whether it should be further evaluated by an experienced engineer. The worldwide practices on RVS are as follows.

A number of guidelines were developed by the Federal Emergency Management Agency (FEMA) in the USA for seismic risk assessment and rehabilitation of buildings. The RVS method was originally developed by the Applied Technology Council (ATC) in the late 1980s and published in the FEMA: 154 in 1988. Later, it was developed in FEMA: 178-1989, 1992 (revised), FEMA: 310-1998 developed as revised version of FEMA: 178-1992, and FEMA: 154-1988, 2002 (revised), for rapid visual screening of buildings. A different RVS procedure was developed based on fuzzy logic technique to categorize buildings into five different damage grades (OASP 2000; Demartinos and Dristos 2006). This technique was applied on 102 buildings, which were affected by 1999 Athens earthquake. At the end, damage score was evaluated through a fuzzy inference system. Another RVS method was suggested by National Research Council, Canada, which is based on a seismic priority index. This method accounts for both structural and non-structural factors including soil conditions, building occupancy, building importance, falling hazards, occupied density and the duration of occupancy (NRCC 1993).

The Japanese procedure is based on seismic index (SI) for total earthquake resisting capacity of a story which is estimated as the product of basic seismic index based on strength and ductility indices, irregularity index and time index (TI). The evaluation is based on few parameters and lacks clarity regarding the ranking of buildings based on a scoring or rating system (JPDPA 2001). The New Zealand code recommends a two stage seismic performance evaluation of buildings (NZSEE 2006). The Switzerland applies a three stage concept for evaluating seismic risk. In first stage, seismic risk of a building is roughly estimated and in second stage, seismic risk is studied in detailed. Strengthening measures are employed in the last stage of seismic risk (SIA-2018 2004). The RVS method developed was based on the ratio of roof displacement capacity to roof displacement demand determined for life safety performance criteria and collapse prevention performance criteria by Bogadici University and Istanbul Technical University. Later, this method was improved on the basis of 454 reinforced concrete buildings surveyed after the 1999 Duzce earthquake and classified into four damage grades (Sen 2010; Hassan and Sozen 1997; Ozdemir and Taskin 2006; Sucuoglu et al. 2007).

There have been some efforts in India towards developing RVS methods. Sinha and Goyal (2004) have proposed a methodology for RVS of ten different types of buildings. The procedure requires identification of the primary structural load carrying system and the building attributes that are expected to modify the expected seismic performance for the lateral load resisting system under consideration. A statistical analysis has been performed to develop Expected Performance Score (EPS) for RC buildings based on the rapid visual surveys in Ahmadabad, India (Jain et al. 2010; Keya 2008).

2 Vulnerability studies of cities

Vulnerability assessment of cities has been performed in the past based on population loss estimation and estimation of direct and indirect losses due to various disasters. Some recent vulnerability assessment studies are discussed in the following sections:

2.1 Tehran, Iran

Seismic vulnerability assessment was done in the city of Tehran based on buildings data and damage for buildings for two earthquake scenarios (Nateghi 1998, 2000; Motamed and Ashtiany 2012). Seismic building damage for earthquake scenarios was derived from HAZUS software ranging from minor to major or complete damage (FEMA 1999; HAZUS 2000). Most of the buildings in the study area were found to be vulnerable considering the two earthquake scenarios (South Ray Fault and Floating).

2.2 Dehradun, India

The vulnerability assessment in Uttaranchal was done by Singh (2005). This study developed a methodology for loss estimation based on buildings and population loss. A Geographic Information System based tool was developed for primarily population loss estimation.

2.3 Kanpur, India

Preliminary evaluation was carried out on 30 multistoried RC buildings (IIT-GSDMA Guidelines 2003). The study revealed that large openings, horizontal and vertical projections, presence of soft and weak stories and short column effects are major weaknesses in the buildings at Kanpur from seismic safety point of view (Jain 2006).

2.4 Zeytinburnu, Turkey

This study is an implementation of the earthquake master plan for Istanbul metropolitan area in the Zeytinburnu district with a population of 240,000 and more than 16,000 buildings (Ozcebe et al. 2006) as a part of seismic vulnerability of existing building, a multi stage seismic safety assessment was performed by the Middle Eastern Technical University (METU) in Ankara.

2.5 Gandhidham, India

RVS was conducted on around 20,000 buildings in Gandhidham and Adipur cities (Srikanth et al. 2010). Though there is a large variation in construction practices, about 26% of buildings were RC buildings and 74% were of brick masonry. The study observes an RVS score ranging from 60 to 120 both for masonry and concrete structures which reveals the low quality of construction in this area (Srikanth et al. 2010).

2.6 Nanded, India

RVS was conducted on around 200 buildings in Nanded city. Initial observation reveals that there are wide variety of construction practices; however, predominantly buildings are classified as per material used, i.e., reinforced concrete, stone and brick masonry, tin shade and other buildings, about 70.39% are reinforced concrete buildings, 25.79% are stone and brick masonry, 2.54% are tin shade and 1.26% are other buildings. In RC buildings, score was varying from 40 to 145 and for brick masonry building score was varying from 113 to 130. A detailed study has been done for buildings regarding the structural aspect and effect of earthquakes (Narender 2014).

In the above literature, vulnerability studies of RC and brick masonry buildings have only been considered. An attempt has been made to do RVS for five varieties of building in the state of Himachal Pradesh. The state Himachal Pradesh consists of 1,027,788 buildings with RC, brick masonry, stone masonry, rammed earth, and hybrid building typologies, having a population of 6,856,509 in an area of 55,763 km2. For the purpose of study, 9099 buildings were surveyed with the help of TARU Consultancy Private Limited. The above buildings were selected based on the structural irregularity, terrain condition, construction year, presence of cracks. Figure 1 shows the format of Building Vulnerability Assessment form originally prepared by TARU consultancy and Himachal Pradesh State Disaster Management Authority (HPSDMA). The RVS scores are originally calculated using RVS forms of RC and Brick Masonry. Later, these forms are modified for stone masonry, rammed earth, and hybrid buildings. Figure 2 represents a sample calculations of RVS score for five variety of buildings. As per statistics of surveyed buildings in Himachal Pradesh, around 17% (1541 out of 9099) of buildings are reinforced concrete, 48% (4363 out of 9099) of buildings are brick masonry, 15% (1341 out of 9099) of buildings are stone masonry, 5% (518 out of 9099) of buildings are rammed earth and 15% (1317 out of 9099) of buildings are hybrid. A schematic diagram of assessment of a building is shown in Fig. 3. Present study mainly focuses on RVS analysis of five varieties of buildings and the surveyed buildings distribution is shown in Fig. 4. Preliminary and detailed analyses are beyond the scope of this paper.

Fig. 1
figure 1figure 1figure 1

Proposed format of building vulnerability assessment form prepared by TARU consultancy and HPSDMA

Full size image
Fig. 2
figure 2figure 2figure 2

Calculation of RVS score from RVS sheets for a reinforced concrete buildings, b brick masonry buildings, c stone masonry buildings, d rammed earth buildings and e hybrid buildings

Full size image
Fig. 3
figure 3

Schematic diagram of assessment of a building

Full size image
Fig. 4
figure 4

Location of buildings considered in RVS analysis

Full size image

3 Methodology

The evaluation is based on few parameters of building like building geometry, frame action, hybrid action, pounding effect, structural irregularity, short columns, heavy overhangs, soil conditions, falling hazard, apparent building quality, diaphragm action etc. On the basis of above-mentioned parameters, performance score of the buildings has been calculated. The formula of the performance score (PS) is given in the following equation:

$${\text{PS}} = {\text{BS}} + \sum {[({\text{VSM}}) \, \times \, ({\text{VS}})]}$$
(1)

where VSM represents the Vulnerability Score Modifiers and VS represents the Vulnerability Score that is multiplied with VSM to obtain the actual modifier to be applied to the Basic Score (BS).

The RVS data of existing buildings in the region is plotted as Gaussian (Normal) distribution. This distribution is commonly used for statistical analysis of large data. A normal distribution in a variate X (RVS) with mean μ and variance σ is a statistical distribution with probability density function is given in Eq. 2:

$$f(x) = \frac{1}{{\sigma \sqrt {2\pi } }}e^{{ - \frac{(x - \mu )}{{2\sigma^{2} }}}}$$
(2)

Generally, a cumulative probability refers to the probability that the value of a random variable falls within a specified range (damage range based on design). Cumulative probabilities refer to the probability that a random variable is less than or equal to a specified value. The cumulative distribution function, which gives the probability that a variate will assume a value ≤x, is

$$D(x) = \int\limits_{ - \infty }^{x} {f(x){\text{d}}x = \frac{1}{{\sigma \sqrt {2\pi } }}} \int\limits_{ - \infty }^{x} {e^{{{{^{{ - (x - \mu )^{{}} }} } \mathord{\left/ {\vphantom {{^{{ - (x - \mu )^{{}} }} } {2\sigma^{2} }}} \right. \kern-0pt} {2\sigma^{2} }}}} } {\text{d}}x$$
(3)

In RVS survey, even non-engineers may collect data and calculate RVS score for buildings. Present study considers Himachal Pradesh as a case study to evaluate RVS score for five varieties of buildings. The description of Himachal Pradesh is given in the following section.

4 A case study of Himachal Pradesh

Himachal Pradesh state is located 31.1033°N, 77.1722°E and lies in the Himalayan Mountains, and is part of the Punjab Himalayas. Since the earthquake database in India is still incomplete, especially regarding the earthquakes prior to the historical period (before 1800 A.D.), the largest instrumented earthquake in Himachal Pradesh was 1905 Kangra earthquake (Mw7.8) (Ambraseys and Bilham 2000). The Himalayan Frontal Thrust, the Main Boundary Thrust, the Krol, the Giri, Jutogh and Nahan thrusts lie in this region. However, it must be stated that the proximity to faults does not necessarily translate into a higher hazard as compared to areas located far away. As it is located on hilly terrain, land by its very nature is very fragile; therefore, threat due to earthquakes or landslides is high. Hence, it has been considered as the apt study area. In total of 12 districts, Chamba, Kullu, Kangra, Una, Hamirpur, Mandi, and Bilaspur districts lie in Zone V, whereas Lahul and Spiti, Kinnaur, Simla, Solan and Sirmaur districts lie in Zone IV. The building’s data are collected in every district of HP and procedure for data collection is described in the following section.

4.1 Data collection

An RVS format was developed for the android platform using ODK (Open Data Kit) framework. After designing RVS format for android platform, it is stored on all the tablet computers to be used for data collection. Once it is installed, it does not require Internet connectivity to fetch the blank form. Data are collected from the field on day-to-day basis. On a regular basis, surveyors were provided maps of the marked area along with the number of buildings to be surveyed in that area. After filling all the required information in the RVS format, latitude/longitude is recorded by enabling inbuilt GPS in the tablet computer. Three photographs are taken for each building type in which first two photographs are taken as front and side elevation of the building and third photograph belongs to any major building vulnerability feature. Data are sent to the server using Internet connectivity. These data can be accessed from anywhere using the web application. Surveyors sent the data every evening or next day (in case of non-availability of signal in the remote area) after finishing the work.

Following were few challenges faced during the process of data collection.

  1. 1.

    More time was needed to reach the rural location due to limited travel connectivity.

  2. 2.

    It was difficult to get the information on few parameters such as horizontal bands in the masonry buildings for well-plastered buildings.

  3. 3.

    Foundation details and age of construction was difficult to obtain, if the building was not constructed by the owner.

4.2 Housing potrayal

It has been observed that buildings in Himachal Pradesh are primarily classified into five types viz., reinforced concrete buildings, brick masonry buildings, rammed earth buildings, stone masonry and hybrid buildings from the surveyed buildings data. It also portrays varied building construction methodologies in the state. From the surveyed buildings database, it is categorized into three levels in the preliminary stage of analysis: (a) district wise, (b) type of buildings and (c) building components. Following are few observations from the database which explains the housing scenario in Himachal Pradesh.

  1. 1.

    Huge number of brick masonry buildings and wide variety of building construction are present in Kangra district.

  2. 2.

    New construction of buildings has evolved in Kangra and Solan districts.

  3. 3.

    Presence of medium type of soil is more in every district except Kinnaur and Lahul Spiti.

  4. 4.

    Existence of corner openings of buildings is more in Kangra and Solan districts and substantial openings are more in Solan district.

  5. 5.

    Existence of horizontal bands of a building is more in Kangra district.

  6. 6.

    Presence of soft story, heavy overhangs and short column is rare in most of the districts.

  7. 7.

    Quality of construction is moderate in all districts of Himachal Pradesh.

  8. 8.

    Because of its geographical conditions, only few buildings are observed to be on the flat ground, whereas large number of buildings is located on slopes.

The building parameters according to district are shown in Fig. 5. The primary observation from housing classification is the complexity involved in construction of buildings in the state.

Fig. 5
figure 5figure 5

Statistics of building parameters for 12 districts in HP state

Full size image

4.3 RVS score calculation

Based on the factors given in the form, RVS scores have been calculated for the buildings present in the database. The scores for RC, brick masonry, stone masonry, rammed earth and hybrid buildings are 50–160; 40–220; 30–190; 50–160; and 60–140, respectively. Normal distribution curves are generated for each district and building variety based on RVS scores. District wise normal distribution curves for buildings are shown in Figs. 6, 7, 8, 9, 10 and 11.

Fig. 6
figure 6

Normal distribution curves for RC buildings obtained from RVS score

Full size image
Fig. 7
figure 7

Normal distribution curves for brick masonry buildings obtained from RVS score

Full size image
Fig. 8
figure 8

Normal distribution curves for stone masonry buildings obtained from RVS score

Full size image
Fig. 9
figure 9

Normal distribution curves for rammed earth buildings obtained from RVS score

Full size image
Fig. 10
figure 10

Normal distribution curves for hybrid buildings obtained from RVS score

Full size image
Fig. 11
figure 11

Normal distribution curves as per building variety

Full size image
  1. 1.

    The number of buildings surveyed in the districts of Bilaspur, Hamirpur, Kinnaur, Kullu, Lahul Spiti, Shimla, Solan, Chamba, Kangra, Mandi, Sirmaur and Una are 383, 789, 149, 619, 70, 401, 1553, 513, 1929, 692, 585 and 637, respectively.

  2. 2.

    Around 62% of buildings were constructed majorly in last three decades, which indicates the developmental growth of the state. The construction of buildings rapidly increased during this period, especially in Kangra and Solan districts of HP. There were few buildings that were constructed in 1970s. Most of the surveyed buildings are single, two or three storied and a few are four and five storied. It is observed that the quality of construction was moderate for 45% of RC buildings, 63% of brick masonry buildings, 63% of stone masonry buildings, 60% of rammed earth, and 68% of hybrid buildings. The quality of construction needs to be improved for hybrid, rammed earth and stone masonry buildings. The construction practices are mostly done on hilly terrain due to unavailability of space on flat terrain, and buildings are built in close proximity to each other. Pounding effect was also observed for 55% of buildings.

  3. 3.

    Around 88% of RC buildings, 90% of brick masonry buildings, 80% of stone masonry buildings, 86% of rammed earth, 82% of hybrid buildings are located on medium type of soil. Around 10–15% of buildings are located on hard and soft soils. Buildings located on soft soils are more vulnerable than medium soils and hard soils.

  4. 4.

    Presence of horizontal bands for buildings located in zone IV and V, increase the seismic response of building. In the present survey only 32% of buildings are observed to have bands.

  5. 5.

    Other parameters like openings, short column effect, slope angle heavy overhangs etc. are classified district wise as shown in Fig. 5, but full access of the building is required to get overview of it. The configuration of building is either rectangle, L-shape, T-shape or U-shape for all types of buildings. It is one of the important virtues of earthquake resistant buildings. For buildings, whose configuration is irregular, should make it regular by providing expansion joints, so that, pounding effect due to earthquake can be minimized in high seismic areas.

  6. 6.

    Many of the buildings use less than 230 mm as column size, which is not advisable in seismic areas of zone IV and V. But, IS: 13920—2014 recommends that the minimum size of column should be 300 mm. Around 63% of brick masonry buildings used 150–250 mm, 73% of hybrid buildings used 100–150 mm, 86% of RC buildings used 100–200 mm as column size.

4.4 Estimation of damage using RVS score

From literature, it is observed that the predominant RVS score lies in the range of 60–120 for both RC and brick masonry buildings in Gandhidham and Adipur cities in Gujarat (Srikanth and Pradeep 2010; Srikanth et al. 2010). The mean and median values for both RC and brick masonry buildings are 89.3, 90.0; 78.74, 80.0. Similarly, for Himachal Pradesh, the RVS score lies in the range of 50–160 for reinforced concrete and 40–220 for brick masonry buildings. For which the mean values for both RC and brick masonry buildings are 106 and 120. The analysis was also carried out for stone masonry, rammed earth and hybrid buildings. The RVS scores of five varieties of buildings were given in the earlier section. The mean values are slightly shifted towards right when compared to that of study done in Gandhidham and Adipur cities. It is difficult to set a benchmark for RVS score to understand the damage state of building. A normal distribution curves have been drawn for RC, Brick Masonry, Stone Masonry, Rammed Earth and Hybrid type of buidlings are shown in Figs. 12–16 respectively. Also, all type of buildings is represented in single plot and is shown in Fig. 17. For this purpose, damage is classified as no damage, slight, moderate, severe and collapse based on μ − 2σ, μ − 1σ, μ, μ + 1σ, and μ + 2σ. The mean and standard deviation of all building typologies are represented in Table 2. A representative diagram is shown in Fig. 18. Mean and standard deviations are calculated for each type of building and the state of damage is estimated. Table 3 represents damage state of five types of buildings and Fig. 18 shows the classification of damage of buildings in Himachal Pradesh State.

Fig. 12
figure 12

Normal distribution curve for RC buildings

Full size image
Fig. 13
figure 13

Normal distribution curve for brick masonry buildings

Full size image
Fig. 14
figure 14

Normal distribution curve for stone masonry buildings

Full size image
Fig. 15
figure 15

Normal distribution curve for rammed earth buildings

Full size image
Fig. 16
figure 16

Normal distribution curve for hybrid buildings

Full size image
Fig. 17
figure 17

Normal distribution curve as per building variety

Full size image
Table 2 Mean and standard deviation of buildings as per district wise and building variety wise
Full size table
Fig. 18
figure 18

Classification of damage from normal distribution curve

Full size image
Table 3 State of damage and number of damage buildings of RC, brick, stone, rammed earth and hybrid buildings
Full size table

The distribution of number of buildings in each district of HP with respect to RVS score is shown in Fig. 19. The buildings whose RVS score lies between 130 and 150 are more in number in all Mandi and Sirmaur districts and RVS score lies between 105 and 110 in other districts. These buildings will suffer moderate damage as per damage classification shown in Fig. 18. It means moderate damage buildings are sparsely distributed in all districts. The mean and median of buildings range 110–130 and 105–135, respectively. Around 5.3% of buildings in Sirmaur district and 2% of buildings in other districts will have collapse stage. Buildings whose RVS score is less than or equal to 100 need to be analyzed in detail. These analyses are beyond scope of the paper.

Fig. 19
figure 19

Distribution of number of buildings with respect to RVS score and each district of HP

Full size image

5 Conclusions

An attempt has been made to perform RVS of five varieties of buildings in Himachal Pradesh state. RVS score are calculated for 9099 buildings and normal distribution curves are plotted for each typology of building to understand the distribution of buildings in Himachal Pradesh state. In this study, a detailed analysis of RVS has been incorporated to that of existing one. This RVS study enables us to calculate the risk by assessing the structural vulnerability of different building typologies. It is observed that overall RVS score for different building typologies range from 30 to 220. A statistical approach has been considered to understand the building state using RVS scores. However, major buildings turn out be in middle range of damage index and hence drawing meaningful conclusion is a difficult task. However, there are low RVS score buildings which are potentially vulnerable to future earthquakes. Also it is suggested that preliminary analysis needs to be performed on few buildings and detailed analysis should be done for selected buildings for calibrating RVS scores.

5.1 Recommendations

  1. 1.

    Implementation of the building code regulations for rammed earth, hybrid and stone masonry buildings needs to be initiated in India.

  2. 2.

    Performance of the low-rise buildings constructed using locally available materials must be improved. This factor could lead to a significant reduction of casualties in future earthquakes.

  3. 3.

    Research is needed to investigate and improve the performance of the above buildings.