A Preliminary Long-Term Housing Evaluation System Study in Pearl River Delta, China: Based on Open Building and “Level” Strategy

Abstract

As the region with the earliest housing stock market and the most advanced development in China, the Pearl River Delta has experienced extensive housing demolition and construction, leading to buildings having short lifespans. The environmental pollution generated during this process has brought attention to the concept of green buildings. However, whether due to previous patterns of demolition and construction or the significant impacts of social and economic changes in the current and future housing stock contexts, the comprehensive adaptability of human-centered living spaces remains a crucial issue. This focus is strongly related to the residents’ psychological responses, such as sense of belonging, safety, and atmosphere, across different scales of physical environment. However, most housing evaluation systems regarding sustainable issues are green building evaluation systems. And their concept and practice are often accompanied by a neglect of the interrelationship between people and the built environment, as well as a lack of an appropriate methodological framework to integrate these elements in the temporal dimension. This paper primarily tries to provide new answers to old questions about housing durability by reconceptualizing evaluation systems beyond ecological metrics, while simultaneously challenging accepted answers that privilege material and energy indicators over sociocultural embeddedness. Moreover, an effective housing evaluation framework must transcend purely technical or ecological indicators to systematically integrate the temporal and sociocultural factors that sustain long-term residential quality, particularly in rapidly transforming urban contexts. Therefore, theories closely related to building longevity, such as open building and the “level” strategy, were introduced. Based on this combined methodological framework, selected cases of local traditional housing and green building evaluation systems were studied, aiming to identify valuable longevity factors and improved evaluation methods. Furthermore, two rounds of expert consultation and a data analysis were conducted. The first round helped determine the local indexes and preliminary evaluation methods, while the second round helped confirm the weighting value of each index through a questionnaire study and data analysis. This systematic study ultimately established a preliminary long-term housing evaluation system for the Pearl River Delta.

1. Introduction and Research Question

From 1990 to 2015, urban land expanded 350% in developing regions around the world, which was much faster than their population growth [1]. A similar urbanization movement can be found in the Pearl River Delta (PRD). As one of the most densely urbanized and economically active areas in China, the PRD has consistently served as a pioneer and experimental zone, where it is currently undergoing remarkable transformations in terms of its society, economy, and information technology. Within this complicated context, the issue of the short lifespan of housing in this region remains a serious problem.

This phenomenon first emerged during the rapid development era that started in the 1990s: in order to free up enough land for new projects, a large number of housing units built in the 1970s and the 1980s were demolished [2]. This large-scale demolition and construction, coupled with rapid industrialization, led to a sudden and severe pollution issue. Consequently, scholars and practicing architects have increasingly viewed environmental sustainability and the concept of green buildings as key approaches to addressing the issue of housing longevity. This also resulted in most housing evaluation systems regarding sustainable issues being green building evaluation systems. However, green building concepts tend to emphasize physical environmental aspects, often resulting in a relatively one-sided, ecologically oriented approach to architectural design, construction, and usage. As early as two decades ago, Cole indicated that the processes of comparison and analysis of green buildings were typically limited to technical issues, and that the success of these assessment tools had dwarfed all other mechanisms for instilling environmental awareness [3].

Although extensive demolition and construction, along with the limitations of ecologically oriented design, have contributed to notable environmental crises, the key motivation behind the last wave of demolition lies in inappropriate design and poor adaptability of housing (from the urban to the interior scale) in the context of social and economic development [4]. Frequent demolition and reconstruction not only cause environmental problems but also negatively affect the psychological well-being of residents, including their sense of security and belonging.

Further speaking, in the present and future housing stock era, cities and other regions are unlikely to return to large-scale demolition and construction. Instead, the uncertainties associated with social, cultural, and economic changes are expected to intensify, making the demand for adaptable housing and the consideration of residents’ desires and behaviors highly crucial. At the macrolevel, the PRD is undergoing industrial transformation and deeper integration within the Guangdong–Hong Kong–Macao Greater Bay Area. More importantly, at the microlevel, the focus of the housing industry is shifting towards meeting the highly diverse and complex needs of residents and other stakeholders in the stock era. The PRD is one of the few domestic areas in China in which the population is still increasing [5]. However, this growth is accompanied by notable challenges in terms of living space, driven by changes in family structure, an aging society, the rise of the information industry, and mental health issues, which are exacerbated by phenomena such as “villages within cities” and the “post-pandemic era”, and so forth.

Therefore, the research target of this paper is to reject the purely technocratic, ecologically reductive evaluation models. A local long-lasting housing evaluation system, focused on residents’ social and psychological experiences and paying attention to the embedded cultural, technical, and temporal dynamics that shape housing resilience, remains necessary to be developed.

2. Research Methods and Theoretical Framework

2.1. Research Methods

In order to achieve the research target, the research methods of this paper were employed in a systematic framework (Figure 1). A classic definition of long-term housing refers to an architectural typology designed with spatial adaptability and functional flexibility to dynamically accommodate occupants’ evolving needs over their lifecycles, thereby extending the building’s functional lifespan beyond conventional housing.

An initial theoretical study was conducted, aiming to accurately address the contradictions and urgent needs of the local housing sector. This analysis focused on the discussion of long-lasting issues regarding the fundamental interrelation between human behavior and the built environment. Then preliminary long-lasting factors were identified. Later, an appropriate theoretical framework was introduced. It can organize these factors in a temporal dimension by levels, with different lifespan targets of physical condition and activity frequency of participants.

Based on the combined methodological framework, a case study will be carried out first. Since most contemporary housing types in the focused area have a relatively short lifespan, local traditional housing was focused on. These cases were considered long-lasting because they demonstrated stable environmental conditions, including a cultural atmosphere, spaces encouraging communication, and a well-established sense of belonging and safety. Green building evaluation systems were examined, focusing on their contemporary evaluation methods and selected relevant indexes. These two lines of study jointly aimed to identify highly valuable longevity factors and effective evaluation methods.

Two rounds of expert consultation and data analysis were conducted. This research step had been approved by “the Specialized Committee on Scientific Ethics of Guangzhou Academy of Fine Arts”, with an informed consent form provided for the participants when the research was conducted. In the first round, the basic indicator system generated from previous studies was sent to the targeted experts. And a roundtable discussion was organized together with a voting procedure later. It finally identified locally relevant longevity indexes and initial evaluation methods.

The second round employed a questionnaire study and data analysis to determine the weighting value of each index. The analytical weighting methods mainly included the Delphi method, the subjective experience method, multivariate analysis (MVA), and the analytic hierarchy process (AHP). The first two methods cannot fully reflect all interviewees’ suggestions and lack scientific validation, especially when confirming the weighting values. MVA is basically a quantitative method that cannot comprehensively evaluate complex situations. Moreover, the lack of relevant domestic data and instructions indicates that the assessment of long-lasting issues should rely heavily on experts’ experiences. Hence, the AHP was introduced in this study. This process is a structured technique for organizing and analyzing complex decisions, based on mathematics and psychology. By combining expert experience with rational analysis and directly comparing pairs of indexes, the AHP notably reduces uncertainties during the comparison process.

2.2. Preliminary Longevity Factors

Concepts and hypotheses such as obsolescence, typology, open building, and the more recent research and practices from Dietmar Eberle, to varying degrees, emphasized the shortcomings of green building approaches in addressing essential human needs [6,7].

Since the 1970s, studies on obsolescence have sought to further explore the underlying causes of building obsolescence and look for efficient ways to avoid it. Regarding the causes, the most widely acknowledged and applied distinction is related to material processes and behavioral factors, as well as human actions and their interactions.

As early as the 1960s, the open building (OB) theory was proposed by N. J. Habraken, which aimed to accommodate the diverse requirements of residents and long-term uncertainties [8]. In his book Supports: An Alternative to Mass Housing, Habraken first articulated the core principles of OB. He argued that housing must always recognize two domains of action: the action of the community and that of the individual inhabitant [9]. When the inhabitant is excluded, the result is uniformity and rigidity; when only the individual takes action, the result may be chaos and conflict. This emphasis on a necessary balance of controls had far-reaching implications for all parties in the housing process, including architects [10]. OB theory clearly states the uncertainty of human requirements and the dynamic relationships among various stakeholders.

Compared with the obsolescence and OB theories, Aldo Rossi paid more attention to cultural and historical urban issues. In his book The Architecture of the City, which was published in 1966, Rossi sought to develop an understanding of the city beyond purely functionalism by reviving the idea of typology as a means of interpreting buildings and clusters at a larger scale. He emphasized the importance of collective memory in shaping the identity of cities, highlighting the role of monuments and a sense of place as expressions of physical reality and historical continuity [11].

These themes are also the central work of Eberle, including the significance of public spaces and collective memory, the atmosphere enclosed by facade systems within specific cultural contexts, local construction measures, and so forth. But different from Rossi, Eberle adopted a more holistic view that went beyond the traditional architect’s role to encompass traditional longevity issues. As a contemporary architect, he also actively integrated economically efficient and ecologically friendly design strategies, particularly through passive design approaches.

Overall, the international debate on buildings’ longevity, including flexibility, compatibility with past and future needs, cultural representation, climate adaptability, and economic sustainability, is often addressed through architectural measures and comprises many overlapping themes. However, these studies also have certain limitations, such as a lack of consideration for urban patterns (in high-density contexts), connectivity, and sufficient local attempts. Thus, the preliminary longevity indicators identified for further study are listed as follows:

• The public space system and quality of outdoor spaces;
• The density and land use (urban morphological quality);
• Aesthetic quality and sociocultural value;
• Flexibility and adaptability;
• Longevity and durability (local construction methods and material usage);
• Passive design strategy.

All these initial longevity factors have direct and indirect impacts on the mental well-being of residents. At the urban level, public spatial quality, density, and sociocultural value are closely linked to a sense of community belonging and a perception of spatial quality. At the building and interior levels, flexibility, longevity, and passive design strategies are strongly associated with the feelings of comfort, safety, and atmosphere.

2.3. Measurement Framework: Longevity Issues in the Framework of Time, People, and Levels: Constructing the Basic Structure of the Index System

Although these long-term indicators show a tendency closely related to residents’ needs and psychology, their behaviors in various community spaces are not individual but socialized activities. As Granovetter pointed out, individual behavior is deeply embedded in specific social relationship networks, and social trust, cultural norms, and informal relationship networks can significantly influence decision-making and cooperation patterns [12]. He also revealed the significant role of weak ties and the fundamental function of informal relationship networks. However, the relational embeddedness focused more on how social networks affect institutional performance. Under the complex situation of urban resource allocation, restraining the behaviors of participants and coordinating the distribution of different interests are more important. The Institutional Analysis and Development (IAD) Framework can provide ideas, emphasizing how institutions influence collective decision-making through the action arena. Incentives seldom exist in isolation, and the policy context in which they exist may be as important as the incentive itself in affecting behavior. Incentives can only function under appropriate institutional conditions [13]. Hence, Gasometer’s “informal network of relationships” can also be incorporated into the variables of institutional design.

In fact, the previous discussion of the open building concept has a similar idea. In order to balance the two domains of action, Habraken divided residential buildings into “support” and “detachable unit” and eventually proposed the “level” concept, which tried to distinguish the selection, control, and responsibility throughout the construction to usage stages (Figure 2). This concept refers to distinct levels of intervention within the built environment, such as those represented by “tissue”, “support”, and “infill”, or by the urban design and architecture. Then it can ensure the trial-and-error within a prescribed range that has the least impact on other groups. Similar ideas can be found in the works of Stewart Brand and Eberle [14]. Brand proposed that a building system comprises several sharing layers, based on the fact that different rates of change (lifespan) of its components (Figure 2). Meanwhile, Eberle emphasized that design quality depends on the effective coordination of different stakeholder groups and the handling of different information. Thus, a building should be conceptually and technologically separated into several systems according to the material lifecycles, spatial and structural hierarchies, and social responsibilities [15,16]. Lifecycles, including physical and service aspects, are a key issue. Organizing a building in such a way can make change easy and provide a great deal of flexibility when dealing with separate systems.

Both Habraken and Eberle divided the built environment into five levels. However, in the historical context of the PRD, the structure of local housing is closely connected to the urban fabric and can also be clearly separated from the facade system. The interior level is still treated as an independent level. Therefore, this research adopts a four-level framework. In combination with the preliminary longevity factors, a two-dimensional structural model is proposed, aiming to develop an assessment tool specifically for long-lasting housing in the PRD. The four levels are arranged vertically, each corresponding to different lifespan targets, while appropriate longevity factors are placed horizontally within each level (Figure 3).

At level one, a scale with comprehensive factors including social, environmental, and economical issues would be categorized respectively, and the three issues can affect the most important issue from classic long-lasting theories—urban form and public space. Some indicators, such as density and land use, are closely related to policies and regulations. They should be dominantly led by the governments or related departments, with residents’ opinions as a supplement. At levels two and three, social and environmental aspects continue to provide significant influences. Although participating groups normally have a smaller number, their interests must be coordinated through a more balanced organization model. At the same time, as discussed above, the longevity and flexibility would be independently treated as crucial factors corresponding to stability and adaptability. They are also influenced together by social, environmental, and economic aspects. Level four has similar issues with levels two and three, but conflicts and demands mainly exist among some households and within themselves; their autonomy rights need to be maximized. Longevity is not that important since life span at this level is approximately 20 years [18].

To sum up, the first-set indicator system (version one) emphasizes people and time issues. First, the level strategy provides a framework for different parties to participate in the spatial activities at different scales and stages, and suggests different frequencies. It is conducive to coordinating the interests among different parties, especially in a large-scale space, and ultimately meets the maximization of individual needs in a small-scale space. It also integrates the estimated lifespan of physical materials usually used at different scales. Second, among these longevity factors, sociocultural value (the inherent common demands of neighborhood residents) and flexibility degree (the inherent demands of individuals) are emphasized. But technical indicators, such as durability and ecology, are also taken into account.

3. Case Study and Documentary Study

Although a basic measurement framework has been established, a number of issues remain: First, the initial set of primary indexes remains generic and lacks detailed secondary-level indicators. Second, the importance of these indexes, as well as any potentially missing aspects, needs to be thoroughly reviewed and validated. Third, the local features and experiences from the study area must be carefully considered to ensure that the evaluation system is contextually adaptable. And last, relevant indexes and evaluation methods from existing green building evaluation systems can still offer valuable references and should be examined further.

3.1. Longevity Issues in Case Study

Historically, buildings in China have a distinct identity, usually designed and constructed by an open and “level”-based division system. In particular, their modular structure system can be independently organized as a foundation, façade, and roof system [19,20]. Traditional housing in the PRD region clearly follows this principle, while also incorporating locally adaptable strategies.

To some extent, the courtyard house can be considered as the basic typology constituting the human settlement clusters in the Pearl River Delta; its basic unit is Jian (间), and Jian can constitute individual housing (屋). There are many types of individual folk housing, like three-Jian, five-Jian, or seven-Jian. By organizing around a patio, individual housings constitute a courtyard (院落). Different from modern functionalism architectural concepts, the spaces formed by such prototypes can not only fulfill the basic function of a family, but also be treated as the long-lasting elements to adapt to frequent changes through time.

The criterion for case selection was a lifespan of more than 100 years. Clusters and buildings from the “National Cultural Heritage Conservation List” were prioritized because they are the most well-preserved and historically important examples, designed with strong identities that have endured over time. The selected cases also cover a wide range of construction types since they all evolved from a similar prototype, not only limited to residential usage but also including residential function mixed with other functions. Based on this principle, five representative cases were selected: (1) Shawan Town, a typical local town that integrates human intelligence with natural resources, boasting a history of more than 800 years; (2) Donghuali Street, the most completely preserved historic street, containing all the typical housing typologies in the PRD; (3) Chen Clan Academy, one of the biggest clan temples (including residential functions) in Guangdong Province, which is representative of local folk architectural decoration and arts; (4) Shangxiajiu, an important colonnade street, which is closely linked to commercial activity, daily activity and living; and (5) Yuyin Garden, a local private garden designated as a national relic protection site, with high cultural and research value. Eventually, at least two cases were analyzed at each level. Some cases covered several levels, which aided in better understanding their longevity in a holistic and comprehensive manner (Figure 4).

At the infrastructure and public space level, study of the two cases revealed that traditional clusters are the outcome of social, environmental, and economic considerations, which is in accordance with the contemporary definition and concept of sustainability. Social and cultural influences evidently play an important role in their formation and longevity. The hierarchal infrastructure system, which is the prediction and foundation of human settlements in the PRD, is strongly defined by a deep-rooted clan system [21] (Figure 5). The public space system should be emphasized as being characterized by high connectivity, a human-centered scale, and abundant details [22]. This can help strengthen the sense of belonging. As the clan system collapsed, environmental and economic impacts became stronger later; however, the system still retains sufficient flexibility to adapt to the uncertainties in the spatial and temporal dimensions. Environmental constraints influence the detailed arrangements and morphology, such as the orientation and public space dimensions, while economic factors have a greater impact on the formation of cluster size and density.

At the load-bearing structure and staircase level, the structure of vernacular housing from the case study plays a more important role in daily life and is more interconnected with other levels, compared to contemporary concrete frame systems. Although technological constraints and economic concerns have affected the dimensions and investment in basic living units (such as spans of interior space and use of high-quality materials), cultural influences are notably pronounced. Factors such as clan structure, family size, social status, and education level are the key determinants of the entire residential volume and spatial quality [23]. For instance, the use of hierarchical dimension systems, symmetric arrangements, and flexible construction methods, as well as the integration of passive environmental strategies and folk art, are all important elements that contribute to the longevity of the structure (Figure 6 and Figure 7). These efforts enhance the spatial quality perception and leave a lasting, unique impression.

From the case study at the level of the facade and service core, since high-density and human-scale dimensions are determined by the design of the upper level, the facade becomes an important interface between the public and private realms and is responsible for more cultural and public obligations. Elements such as windows and doors serve as ideal integrations of climate-responsive strategies, cultural representation, and user convenience (Figure 8). These features help balance residents’ needs for safety and openness, while also enhancing their sense of pride. From an environmental perspective, such elements notably contribute to the quality of the indoor environment. For example, hollow-carved windows and screens are specifically designed to enhance ventilation (Figure 9). Additionally, the high west–east-oriented walls facing narrow lanes are strategically designed to reduce solar radiation and mitigate the impact of frequent typhoons. Simultaneously, they provide benefits in terms of fire protection and residents’ privacy [26].

Ongoing changeability and adaptability are evident at the interior level in both cases. In the larger-scale case of the Chen Clan Academy, which is constantly serving various functions, the frequent alternation of public and private functions demonstrates that the column-based structural system allows for a high degree of spatial flexibility. Even in the wing-rooms, the use of a precast orbit strategically supports variable and adaptable spatial divisions. In the smaller-scale case of Donghuali, flexible partitioning structures—such as light-weight partitions, gate masks, hanging fascia, and traditional frames—facilitate temporary divisions. Craft technology is important at this level because it is closely related to residents’ daily lives. Partition walls and furniture possess strong cultural value and are mainly made of wood, reflecting localized, environmentally friendly practices (Figure 10 and Figure 11). The rich and integrated atmosphere created by these elements is a key factor in promoting psychological comfort within interior spaces.

Through the case study, an initial version of the two-set index system was developed, revealing many indicators that are specific to the region. However, most of these indicators are directly derived from the traditional case studies, which means that further research is necessary to supplement and validate this preliminary framework.

3.2. Longevity Issues from Documentary Study

Based on the initial longevity issues identified through the case study, a holistic documentary study was conducted. The objective was to extract valuable longevity indicators from selected green building evaluation systems and to reference their applicable evaluation methods. This study also conducted a comparative analysis, aiming to identify which tools provide highly relevant details and which offer less. The selected indexes were categorized under the initial longevity issues, accompanied by a comparative analysis of their weighting values.

GB/T was first selected because it serves as a key domestic reference system [29]. It was formulated under China’s Ministry of Housing and Urban–Rural Development (MOHURD) to advance green building development, conserve resources, and enhance ecological sustainability. Its early version mainly took LEED from the U.S. as a reference. The latest one: assessment standard for green building (GB/T-2019) [29] had been developed into a more comprehensive system, including safety and durability, health and comfort, life convenience, resource conservation, and environment livability. For the score method, GB/T has four tiers of score, and consists of “prerequisite items” and “scored items”; only when the project fulfills all “prerequisite items” can it continue to be assessed.

CASBEE from Japan and DGNB from Germany were also included, as they both cover valuable and comprehensive issues that go beyond ecological sustainability. CASBEE is based on the Japanese situation and has been developed into a two-dimensional evaluation system. When evaluating a building and its surrounding area, CASBEE sets a hypothetical boundary to enclose the urban area. In doing so, it can evaluate the Built-Environment Efficiency (BEE) within the hypothetical boundary. Improvement in environmental quality and activities (referred to as “Quality”, or “Q”) within the enclosed space and reduction in negative environmental impact (referred to as “Load”, or “L”) on the area beyond the boundary lead to higher BEE values, thus a better rating [30].

In the case of Germany, which has high industrialization technology and standards, its green building assessment system (DGNB) is the latest one to be launched, but it has the most comprehensive considered aspects, including social, economic, environmental, and process management. It was developed by the German Sustainable Building Council and has clarified the sub-indexes of each target group for easier understanding and operation. The outcome would be presented as a pie diagram, which can directly indicate the satisfaction degree of each target group both individually and collectively [31].

Unlike GB/T, CASBEE, and DGNB are not single systems but rather families of tools, each comprising several independent systems that are tailored to different scales, building typologies, and specific contexts. This study initially examined the latest version of each system, and then focused on the most representative tool within each family.

By preliminarily comparing the case study and the three selected evaluation tools, several relevant longevity issues could be identified in their second-set index systems. The relatively low level of industrialization in China has led GB/T to focus primarily on the saving of water, land use, energy, and materials. But GB/T-2019 has expanded its scope to include durability, convenience, and livability. In contrast, CASBEE and DGNB have emphasized longevity issues since their early visions, addressing aspects such as public space, aesthetic quality, and flexibility. However, even their latest versions allocate a relatively small proportion of longevity concerns. DGNB assigns a total weighting value of 24.4%, while CASBEE has an average weighting value of around 30% (39% for quality and 22% for load). Additionally, CASBEE includes more factors focusing on durability, reflecting Japan’s location in a seismic zone (

Table 1. Overall view of longevity issues in three studied GB tools.

First Set of Longevity Indexes	GB/T-2019	CASBEE-BD (NC)	DGNB 2020 (International)
Public space system and quality of outdoor space		Q-3.2 Townscape and Landscape (12%)	Soc 1.6 Quality of outdoor space (2.1%)
			Pro 1.6 Procedure for urban and design planning (1.6%)
			Site 1.2 Influence on the district (1.1%)
Density and land use	6.1 Occupant convenience (prerequisite items) (8%)		Env 2.3 Land use (2.4%)
	6.2.1–6.2.5 Occupant convenience (scoring items) (4.1%) 7.2.1 Land saving (2%)		Site 1.3 Transport access (1.1%) Site 1.4 Access to amenities (1.7%)
Aesthetic quality and sociocultural value	9.2.2 Architectural design styles suitable for regional characteristics (2%)	Q-3.3 Local Characteristics and Outdoor Amenity (9%)	Site 1.2 Influence on the district (1.1%)
Flexibility and adaptability	4.2.6 Adaptability measures (1.8%)	Q-2.3 Flexibility and Adaptability (9%)	Eco 2.1 Flexibility and adaptability (7.5%)
			Tec 1.4 Use and integration of building technology (1.9%)
Longevity/durability	4.2.7–4.2.9 Longevity (2.9%)	Q-2.2 Durability and Reliability (9%)	Tec 1.3 Quality of the building envelope (2.5%)
Passive design strategy	7.1.1 General passive design strategy (0.8%) 7.2.4 Optimize thermal performance of building envelope (1.5%) 7.2.17 Recyclable materials (1.2%) 7.2.18 Green building materials (1.2%)	LR-1.2 Natural Energy Utilization (4%) LR-2.2 Reducing Usage of Non-Renewable Resources (18%)	Tec 1.3 Quality of the building envelop (2.5%) Tec 1.6 Ease of recovery and recycling (2.5%)
Total Score Percentage	25.5%	Q: 39%, LR: 22%	24.4%

).

On the other hand, the lack of an appropriate methodological framework for effectively organizing these elements across the temporal dimension in these evaluation systems is evident. They tend to have independent categories for the urban and interior levels, while many other categories, such as material, energy, and water savings, cover multiple levels but are treated as independent groups. This issue could worsen during the evaluation process. For instance, several criteria of GB/T only focus on outcomes. Renewable energy, for instance, is only graded when the renewable energy usage reaches the minimum requirement, without considering the different acquisition processes at different levels or in different local contexts. Furthermore, evaluation categories that are specific to certain levels often remain isolated. In the case of GB/T and DGNB, although site quality belongs to the first set of categories, it is still treated as a relatively independent evaluation item.

At level one, infrastructure and public space, the primary focus of the selected assessment tools is on protecting a site’s ecological conditions and preventing hazards and risks. Second, they emphasize optimizing the physical environmental performance or microclimate conditions of the site, such as enhancing ventilation, ensuring adequate lighting, and reducing noise. Although connectivity is mentioned, very few concerns are related to historical and sociocultural aspects. Despite being a critical issue at this level, the quality of the public space system is largely overlooked, not only in the selected assessment tools but also in many local architectural design projects. These issues are not new concepts or principles, but they have been overshadowed by more visible contemporary environmental problems. Consequently, psychological issues such as a sense of belonging are similarly neglected. As Eberle pointed out, the relationship between human behavior and the physical structure of a city can often be understood as the relation or switching between public and private spaces [15]. Furthermore, although economical concerns can be found in some relative indexes, such as land, energy, water, and material savings, their connection is mostly indirect. A direct explanation and comprehensive consideration of economic factors remain lacking. Only DGNB includes specific references to lifecycle costs and economic feasibility.

At level two, load-bearing structure and staircases, the assessment tools focus on the material selection, durability, and resource-saving strategies, as well as passive design methods such as structure allocation, orientation, and floor height. Security is another focus, mainly through architectural design and management controls, but this focus is limited to physical aspects. In contrast, the social and cultural interactions integrated with the structure system, which are evident in traditional case studies, are largely overlooked.

At level three, facade and service core, the selected tools pay more attention to the mechanical energy and water saving methods, as clearly reflected in the weighting distribution of the indexes. DGNB emphasizes the importance of architectural approaches (such as the envelope quality) to enhance indoor environmental quality, although this is based on the context of a temperate zone. However, a notable omission across all tools is the social considerations, especially the mental communication between interior and exterior spaces. As discussed at level one, the public space system is an important issue, and the facade plays a key role in shaping public spaces. It serves as a carrier of public art and collective interests, but only DGNB includes a few factors relating to this aspect, and even then, they have minimal weighting. Flexibility, while acknowledged in highly industrialized countries, is mainly interpreted in terms of deconstruction, disassembly, and the separation design strategy of living and device spaces. As the only evaluation system from a developing country, GB/T-2019 explicitly recognizes the need for prefabricated facade components and integrated kitchen systems.

At level four, interior and function layout, the key issue of these evaluation systems is the comfort of the indoor physical environment. Some psychological concerns and social issues are also present in CASBEE and DGNB, where aspects such as the basic spatial quality, integrated atmosphere, and interior adaptability are emphasized.

Compared with traditional buildings, process control and operation management are relatively new concepts that can contribute to organizing the projects efficiently from the concept to the operation stage. However, this approach is a passive evaluation that lacks active guidance and a structured framework to manage longevity issues and stakeholder behaviors across different levels within the temporal dimension. Deconstruction action often occurs earlier than the expected end of a building’s service life, typically due to complex and uncertain factors, such as social, economic, and environmental aspects. In order to ensure longevity, key issues must be clearly determined and systematically organized in a flexible and adaptive manner. Overall, the differences and similarities between traditional housing in the PRD and contemporary green building assessment tools can be outlined as shown in the figure below (Figure 12).

4. Establishment of the Index System

4.1. Establishment of the Final Index Structure

Through comprehensive and comparative studies, further analyses were conducted to finalize the first and second sets of indexes. The population of cities in less developed countries doubled between 1990 and 2015—for example, the time period covered in this Atlas—and their urban extents increased on average by a factor of 3.5. Several cities in the Pearl River Delta (Guangzhou, Kaiping, and Shenzhen) all showed the same tendency [1]. Although the per capita living area increased significantly, the urban area expanded too fast. Accordingly, the indicators related to the urban compactness, such as saturation, openness index, and cohesion, were significantly reduced. Therefore, in the discussion of local indicators at level one, appropriate density and high accessibility must be specifically emphasized. These characters are also obvious in the conclusions of traditional local case studies, and they have demonstrated the human-scale perceptibility and high efficiency.

Furthermore, Meng Dancheng et al. [32] from Tongji University conducted a study of how community and housing characteristics affect health, residential satisfaction, housing choices, and their urgent needs. Based on the feedback of questionnaires from over 7000 residents, the most important indicators can be summarized, which are also emphasized in the above studies. Such as appropriate density (300–1000 households), diverse services, sufficient open and green spaces, facade system with high connection to outer space, belonging, security, sufficiency interior space (flexibility), and the rational layout of sanitary and pipe space, etc.

In order to determine indexes that are specific to the Pearl River Delta, the expert panel method was then introduced, accompanied by the suggested evaluation method category for each index. It is a collective prediction method where experts freely exchange views in a structured setting, sparking intellectual collisions to refine conclusions. It can integrate group wisdom through face-to-face debates, rapidly forming consensus via brainstorming, adversarial discussions, or hybrid procedures.

First, the basic indicator system generated in previous studies was sent to the selected experts, and summarized index systems from the case and documentary studies were also provided for their reference. All of the experts held senior titles. These experts included developers, designers, academics, and engineers. Two senior project managers were from China Poly Group, two senior architects were from Guangzhou Provincial Academy of Building Research Group, three professors were from South China University of Technology, and two senior engineers were from Huizhou Municipal Engineering Co., Ltd. After they had reviewed and processed the materials, a roundtable discussion involving the nine experts was organized. During the discussion, most of the experts agreed with the basic index system for each level, while also suggesting several indexes that were specific to this region. In order to finally confirm the local indexes, a vote was carried out. Indexes with more than half of the votes were chosen as the final local indexes. Eventually, six indexes were developed as follows, all of which could cover social, economic, and environmental aspects in the local context, while some included multiple aspects:

• Compact urban form with variety and order;
• Spatial efficiency;
• Highly damp-proof structure with long life span;
• Delicate courtyard/atrium space and high-quality vertical access;
• Local feature facade integrated with multi-cultural elements;
• Passive facade strategy for shading and ventilation.

These local indexes belong to the second set of index systems. They can be used to evaluate a building’s performance within the local context more elaborately and accurately, and they were subsequently placed under the most suitable first set of indexes. In addition, to clearly establish the entire structure, the levels and indexes were named using a combination of English characters and Arabic numbers. The four levels were named with English characters from A to D. Within each level, the first index of the first set in level A was named as A1, the first index of the second-set index (A1) was named as A1.1, and so on by this analogy.

Subsequently, a two-set index system based on level division was established, including specific indicators tailored for this region and the preferred measurement methods for each index. Notably, all indexes in the first set across the four levels have direct or indirect effects on human mental considerations. Some second-set indexes, such as community belonging, spatial quality perception, psychological comfortability, and sense of safety, specifically address mental health issues. The evaluation methods are diverse, including formula calculations, interview scores, compliance with national standards, and assessments through drawings and diagrams (Figure 13).

On the other hand, among the studied green building assessment tools, the points score is the primary method for evaluating each index. However, this approach has a clear limitation: it cannot precisely evaluate the performance degree, especially when assessing the qualitative performance of long-lasting buildings. Hence, a graded evaluation standard was suggested, aiming to comprehensively assess the performance level of each index.

The evaluation methods of each indicator have been preliminarily studied. However, due to avoiding excessive complexity and interminability, they will not be elaborated on in detail in this paper. Figure 13 briefly presents the diverse evaluation methods of each indicator. For some indicators that can be directly quantified, such as spatial efficiency, the formula of CASBEE will be referred to apply (total usable floor area/gross floor area). For some indicators, such as the neutral structural space, it is necessary to apply the formula (wall length/area ratio) together with the evaluation of the design drawings. As for the psychological and social dimensions, due to their complexity, some indicators can be regarded as a sub-research topic. Usually, questionnaires and data analysis are needed. Although neighborhood attachment is not listed in this system, it mainly includes belonging and safety aspects, which are both important indicators of this system. According to Savage and Blockland, belonging is a dynamic inter-construction. It should be emphasized on social relationships, spatial practice, and power aspects. And it is suggested to be evaluated through a combined method, including sense of community index (SCI), behavioral mapping, physical spatial chrematistic (by GIS), social network and function, and policy intervention. Psychology safety can be evaluated through questionnaire surveys based on the Maslow Security–Insecurity Inventory. While high-quality and durable materials can provide the physical prerequisites for psychological safety. These materials are suggested to be applied during the design stage and monitored during the usage stage. Psychological comfortability can be mainly evaluated through the General Comfort Questionnaire (GCQ), which was developed by Kolcaba. In addition, some other indicators, such as diverse amenities and a highly connective public space system, also have a strong relation to the issues of neighborhood attachment.

4.2. Weighting Method and Analysis

4.2.1. Questionnaire Design

Determining the importance degree of each index within the structure is the next crucial step in establishing the final evaluation system, by employing a questionnaire study and data analysis. The questionnaire was first divided into four large groups corresponding to the four levels. Each group was then further divided into several subgroups based on the two-set index system within each level. For instance, questionnaires A, B, C, and D focused exclusively on the first set of indexes for each level. Questionnaire A1 only covered the second set of indexes belonging to the first-set index of A1, and so on by analogy. Additionally, the degree of importance was measured using a five-level scale.

Aiming to comprehensively evaluate the importance of interrelations of indexes, the questionnaire interview attempted to cover a wider range of professional groups and regions. Profession groups included developers, consultants (consultant company and academic), designers (architects, planners, and landscape designers), and contractors, while regions included the areas outside China, the PRD, Hong Kong, and other regions in China (Figure 14). This strategy helped prevent any single interest or preference from being dominated by a specific group of people. Even within some groups, such as designers, including diverse types of firms and organizations, allowed for different angles and ensured comprehensiveness. The main professions that were targeted for interviews were consultants and designers, primarily from the PRD, to ensure the professional expertise and local relevance.

Eventually, over 80 questionnaires were collected, mostly from professionals with more than 10 years of experience, intermediate professional titles, and bachelor’s degrees. In order to effectively collect important information, three principles were applied to limit the number of questionnaires. The first was to balance the regions that they came from; the second was to ensure that all relevant professional roles were included; and the third was to select experts with more extensive experience. Therefore, 40 questionnaires were selected for further analysis.

4.2.2. Data Analysis and Weighting Confirmation

In the AHP method, experts provide their opinions on the relative importance of different levels within an index system; based on this, the weight values of various indicators can be calculated. A critical challenge is how to synthesize these experts’ opinions. Normally, two main methods are used: the arithmetic average method and the weighted average method. The arithmetic average method assumes that the experts are completely independent and that their opinions are totally irrelevant. The assessment of the relative importance of different index weights may be similar and possibly nearly consistent. Accordingly, experts with similar opinions should be assigned similar weight coefficients, indicating that a larger group of experts would receive a greater weight coefficient, while a smaller group would have less weight.

Therefore, the next step is to identify a method for classifying the experts’ opinions, whose influencing factors are different and complex. Experts’ opinions on the relative importance of certain indexes may be similar for some, while they will significantly differ for others. Cluster analysis, a multivariate statistical method that is used to classify samples based on their individual characteristics, is a suitable approach for classifying experts in this situation. Cluster analysis is also called segmentation, as it can categorize data into groups or clusters. These clusters are formed in such a way that objects in the same cluster are very similar, while objects in different clusters are very distinct. The measure of similarity depends on different application situations. Cluster analysis generally includes two main methods: k-means clustering and hierarchical clustering. Unlike hierarchical clustering, k-means clustering operates directly on actual observations (rather than the larger set of dissimilarity measures) and creates a single level of clusters. Because the number of categories was unknown and needed to be explored, hierarchical clustering was more suitable for the current study. Hierarchical clustering groups data across various scales by creating a cluster tree or dendrogram. This tree is not a single set of clusters, but rather a multilevel hierarchal system, where several clusters at one level can join to form a new cluster at the next level. Therefore, an appropriate level or scale of clustering is crucial.

Based on the AHP method, the weighted values of the indexes from different sets were first calculated using Modern Comprehensive Evaluation. The calculation results are shown in Appendix A. In the second step, based on hierarchical clustering, the weighted values of different sets of indexes were classified using the SPSS 30 software. In detail, using Agglomeration Schedule for the statistics, Furthest Neighbor for clustering, Squared Euclidean Distance for interval measurement, and dendrogram for the final diagram representation. A classification standard threshold of five was to avoid having too many or too few categories. For instance, the dendrogram of the first set of indexes for the infrastructure and public space level is shown in Figure 15 below.

Finally, the weighting of each index in the entire system is summarized in the following (Figure 16). According to the results, the weighting values of locally specific indexes are relatively high (normally higher than 0.5, regardless of the number of indexes within the same set). This finding applies not only to the second-set indexes themselves but also to the first-set indexes to which they belong.

5. Conclusions and Discussion

5.1. Conclusions

This study begins with the discussion of a new challenge in China’s most advanced area, the Pearl River Delta, which is facing short housing lifespans under complex social changes, and urgently needs to develop an appropriate evaluation method for long-term housing.

Long-term housing not only achieves environmental sustainability but also promotes residents’ mental health and well-being by providing a stable living environment. Aiming to systematically, comprehensively, and locally develop such a system, theoretical research, local case studies, and documentary analysis were first conducted. These efforts aimed to identify initial longevity issues and to establish a preliminary first-set and second-set index system, respectively. Through a roundtable discussion with local experts, the final version of the two-set index system, including specific indicators tailored to the study area, was established. The evaluation method for each index was categorized. Furthermore, a questionnaire interview involving a wide range of experts was conducted. Combined with data analysis, this process ultimately confirmed the weighting value of each index.

Compared with other evaluation systems, this framework explicitly incorporates multi-scalar temporality, sense of belonging, and adaptability into a structured evaluation system that responds to the specific material, social, and participation realities of the Pearl River Delta, thereby avoiding universalist models that fail to recognize contextually rooted patterns of embeddedness. It is also a flexible system since it is based on a multilevel strategy. The system allows for the independent evaluation of projects at different levels, each with its own targeted lifespan. For instance, when a project is fully evaluated, if one level receives a high final assessment grade, then it is only likely to reach the targeted lifespan for this level. It can also serve as an ideal prediction for lower levels or future use within the same level. On the other hand, the system can also independently evaluate one level of a project, providing a focused outcome specific to that level.

5.2. Limitations and Discussion

Physical space and social participation are two major aspects that can be effectively organized through a level strategy; the higher level can accommodate more participants, including government, developers, designers, builders, resident groups, etc. The lower level can provide residents with more operational spaces and usage rights. In this process, designers simultaneously play the role of coordinators.

To be more concise and relatively easy to operate, the indicators of this evaluation system cannot cover all aspects of society, economy, and environment, but only extract the most important parts. At the same time, it is an initial ideal model based on the studies from respected urban researchers and architects, and is suggested to be applied in the concept and design stages. There is no further discussion on the detailed governance process and the operation process. The purpose of this paper is to build a reasonable basic framework and methodology for the front-end of the design and planning stage. It can provide a prerequisite for social governance and the operation stage, while transcending the limitations of geography and different institutional governance models, such as dominance of government, dominance of the technical management department, and collaboration of multiple groups.

However, in order to cover the entire life cycle of a project and to be more comprehensively concerned, further study is needed. The combination of flexible spatial strategies and resilient governance will have great potential for further analysis, which can better balance the contradictions of multi-group collaborative models.

Moreover, this research is a preliminary study; detailed evaluation methods for each index still require further analysis and elaboration. Although it is a relatively simple system compared with CASBEE and DGNB, it still needs a training course and some professional groups to instruct the users when applying to the project evaluation, especially for those indicators that need comprehensive measurement methods. This process will involve mutual monitoring and evaluation, which can not only improve the evaluation system but also help identify the problems and insufficient knowledge of these representative projects within the region.