Assessment of social sustainability performance for residential building

Abstract The need for residential buildings has been increasing over time. Because a residential building has a long-life cycle and influences the life of the community in which it is situated, it is necessary to assess performance in terms of social sustainability. Social sustainability is a component of sustainable development that has a relationship with community interest. This study proposes a framework for evaluating the social sustainability of residential buildings using social network analysis (SNA) and fuzzy comprehensive evaluation (FCE). We use SNA to visualize relationships between the stakeholders with their interests pertaining to social sustainability variables or criteria. FCE is applied to quantify the perceptions of stakeholders regarding the performance of the social sustainability criteria. The assessment framework compares the stakeholders’ expectations with existing performance to determine whether the building has provided maximum benefits to society. We applied the framework to a real case involving a residential building in Indonesia to demonstrate its efficacy. The results demonstrate that the model is capable of assessing the social sustainability performance in this type of construction.


Introduction
Adequate housing is a basic need and has been declared by the United Nations to be a human right (UN Habitat 2014).Shelter provides not only a place in which to live but is also a place where relationships between society and the environment exist to support various human activities such as learning and hosting a business meeting.Housing is an essential factor in determining the quality of life and the well-being of people (Golubchikov and Badyina 2012).In many nations, especially countries with low and moderate per capita incomes, population growth and urbanization are increasing while land availability for housing is limited.These circumstances mean that housing patterns are based not on horizontal spread but rather on vertical growth.Consequently, the advent of living arrangements that require residents to reside on different floors changes the way that they socialize.
Offsetting certain positive benefits, the construction of multi-story residential buildings can negatively affect existing communities.The existence of the building creates long-term footprints that could change the behavior of social communities (Almahmoud and Doloi 2020).For example, during the construction process a building can have adverse impacts on the community and its environment such as construction-caused damage, dust, flooding, and traffic interference.Therefore, it is important to ensure that the building can reduce the negative impacts to stakeholders as well as the environment during the construction and operation phases to achieve overall efffectiveness (Rohman, Doloi, and Heywood 2017).
The concept of project success suggests that satisfaction should be measured from the point of view of its constituent stakeholders.According to De Wit (1988, 165), "a project can be considered an overall success if the project meets the technical performance specification and/or mission to be performed, and if there is a high level of satisfaction concerning the project outcome among key people in the parent organization, key people in the project team and key users or clientele of the project effort."As such, identifying different views and interests of stakeholders is essential for achieving project objectives.In the context of a building or residential construction the stakeholders can comprise both individuals and groups that affect or can be affected by decisions associated with the design and implementation of a residential project (Freeman and McVea 2001;Phillips, Freeman, and Wicks 2003).Generally, they can be categorized as the construction industry, the users, and the neighboring community (Almahmoud and Doloi 2015).
Indeed, satisfaction by stakeholders is an important criterion for the success of a particular project (Silvius and Schipper 2022).This satisfaction can be understood as achievement of expectations in actual performance (Li, Ng, and Skitmore 2013) and inabilty to meet expectations can cause stakeholder opposition which can disrupt the project and create circumstances where conflicts and controversies ultimately lead to project failure (Olander and Landin 2005;El-Gohary, Osman, and El-Diraby 2006;Majamaa et al. 2008).Accordingly, stakeholder management needs to be implemented appropriately to meet the expectations of relevant individuals and communities.
With regard to communities, the concept of sustainable development is quite relevant in assessing the success of construction projects.Referring to the definition originally set forth by the Brundtland Commission (WCED 1987), sustainable development is "development that meets needs of the present without compromising the ability of future generations to meet their own needs."There are three main aspects of sustainability that are often referred to as "the triple bottom line," namely economic, environmental, and social (Elkington 1998).However, by comparison to the economic and environmental dimensions, the social aspect has received the least attention (Bostr€ om 2012), especially in the construction-management area (Rohman, Doloi, and Heywood 2017;Almahmoud and Doloi 2020;Fatourehchi and Zarghami 2020).Social sustainability can be defined as a life-enhancing condition and a process within communities that can achieve that condition (McKenzie 2004).It is closely related to fulfilling a community's needs to enhance the quality of life of its residents.Therefore, assessment of social sustainability is essential for evaluating and improving a project's performance and ensuring that its social benefits accrue to the stakeholders.
Previous studies have proposed methods to assess the social sustainability performance of construction projects by considering the needs and interests of their stakeholders.Sodangi (2019) developed a model using mean score and expert judgment to determine the weight of each indicator.Almahmoud and Doloi (2015) and Doloi (2018) proposed social network analysis (SNA) as a tool to quantify the needs and interests of stakeholders and subsequently estimated the level of social satisfaction by comparing their interests and expectations.While these works have enhanced current understanding of how to assess social value, they still have limitations related to the accuracy of quantifying stakeholder satisfaction which are often subjective and vague.The results of these analyses thus tend to be excessively optimistic or pessimistic and such outcomes can lead to suboptimal decision making (Li et al. 2015).Fuzzy comprehensive evaluation (FCE) is a tool which is used to deal with ambiguity and is capable of overcoming these limitations by producing more comprehensive results (Liu and Leng 2019).
This study aims to demonstrate how SNA and FCE can be combined to more adequately assess social sustainability and to accommodate the inherent subjectivity of stakeholders.We employ SNA to illustrate stakeholder interests by considering appropriate social criteria and FCE to determine how project performance can accommodate uncertainty.We then demonstrate the application of the framework to assess performance in terms of social sustainability in a case study of a residential building in Alam Sutera, South Tangerang, Banten Province, Indonesia.

The concept of stakeholder satisfaction
Measuring social sustainability as related to community satisfaction cannot be separated from stakeholder management.Li, Ng, and Skitmore (2013) defined stakeholders as "those who can influence the project process and/or final results, whose living environment is positively or negatively affected, who associated direct and indirect benefit and or losses."To make such determinations it is necessary to identify the different views and interests of stakeholders and the ways in which project costs and benefits are distributed (Almahmoud and Doloi 2015).This process can be confounded by the fact that once a project is operational the interests of various stakeholders can change (Almahmoud and Doloi 2015).Regardless of this critical source of complication, it is necessary for a project's sponsors to seek to satisfy all stakeholders.Li, Ng, and Skitmore (2013) defined stakeholder satisfaction as the perceived achievement of reality against expectation.
According to the project-success concept, satisfaction should be measured from the perspective of a project's stakeholders according to three criteriaoften referred to as the "Iron Triangle" and comprising cost, time, and quality.However, the notion of social sustainability requires a more expansive approach (Rohman, Doloi, and Heywood 2017).Baccarini (1999) has divided the criteria for success into "project-management success" and "product success."There are, in turn, three components of project-management success: (1) meeting time, cost, and quality, (2) quality of the project-management process, and (3) satisfying the needs of a project's stakeholders that is related to the project-management process.Meanwhile, the components of product success are (1) meeting the strategic organizational objectives (project goals) of the project's sponsors; (2) satisfaction of user needs (which are related to the project purpose); and (3) satisfaction of stakeholder needs (which are related to the product).
With regard to a construction project, stakeholder satisfaction can be achieved by minimizing the negative impacts of the project's existence.Figure 1 illustrates how each stakeholder has different interests with respect to each criterion in a project.Focusing on one criterion in all likelihood cannot increase the satisfaction of all stakeholders.Therefore, it would be more appropriate to formulate strategies to include all the criteria as part of a process of engaging with the full range of stakeholders.

Sustainable construction and social sustainability assessment
The concept of sustainable development is relevant for assessing the success of construction projects.For decades, construction practitioners have recognized the negative impacts of their activities on communities and the environment (Wirahadikusumah and Ario 2015).Sustainable construction is the application of sustainable development in the construction industry.Gunatilake (2013) identifies four key features of sustainable construction that require consideration (1) the whole life cycle of the project from inception until demolition, (2) all three dimensions of sustainability, namely environmental, social, and economic, (3) both technological and non-technological solutions, and (4) needs of present and future stakeholders.
Social sustainability is a feature of sustainability more generally that can be defined as improving the quality of life by creating harmonic living (Enyedi 2002).McKenzie (2004) added that social sustainability is a life-enhancing condition and a process within communities that can achieve that condition.
Additionally, social sustainability has been described as satisfying people's extended needs, preserving nature, fulfilling social justice, and fostering human dignity and participation (Littig and Grießler 2005).Santa-Cruz et al. (2016) stated that in principle, social sustainability consists of two dimensions: social equity (access to services, opportunities, facilities, and adequate transportation and adequate infrastructure) and sustainability of the community itself (security, social interaction, and public participation).
Numerous criteria for social sustainability have been developed in the context of literature on residential building.Ardda, Mateus, and Braganc ¸a (2018) proposed six categories of social sustainability in a building namely: cultural, heritage, indoor environmental quality, health and well-being, safety and service quality, and accessibility.In another study, Maleki et al. (2019) identified several quite similar criteria: safety, security, sense of belonging to a place, comfort, and esthetics.Ahmad and Thaheem (2017) developed two functional indicators that they termed (1) esthetic and innovative design and (2) user comfort and safety.Considering another perceptive, Yigitcanlar, Kamruzzaman, and Teriman (2015) focused on access to public facilities and emergency services, crime prevention and safety, and traffic calming as essential to social sustainability.Fatourehchi and Zarghami (2020) highlighted five categories: site considerations and equipment, health and comfort, safety and security, practitioner interactions, and architectural factors.
With regard to the assessment of the social sustainability of construction projects, several previous studies introduced a framework to assess their performance.Valdes-Vasquez and Klotz (2013) suggested that social sustainability should fully reflect the perspectives of the stakeholders in a project.In response, Almahmoud and Doloi (2015) and Doloi (2012) proposed a framework to assess social sustainability by managing the needs and interests of different stakeholders using SNA to map their existence and perspectives.Xiahou et al. (2018) combined fuzzy analytical hierarchy process (FAHP) and FCE.The authors use FAHP to obtain the weight of an indicator and FCE to convey social performance.All three studies rely on the opinions of experts to obtain the weight and performance of each social indicator.However, only relying on the use of weighting based on expert appraisals can lead to certain sources of bias (Reza, Sadiq, and Hewage 2014).
While these studies have contributed to improving the assessment of social sustainability performance by accommodating stakeholder interests, they are not fully adequate for accommodating stakeholder subjectivity.The application of FCE can overcome the limitation by assessing the social aspects in more objective terms using a quantitative approach that considers the importance of the criteria and the weight of each stakeholder to provide more comprehensive assessment (Li, Ng, and Skitmore 2013).

Research design and framework
To address the limitations of existing methods, this article proposes a framework to estimate social sustainability performance by integrating SNA and FCE.We use SNA to visualize stakeholder interests by connecting them to appropriate social criteria based on their needs and calculating the weight of each criterion.The analysis then relies on FCE to accommodate subjective judgments to provide more objective results.

Social network analysis (SNA)
SNA is a methodology that analyzes the structure of relations within a network by capturing the interaction and interrelationships (ties) among the actors (nodes) (Wasserman and Faust 1994).This procedure has been used in construction-project management because it can provide insights into relationship structure and integrates a large amount of information which is illustrated in the graph that is produced as part of the SNA (Liang, Yu, and Guo 2017).
There are two types of SNA networks, namely one-mode and two-mode.The one-mode network only links one actor to another actor while the twomode network analyzes the relationships between actors and their associated attributes.Weighting is used to determine the priority of the stakeholders or criteria.SNA identifies the relationships of the stakeholders in the network and shows how they influence the relevant criteria and how they relate to one another.Hadiana dan Witanti asserted that the steps of two-mode SNA mainly involve: (1) identifying nodes and the boundary of the network, (2) linking the nodes, (3) visualizing and projecting the network, (4) analyzing the network, and (5) finding the result.

Fuzzy comprehensive evaluation (FCE)
In the case of FCE, fuzzy refers to the unclear boundaries that represent two or more conditions using linguistic terms such as low, medium, or high.The fuzzy theory which was proposed by Zadeh (1965) introduced three values: (1) the value is not completely true (1); (2) the value is not completely false (0); and (3) both values are inconclusive.The value varies between completely true and completely false which is represented in a fuzzy membership function.Fuzziness is often encountered when evaluating or assessing a condition or status of something in accordance with the previously identified linguistic terms.
FCE is used to solve vagueness problems due to uncertainty and incomplete information in real life (Li, Ng, and Skitmore 2013).It is a comprehensive evaluation method based on fuzzy mathematics and predicated on the principle of fuzzy relation synthesis.This concept helps to quantify variables that do not readily lend themselves to quantification as well as to evaluate the condition of the variables comprehensively (Zhu 2022).FCE provides a more reasonable reference for decision makers because the membership function and the factor are considered comprehensively, which makes the evaluation more reasonable and accurate (Gu et al. 2020;Liu and Leng 2019).With the traditional evaluation method, it is usually difficult to accommodate vagueness problems with the results that are often encountered in decision making (Li et al. 2015).

Development of the framework for residential social benefit assessment
The proposed framework is referred to as residential social benefit assessment (RSBA).It is designed to measure building performance and to satisfy the interests of stakeholders related to social sustainability.Figure 2 shows how the RSBA model works.Stakeholders' interests are mapped based on their expectation regarding the social criteria which are compared to the actual social performance of the building.The expectation of the stakeholders is called the targeted residential social benefit (TRSB) and the actual or real performance is referred to as actual residential social benefit (ARSB).
In a case where the actual score (ARSB) does not meet the target (TRSB), the building performance needs to be improved by identifying the criteria that have not been met.Figure 3 illustrates the comparison of ARSB and TRSB over time across residential building timelines.RSB GAP is the difference in the ARSB and TRSB scores.The timescale shows the change of TRSB and ARSB scores through time-phased assessment whether there is an improvement or not.
Before assessing a residential building, it is essential to identify the stakeholders and the social sustainability criteria.In this article, the stakeholders are classified into three categories by distinguishing the industry, the users, and the neighboring community as shown in Table 1 (see also Almahmoud and Doloi 2015).
First, industrial communities are comprised of people who provide or operate the building such as the owner, operational manager, and staff/employees.Second, users are people who use the building such as residents and visitors.Finally, the neighborhood refers to people who live nearby and/or are affected by the building such as residential neighbors, commercial neighbors, and those who use the roads and/ or sidewalks in surrounding areas.Based on a literature review, we identified 21 social sustainability criteria and validated them through a preliminary survey involving experts in the field (Rohman, 2022).We interviewed three experts with academic backgrounds who had 16-20 years of work experience and had been involved in numerous residential building projects.The social sustainability criteria which are relevant to this research based on the opinions of the experts are presented in Table 2.
We used SNA to visualize and determine the importance of stakeholders and social sustainability criteria.As outlined above, there are two network types in SNA, one mode and two modes.Since a one-mode network can only link one actor to another, we employed the two-mode network to analyze the relationship between actors and their associated criteria or attributes such as the stakeholders and their interests to better understand the social network in this case study.The two-mode matrix between stakeholders (S) and social sustainability criteria (C) was formed as the input of the SNA which was C i,j S n,m ¼ 1 if the stakeholders i,j are interested in criteria n,m and C i,j S n,m ¼ 0 if otherwise.
Eigenvector centrality (C EV ) determines the weight of the stakeholders and the importance of the sustainability criteria.C EV can be shown in Equation 1 (Borgatti and Everett 1997) as follows: where n 0 is defined as the size of the vertex set in which the node belongs.
To make the results more intuitive, C EV is normalized as outlined in Equations 2 and 3: where RC EV is the sum of C EV , s n,m is defined as the weight of each stakeholder and c i,j is defined as the weight of each criterion.
We then subsequently used FCE to compare the TRSB and ARSB.FCE was adapted from (Li, Ng, andSkitmore 2013, Li et al. 2015) which involves three steps as explained below.
1. Determine the weight of the stakeholders and the social sustainability criteria: Set the weight of the stakeholders and the social sustainability criteria obtained from the previous stage so that    they can be described using the following matrices: where s i (i ¼ 1, 2, 3,., m) is defined as the weight of stakeholders and c i (i ¼ 1, 2, 3,., m) is the performance assessment.The weight of each criterion must follow Equations 6 and 7: 2. Calculate the membership function: The qualitative result is described by qualitative fuzzy lan- where v i (i ¼ 1, 2, 3, 4, 5) represents five-level classifications.As mentioned earlier, stakeholder satisfaction can be defined as achieving stakeholders' expectations and actual reality performance (Li, Ng, and Skitmore 2013).Based on these definitions, the core of TRSB will follow an amount of expectation where the five levels can be described as very low, low, neutral, high, and very high.The score of ARSB will follow the amount of actual performance where the five levels can be described as (1) very poor, (2) poor, (3) neutral, (4) good, and (5) very good.3. Set up a fuzzy evaluation matrix: Classify stakeholders by their level of perception on each social sustainability criteria by considering their weight and then normalizing it, as showed in Equation 8.
where P T t¼1 s is the sum of all stakeholders' weights.After membership is calculated, the fuzzy matrix is determined by Equation 9.The overall scores of TRSB and ARSB can finally be quantified by taking into account the set of grade perception V ¼ (V¼ [1,2,3,4,5]) as in Equation 12.
Application of the RSBA model in a case study We applied the framework in a actual case study to demonstrate the application of the RSBA model.The focus of this exercise was a high-rise residential building in Alam Sutera, South Tangerang, Banten Province, Indonesia (see Figure 4).South Tangerang is proximate to Jakarta (about 30 kilometers to the west of the capital city of Indonesia).The building has been operated for approximately four years by two companies as part of a joint-ownership scheme.Its land area is around 10,000 square meters (m 2 ) and a building area of around 90,000 m 2 with 27 floors and three basements.
There are nine floors for office space and housing.

Mapping interests
The first step involved mapping the stakeholders' interests by collecting data from eleven stakeholders of the residential building.We identified the stakeholders based on their knowledge about the building and categorized them into three key communities according to their roles in the building.The respondent stakeholders were asked to complete a questionnaire that provided a checklist of social criteria related to their needs and interests.They were advised to select "zero" if the items in the checklist were not related to the social criterion and to select "one" if the relation did exist between them.The respondents were then asked to indicate their expectations and perceived reality using a Likert scale range from 1 to 5. To map the stakeholder interests, eigenvector centrality (C EV ) from the SNA output was used to determine the weight of stakeholder's importance and the social sustainability criteria.
Figure 5 shows the social network of the stakeholders and the social sustainability criteria in the case study.The existence of a line provides information about the relationship between the stakeholders and variables.The thicker the line connecting the nodes, the stronger the relationship.Accordingly, we can infer that almost all stakeholders are related in the social network with respect to this study case.
Table 3 shows the importance level of the stakeholders' role (Si) in the case study based on C EV value.The higher the C EV , the more important the role of the stakeholder and the social sustainability criteria are in the case-study network.The eigenvector centrality obtained from the SNA needs to be normalized (normalized weight) based on Equations 2 and 3 to be used in the performance assessment in the next section.
According to Table 3, Owner 1 and Owner 2, as well as the building management, play important roles regarding the social sustainability performance of the apartment building.This is likely because they are the decision makers regarding policies determining building operations.In addition, building users have a relatively important role because they can directly influence the policies carried out by the building operators.
Concomitantly, Table 4 presents the relative importance of the social sustainability criteria (Ci) based on C EV value.The table offers insight on the most important social sustainability criteria in the building (highlighted in bold in the table), namely: (1) ensuring proper traffic management (C13), (2) providing parking area (C12), (3) paying attention to health from pollution/environmental problems (C3), (4) creating a sense of safety (C2), and (5) offering easy access to public facilities (C14).

Social sustainability performance assessment
We analyzed the performance of the social sustainability criteria based on the relative weight of the stakeholders and the social sustainability criteria using FCE.All of the social sustainability criteria are represented using a membership function which is calculated using Equation 8 based on the stakeholders' opinions.For example, with regard to the most important social sustainability criteria (providing proper traffic management (C13), there were nine stakeholders' opinions.According to responses on the questionnaire, three stakeholders (S6, S8, S11) had high expectations and six stakeholders (S1, S2, S3, S4, S7, and S9) had very high expectations for C13.
Based on the stakeholders' normalized weight in Table 3, the assessment is carried out using an average which produces an absolute value.It can be difficult to determine whether the value is in the high or very high category.Therefore, the membership function of high expectation with respect to C13 can be calculated as the total weight for S6, S8, and S11 and we obtained 0.110 þ 0.128 þ 0.010 ¼ 0.247.Meanwhile, very high expectations were proposed by S1, S2, S3, S4, S7, and S9 and we obtained 0.121 þ 0.156 þ 0.156 þ 0.083 þ 0.104 þ 0.073 ¼ 0.693.Therefore, the overall score of C13 can be calculated as 0.247 þ 0.693 ¼ 0.940.
Based on a similar method, the membership function TRSB for C13 can be obtained as follows: TRSB of S 0 13¼ s 0 13:1 , s 0 13:2 , s 0 13:3 , s 0 13:4 , s 0 The above membership function can be normalized as C13 ¼ (0,0,0,0.263,0.737)which means the expectation value of TRSB for C13 is 73.7% as very high and 26.3% as high.This is different if the assessment is carried out using the mean which produces an absolute value; it is difficult to explain whether the value is categorized as high or very high.
The membership function of TRSB for all social sustainability criteria is presented in Table 5 and the membership function of ARSB in all social sustainability criteria is shown in Table 6.
Meanwhile, the level of expectation and actual performance can be quantified with Equations 10-12.
Based on these calculations, we determined that the difference between the TRSB value and the ARSB value is 0.04.As described above, the unit is based on a Likert scale from 1-5.This means that the gap between the actual performance and the expectation is not very large even though the residential building has not achieved overall stakeholder satisfaction according to its social sustainability performance.As such, improvement is still necessary to close the outstanding gap.
To better understand the social sustainability performance in this case, we developed the spiderweb diagram shown in Figure 6 to compare the difference between TRSB and ARSB.TRSB is shown as an unbroken line and ARSB as a dashed line.The figure clearly displays the criteria that have fallen short of expectations, those that have met expectations, and those that have exceeded expectations.
Based on the assessment, it is apparent that the building has met or even exceeded stakeholders' expectations on several criteria.Table 7 shows several social sustainability criteria (10 out of 21) that exceeded expectations.Based on the ranking, the first criteria is "providing local job opportunities" (C5) and the second is "providing economic benefits for the surrounding community" (C6).Meanwhile, the third criteria is "providing a sense of safety" (C2).With regard to improvement in social sustainability performance, it is necessary to highlight several criteria that have fallen short of expectations (see Table 8).A total of 10 out of 21 criteria have not met stakeholders' expectations.Therefore, the building management should implement improvements to increase the score on these performance criteria.The first criterion that should be more effectively addressed by the building management is "providing parking area" (C12) because this is the second most important criteria overall.The most important criterion -"providing proper traffic management" (C13)is also still below stakeholders' expectations.

Conclusion
This study proposes a framework for assessing the social sustainability performance of a residential building by combining SNA and FCE.This framework is intended to accommodate the stakeholders' needs and interests in the assessment of social sustainability performance based on their perceptions which are often vague and difficult to quantify.The framework is implemented and tested in a case study of an existing residential building in Alam Sutera, South Tangerang, Banten Province, Indonesia to demonstrate its applicability.The results revealed that the framework can assess social performance by linking stakeholders' interests and needs to social sustainability criteria.We found that several variables associated with the case-study building have met stakeholders' expectations.However, the building management should address underperforming variables to increase the social sustainability score so as to achieve the target performance.The model can help decision makers to quantify and assess social performance in selecting appropriate management strategies that maximize social benefits and facilitate project success.Each variable makes visible stakeholders' opinions by mapping the SNA and FCE results.This research can contribute to the design of evaluation frameworks for social sustainability and improvement of building performance to achieve more sustainable construction.
It is though important to note that regardless of the care applied in executing the analysis, this research has limitations due to the fact that we focused on only on one residential building in Indonesia.The methodology described here merits replication for other types of buildings to enhance social sustainability.
Calculate and normalize the fuzzy evaluation matrix and the weighting criteria: The fuzzy evaluation matrix (S 0 ) can be calculated and normalized by considering the weighting matrix of social sustainability criteria as shown in Equations 10 and 11.B ¼ C Á S 0 ¼ c 1 , c 2 , c 3 , :::, c n ½ :

Figure 5 .
Figure 5.The social network of case study.

Table 1 .
Identification of Stakeholders.

Table 3 .
Eigenvector centrality (C EV ) and the weight of stakeholders (Si).Roles denoted in bold indicate stakeholders that have a higher influence according to the eigenvector centrality value.

Table 4 .
Eigenvector centrality (C EV ) and the weight of sustainable criteria (Ci).

Table 7 .
Social sustainability criteria that exceeded the expectation.

Table 8 .
Social sustainability criteria that not yet fulfill the expectation.