A Fuzzy Decision-Making Approach to the Health Assessment and Optimization of Architecture-Dominated Outdoor Spaces in High-Density Urban Environments

Abstract

As urbanization rapidly increases, the design of outdoor spaces in high-density urban environments has become crucial for promoting public health. This study investigates the health impacts of architecture-dominated outdoor spaces, particularly focusing on small decentralized spaces around buildings. The research aims to develop a comprehensive health evaluation framework that quantifies the influence of various design factors such as comfort, safety, diversity, and ecology. Using a fuzzy Delphi method (FDM) and the Analytic Hierarchy Process (AHP), 31 key design indicators are identified and weighted based on expert opinions. A multi-level fuzzy comprehensive evaluation model is then applied to assess the outdoor space of Wuhan Citizen’s Home. The results show that the space performs well in promoting health, particularly in comfort, safety, and ecological design. However, there are areas for improvement, such as enhancing cultural representation and increasing the frequency of health-promoting activities. This study concludes that the proposed evaluation framework can provide valuable insights for optimizing the design of outdoor public spaces, supporting healthier urban environments, and improving residents’ physical and mental well-being, offering valuable reference for future urban space design.

1. Introduction

The World Health Organization defines health as “a state of complete physical, mental, and social well-being, beyond the mere absence of disease or infirmity [1]”. Amid rapid urbanization, the interplay between the built environment and human health has garnered significant research attention. Studies consistently demonstrate that the built environment influences residents’ physical and mental well-being [2,3,4], with its components broadly categorized into outdoor and indoor architectural spaces [5]. Outdoor spaces range from large, centralized public areas—such as urban green spaces, parks, and squares—to smaller, decentralized, building-oriented zones [6], while indoor spaces encompass residential, work, and leisure settings [7].

Global urbanization trends, as outlined in the United Nations’ “World Urbanization Prospects: The 2018 Revision”, project a 68% urbanization rate by 2050, with notable growth in Asia and Africa [8]. East Asian cities, shaped by scarce land and high population density, exemplify this shift, often prioritizing dense building layouts over expansive public spaces [9]. Consequently, urban green spaces in these regions tend to be small, fragmented areas surrounding buildings, contrasting with the larger, contiguous parks typical of Western cities. Despite their prevalence, these scattered spaces often lack the scale and connectivity needed to fully support community interaction and health benefits. This underscores the growing role of outdoor public spaces in fostering well-being, particularly in high-density urban contexts [10,11,12,13].

While the health impacts of indoor environments and large outdoor spaces are well-documented [14,15,16,17,18,19], systematic research on small, building-oriented outdoor spaces in dense urban settings remains limited. In East Asia, where such spaces are integral to daily life, their potential effects on physical and mental health are underexplored. This gap hampers both our understanding of urban environment-health dynamics and the development of targeted design strategies. Existing evaluation frameworks, critical for quantifying spatial impacts on health [20], predominantly focus on large outdoor areas [21,22,23] or indoor settings [24], leaving a void in tools tailored to small, decentralized outdoor spaces.

To address this, this study develops and validates a health evaluation framework for building-oriented outdoor spaces in high-density urban environments. The proposed framework targets small-scale public areas outside urban parks, streets, or waterfronts—such as those defined by single- or multiple-building complexes—and aims to enhance our understanding of their role in improving residents’ quality of life. Specifically, this research pursues three objectives:

RQ1. 

Develop a health-oriented evaluation index system using theoretical insights and field surveys, applying fuzzy set theory to establish standards and weights for a comprehensive evaluation model.

RQ2. 

Validate the framework by assessing typical outdoor spaces in Wuhan using fuzzy comprehensive evaluation.

This study unfolds in three stages (Figure 1). First, a literature review and the fuzzy Delphi method (FDM) identified 31 health-related design factors from an initial pool of 53. Second, the Analytic Hierarchy Process (AHP) determined factor weights and rankings. Third, a multi-level fuzzy comprehensive evaluation model was constructed and tested on a Wuhan case study to confirm its applicability. This paper proceeds as follows: Section 2 reviews the literature in order to extract key design factors; Section 3 details the methodology; Section 4 presents the evaluation model and its application; Section 5 discusses results; and Section 6 concludes with limitations and future directions.

2. Literature Review

2.1. Relevant Evaluation System

The World Health Organization’s “Healthy Cities Standard Evaluation”, launched in Barcelona in 1893, marked the inception of healthy community standards. This framework evaluated health across seven dimensions—physical environment, ecosystem, community equity, governance, material needs, community services, and residents’ health status—serving as a foundational reference for global standards. Notable advancements emerged in the United States, Canada, Japan, and Australia. The U.S. WELL Community Standard assesses health through ten factors, including air, water, light, fitness, and spirit, emphasizing environmental design’s role in well-being. Canada’s framework, initiated in the 1990s and enhanced by the 2014 “Healthy Community Guidelines for Halton Region”, integrates physical and social environments, governance, and sustainable design principles. Australia’s “Healthy Community Planning” (2003) and “Victoria Public Health Guidelines” (2011) evolved from basic community transformation to promoting well-being via environmental and social strategies. Japan’s 2013 “Healthy Community Check Overview” prioritizes safety and smart infrastructure, as exemplified by Fujisawa’s age-friendly community, which integrates elderly care and medical services to foster health-oriented lifestyles.

2.2. Index Selection

This study synthesizes international health-related building evaluation standards to identify design factors influencing the healthiness of outdoor architectural spaces, categorized into five dimensions: comfort, diversity, safety, ecology, and humanism (

Table 1. Summary of clusters and their constituting factors with evaluation standards.

Evaluation Standards	Cluster	Factor
WELL v2™, WELL Community Standard [25]	Comfort	Air
		Water
		Light
		Thermal Comfort
		Sound
	Diversity	Movement
		Community
	Safety	Materials
	Humanities	Mind
	Others	Nourishment
		Innovation
Reference Guide for the Fitwel Certification System: Community [26]	Comfort	Community Open Space: Design
		Outdoor Environment
	Diversity	Community Composition and Location
		Community Open Space: Design
	Safety	Emergency Preparedness
	Ecological	Community Open Space: Design
	Humanities	Community Open Space: Management
		Community Assets
		Community Resiliency
	Others	Site Access
		Building Certification
		Healthy Food Environment
Building Healthy Places Toolkit [27]	Comfort	Ban Smoking
		Use Materials and Products that Support Healthy Air Quality
		Facilitate Proper Ventilation and Airflow
		Maximize Lighting Quality
		Minimize Noise Pollution
	Diversity	Incorporate a Mix of Land Uses
		Design Well-connected Street Networks at the Human Scale
		Provide High-quality Spaces for Multi-generational Play and Recreation
		Build Play Spaces for Children
	Ecological	Increase Access to Nature
	Humanities	Facilitate Social Engagement
	Others	Adopt Pet-friendly Policies
		Promote healthy food retail

). These factors, derived from the literature and established frameworks, inform the health evaluation indices for this research.

2.2.1. Comfort

Comfort, a pivotal quality metric for outdoor spaces, encompasses physiological, psychological, and sensory responses to environmental conditions such as air quality [28,29], lighting [30], thermal comfort [31], water features, and noise levels [32,33]. You You Peng et al. [34] argue that comfort extends beyond sensory perception, reflecting a dynamic interplay between environmental factors and individual needs, crucial for health-oriented design.

2.2.2. Diversity

Diversity gauges a space’s adaptability and inclusiveness, enabling it to accommodate diverse user groups and activities [35]. In high-density urban contexts, multifunctional design is vital due to limited land resources [36]. Research shows moderate physical activities—e.g., walking, running, and cycling—enhance cardiopulmonary function, mental health, and disease prevention [37,38,39,40,41,42]. For instance, Zhaoyuan Huang et al. [42] found that natural walking environments improve mood and reduce fatigue. Green roofs, although often commercially focused with simple designs [43], can alleviate stress and enhance attention when optimized with diverse plant species and accessible settings [44].

2.2.3. Security

Safety, fundamental to user health and comfort, addresses risks in dense urban environments [45,46]. Gökçen Firdevs Yücel et al. [47] demonstrated, via Istanbul’s Ortaköy Square, that safe public spaces boost environmental comfort and well-being. Design strategies like pedestrian-vehicle separation, slip-resistant surfaces, and signage enhance safety [48], while accessible facilities for disabled users increase participation [49].

2.2.4. Ecology

Ecology emphasizes sustainable spatial design that harmonizes with the natural environment, improving microclimates and user health [50]. Jun Zhang [51] identified a 40–45% greening rate as optimal for mental fatigue recovery, while studies link urban green spaces to health equity and reduced depression [35,52,53]. In high-density cities, ecological design mitigates environmental pressures effectively.

2.2.5. Humanism

Humanism integrates cultural identity, social needs, and management into space design. Linus W. Dietz et al. [54] highlight site-specific cultural elements in parks as health promoters, while Michael J. Chandler [55] links cultural continuity to lower suicide rates. Effective management, including maintenance and community engagement, enhances comfort and well-being [19,56].

2.3. Health-Related Theory

2.3.1. Health Behavior Change Theory

Health Behavior Change Theory, which originated in psychology, has been extensively applied in various fields, particularly in health and education, where it has played a pivotal role in advancing the study of health behavior promotion at both the individual and group levels. Health behaviors encompass actions individuals engage in to maintain health and prevent illness, such as adopting healthy eating habits, engaging in regular physical activity, quitting smoking, and limiting alcohol consumption. The theory of behavior change seeks to identify the key factors influencing health behavior modification, with a particular focus on how individual cognition, social context, and physical environment collectively influence behavioral decisions.

In the 1950s, Health Behavior Theory introduced foundational frameworks such as the ecological perspective, social cognitive theory, and the health belief model. The ecological perspective emphasizes the multi-layered influences on health behavior, positing that an individual’s actions are shaped not only by personal factors, but also by external forces, including the social and physical environments. Social cognitive theory focuses on how individuals modify their health behaviors by observing and modeling others’ actions, highlighting the influence of social interactions and contextual factors in facilitating health-promoting behaviors. The health behavior change model investigates how individuals’ perceptions of health risks, as well as the perceived benefits and barriers to behavior change, influence their health-related decision-making processes.

As research advanced, scholars expanded the theoretical framework of health behavior change, highlighting the extensive and long-term influence of policies and the built environment on health behaviors. Studies have shown that critical factors—such as architectural design, the configuration of public facilities, and the accessibility and safety of surrounding areas—play a significant role in shaping individuals’ willingness to engage in health-promoting activities and their actual participation. Consequently, changes in health behavior are not solely the result of individual decisions, but also the outcome of the complex interplay between environmental and social factors.

In the fields of architecture and urban planning, the theory of health behavior change has become a crucial foundation for designing outdoor spaces. Optimizing the surrounding built environment can effectively encourage residents to participate in physical activities, enhance social interactions, and enjoy recreational opportunities, thereby improving the overall health and quality of life for all residents. The design of outdoor spaces should prioritize not only esthetics and functionality, but also the creation of environments that promote healthy living, ultimately contributing to the broader improvement in residential quality.

2.3.2. Healthy City Theory

The theory of healthy cities originated from extensive global research on the relationship between urban environments and public health. Since the 1970s, with the intensification of global economic and social instability, issues such as environmental degradation, rising unemployment, and increasing population mobility have become more pressing, catalyzing the development of the healthy city theory. In 1994, the World Health Organization (WHO) defined a healthy city as follows:

“A healthy city is one that continuously develops and enhances its natural and social environments and expands social resources, enabling individuals to support each other in enjoying life and fully realizing their potential”.

This concept aims to create an urban environment conducive to health through scientific planning and effective management. It stresses that cities must provide not only basic healthcare and sanitation facilities, but also address a wide range of factors, including ecological environments, social functions, and human-centered care, ensuring that the city becomes a livable space that fosters health.

As the development of healthy cities progresses, outdoor urban spaces assume a pivotal role in national development. These spaces not only provide areas for residents’ daily activities, but also foster the dissemination of health-promoting knowledge. The design of outdoor spaces, such as urban parks, plazas, pedestrian streets, and bike paths, profoundly influences public engagement in physical activities, social interaction frequency, and the mental health of the community at large. The concept of healthy cities seeks to establish an optimal living environment that promotes citizens’ health, with an emphasis on ensuring that architectural and outdoor space planning prioritize both comfort and safety. It strongly encourages public participation in daily physical activities such as walking, running, fitness routines, and cycling, thereby reinforcing the social cohesion of urban residents and enhancing the nation’s overall quality of life.

In response to the COVID-19 pandemic, the critical importance of strengthening public health systems and emergency response mechanisms in the development of healthy cities has become even more pronounced. During key stages of the pandemic, the creation of “healthy cities” necessitates the expansion and improvement of medical facilities, as well as immediate adjustments to the design and layout of buildings and outdoor spaces. Under such exceptional circumstances, adjusting spatial configurations can facilitate greater physical distancing while bolstering the overall health defense framework. Through meticulous planning and thoughtful design, the strategic development of outdoor spaces can substantially reduce the risk of infection in high-density areas, thereby enhancing both the mental and physical well-being of citizens.

2.4. Problem Description

Despite consensus on designing healthy outdoor public spaces, existing evaluation frameworks primarily target large urban areas, overlooking building-oriented spaces. This study addresses this gap by developing a health evaluation framework to quantify design impacts on user health, aiding designers in identifying and rectifying health-related deficiencies.

3. Research Methodology

3.1. Fuzzy Delphi Method

The Delphi method was coined by the RAND Corporation [57]. This is a qualitative method often used to help a group of experts reach a consensus when identifying and reviewing the relevant factors in multi-criteria decision-making problems [58,59]. Although the traditional Delphi method is used to identify these factors, the fuzziness and uncertainty of expert opinions still persist. Akira Ishikawa et al. [60] combined fuzzy set theory with the Delphi method, which helps to overcome the vagueness and subjectivity of human thinking, judgment, and expression. Ching-Hsue Cheng and Yin Lin [61]. In the military field, the fuzzy Delphi method is used to reach a consensus among experts through triangular fuzzy numbers. Therefore, this study employs the fuzzy Delphi method (FDM) to identify and review the design factors affecting the health of outdoor architectural spaces. Below is the standard calculation process for implementing the FDM.

Ask the experts to use a linguistic scale to rate the importance of each factor affecting the health design of outdoor spaces in buildings, as shown in

Table 2. Linguistic scales for the FDM.

Rating	Linguistic Terms	Description
1	Very Unimportant	Has a negligible impact on the health of outdoor public spaces, almost ignorable.
2	Very Low Importance	Minimal impact, rarely needs consideration.
3	Low Importance	Occasionally impactful, but generally not a priority.
4	Slightly Low Importance	Somewhat impactful, but not usually a key factor.
5	Moderate Importance	Noticeably affects the health of outdoor spaces, deserves attention.
6	Moderately High Importance	Significantly impacts the health of public spaces, often needs to be considered.
7	High Importance	One of the key factors affecting outdoor space health, should be prioritized.
8	Very High Importance	Has a major impact, is very critical.
9	Extremely High Importance	Almost determines the health status of outdoor spaces, must be prioritized.
10	Crucial	The most central factor, vitally important to the health of outdoor spaces.

. This scale helps to aggregate expert opinions using the triangular fuzzy method [62], which can be represented as follows:

$${\mathrm{V}}_{\mathrm{i}}=({\mathrm{C}\mathrm{i}}_{\mathrm{x}},{\mathrm{O}\mathrm{i}}_{\mathrm{x}})$$

(1)

In the equation, ${\mathrm{V}}_{\mathrm{i}}$ denotes the expert’s set of evaluations for factor x, ${\mathrm{C}\mathrm{i}}_{\mathrm{x}}$ represents the expert’s conservative estimate of the factor’s importance, and ${\mathrm{O}\mathrm{i}}_{\mathrm{x}}$ reflects the expert’s optimistic estimate of its significance.

Construct the conservative cognitive dual-triangular fuzzy number set and the optimistic cognitive dual-triangular fuzzy number set, using the geometric mean method to calculate the comprehensive rating of each factor. The specific formula can be expressed as follows:

$$\mathrm{C}=({\mathrm{C}}_{\mathrm{L}}^{\mathrm{i}},{\mathrm{C}}_{\mathrm{M}}^{\mathrm{i}},{\mathrm{C}}_{\mathrm{U}}^{\mathrm{i}})=({\mathrm{C}}_{\mathrm{L}}^{\mathrm{i}},{\displaystyle \frac{1}{\mathrm{P}}}{\sum }_{\mathrm{p}=1}^{\mathrm{P}} \mathrm{C}{\mathrm{i}}_{\mathrm{x}}^{\mathrm{p}}{\displaystyle \frac{1}{\mathrm{P}}},{\mathrm{C}}_{\mathrm{U}}^{\mathrm{i}})$$

(2)

In the equation, $\mathrm{C}$ denotes the set of conservative triangular fuzzy numbers, ${\mathrm{C}}_{\mathrm{L}}^{\mathrm{i}}$ represents the minimum value within the conservative estimate, ${\mathrm{C}}_{\mathrm{M}}^{\mathrm{i}}$ corresponds to the geometric mean of the conservative estimate, and ${\mathrm{C}}_{\mathrm{U}}^{\mathrm{i}}$ represents the maximum value within the conservative estimate.

$$\mathrm{O}=({\mathrm{O}}_{\mathrm{L}}^{\mathrm{i}},{\mathrm{O}}_{\mathrm{M}}^{\mathrm{i}},{\mathrm{O}}_{\mathrm{U}}^{\mathrm{i}})=({\mathrm{O}}_{\mathrm{L}}^{\mathrm{i}},{\displaystyle \frac{1}{\mathrm{P}}}{\sum }_{\mathrm{p}=1}^{\mathrm{P}} \mathrm{O}{\mathrm{i}}_{\mathrm{x}}^{\mathrm{p}}{\displaystyle \frac{1}{\mathrm{P}}},{\mathrm{O}}_{\mathrm{U}}^{\mathrm{i}})$$

(3)

In the equation, $\mathrm{O}$ denotes the set of optimistic triangular fuzzy numbers, ${\mathrm{O}}_{\mathrm{L}}^{\mathrm{i}}$ represents the minimum value within the optimistic estimate, ${\mathrm{O}}_{\mathrm{M}}^{\mathrm{i}}$ corresponds to the geometric mean of the optimistic estimate, and ${\mathrm{O}}_{\mathrm{U}}^{\mathrm{i}}$ represents the maximum value within the optimistic estimate.

$${\mathrm{Z}}^{\mathrm{i}}={\mathrm{C}}_{\mathrm{U}}^{\mathrm{i}}-{\mathrm{O}}_{\mathrm{L}}^{\mathrm{i}}$$

(4)

In the equation, ${\mathrm{Z}}^{\mathrm{i}}$ denotes the intersecting region of two triangular fuzzy numbers.

If ${\mathrm{C}}_{\mathrm{U}}^{\mathrm{i}}$ > ${\mathrm{O}}_{\mathrm{L}}^{\mathrm{i}}$, it indicates that the two fuzzy triangular numbers do not overlap, and there is no gray area, which means the opinions of industry experts are consistent. Then,

$${\mathrm{G}}^{\mathrm{i}}={\displaystyle \frac{({\mathrm{C}}_{\mathrm{M}}^{\mathrm{i}}+{\mathrm{O}}_{\mathrm{M}}^{\mathrm{i}})}{2}}$$

(5)

If two fuzzy triangular numbers overlap, a gray area exists (${\mathrm{Z}}^{\mathrm{i}}$), meaning that ${\mathrm{C}}_{\mathrm{U}}^{\mathrm{i}}$ ≤ ${\mathrm{O}}_{\mathrm{L}}^{\mathrm{i}}$:

$${\mathrm{G}}^{\mathrm{i}}={\displaystyle \frac{\mathrm{m}\mathrm{i}\mathrm{n}({\mathrm{C}}_{\mathrm{U}}^{\mathrm{i}},{\mathrm{O}}_{\mathrm{L}}^{\mathrm{i}})+\mathrm{m}\mathrm{a}\mathrm{x}({\mathrm{C}}_{\mathrm{L}}^{\mathrm{i}},{\mathrm{O}}_{\mathrm{U}}^{\mathrm{i}})}{2}}$$

(6)

$${\mathrm{M}}^{\mathrm{i}}={\mathrm{O}}_{\mathrm{M}}^{\mathrm{i}}-{\mathrm{C}}_{\mathrm{M}}^{\mathrm{i}}$$

(7)

In the equation above, ${\mathrm{G}}^{\mathrm{i}}$ denotes the expert consensus value between ${\mathrm{O}}_{\mathrm{L}}^{\mathrm{i}}$ and ${\mathrm{C}}_{\mathrm{U}}^{\mathrm{i}}$ in the function graph, while ${\mathrm{M}}^{\mathrm{i}}$ represents the test value for the gray zone.

If ${\mathrm{Z}}^{\mathrm{i}}$ > ${\mathrm{M}}^{\mathrm{i}}$, it indicates that the experts’ opinions are inconsistent. The indicators need to be modified according to the experts’ opinions, and the experts’ opinions should be collected again until they converge and reach a consensus. The double-trigonometric function graph is shown in Figure 2.

Based on the expert consensus value ${\mathrm{G}}^{\mathrm{i}}$, determine the threshold value for the indicators and filter the indicators according to the threshold value. The formula for calculating the threshold value is as follows:

$$\mathrm{S}=\sqrt[\mathrm{n}]{{\mathrm{G}}^1{\mathrm{G}}^2{\mathrm{G}}^3...{\mathrm{G}}^{\mathrm{n}}}$$

(8)

In the formula, S represents the threshold value of the indicator.

By comparing the consensus value (${\mathrm{G}}^{\mathrm{i}}$) with the threshold value($\mathrm{S}$), if ${\mathrm{G}}^{\mathrm{i}}$ ≥ $\mathrm{S}$, the indicator is retained; if ${\mathrm{G}}^{\mathrm{i}}$ < $\mathrm{S}$, the indicator is deleted.

3.2. Analytic Hierarchy Process

After identifying the design factors affecting the health of outdoor spaces through the FDM, we constructed a multi-level hierarchical model and employed the AHP method to calculate the weight of each factor [63,64]. These factors were then ranked in ascending order based on their respective weights. The Analytic Hierarchy Process (A.H.P.), introduced by Saaty in 1988, has since been widely adopted for multi-criteria and multi-factor decision-making problems to establish rankings and determine priorities [65]. This method evaluates the relative importance of each criterion, calculates the weights of the standards, and formulates the evaluation framework [66]. The following outlines the key steps involved in calculating the weight of each factor:

• Experts are instructed to compare the relative importance of indicators at the same level in pairs, based on the AHP language scale presented in

  Table 3. Linguistic scales for the AHP.

  Rating	Linguistic Terms	Description
  1	Equally important	Both elements contribute equally to the goal.
  3	Slightly more important	One element is slightly more important than the other.
  5	Significantly more important	One element is significantly more important than the other.
  7	Strongly more important	One element is extremely more important than the other.
  2, 4, 6	Intermediate even values	Express intermediate levels of importance between adjacent odd values.
  1/2 to 1/7	Reciprocal values	Indicates varying degrees to which the other element is less important.

  .

Using the 1–9 scale, values are assigned to each indicator, and the pairwise decision matrix produces a comparison judgment matrix ($\tilde{\mathrm{A}}$), as demonstrated in the following equation:

$$\begin{array}{c}\tilde{\mathrm{A}}=\left[\begin{array}{ccccc}\mathrm{A}& {\mathrm{B}}_1& {\mathrm{B}}_2& \cdots & {\mathrm{B}}_{\mathrm{n}}\\ {\mathrm{B}}_1& {\mathrm{a}}_{11}& {\mathrm{a}}_{12}& \cdots & {\mathrm{a}}_{1\mathrm{n}}\\ {\mathrm{B}}_2& {\mathrm{a}}_{21}& {\mathrm{a}}_{22}& \cdots & {\mathrm{a}}_{2\mathrm{n}}\\ \cdots & \cdots & \cdots & {\mathrm{a}}_{\mathrm{i}\mathrm{j}}& ⋮\\ {\mathrm{B}}_{\mathrm{n}}& {\mathrm{a}}_{\mathrm{n}1}& {\mathrm{a}}_{\mathrm{n}2}& \cdots & {\mathrm{a}}_{\mathrm{n}\mathrm{n}}\end{array}\right]\end{array}$$

(9)

In the equation above, the ratio of the relative importance between elements i and j in the judgment matrix $\tilde{\mathrm{A}}$ is represented by ${\mathrm{a}}_{\mathrm{i}\mathrm{j}}$. The greater the ratio, the higher the importance of element i, and the following relationship holds:

$${\mathrm{a}}_{\mathrm{i}\mathrm{j}}\ne 0,{\mathrm{a}}_{\mathrm{i}\mathrm{j}}={\displaystyle \frac{1}{{\mathrm{a}}_{\mathrm{i}\mathrm{j}}}},{\mathrm{a}}_{\mathrm{i}\mathrm{j}}=1,\mathrm{i}\mathrm{j}=\mathrm{1,2},\dots ,\mathrm{n}$$

(10)

• Use the root-finding method to calculate the approximate value of the eigenvector of the judgment matrix; the formula is as follows:

$${\mathrm{W}}_{\mathrm{i}}={\displaystyle \frac{(\prod _{\mathrm{j}=1}^{\mathrm{n}} {\mathrm{a}}_{\mathrm{i}\mathrm{j}}{)}^{\frac{1}{\mathrm{n}}}}{\sum _{\mathrm{i}=1}^{\mathrm{n}} (\prod _{\mathrm{j}=1}^{\mathrm{n}} {\mathrm{a}}_{\mathrm{i}\mathrm{j}}{)}^{\frac{1}{\mathrm{n}}}}},\mathrm{i}\mathrm{j}=\mathrm{1,2},\dots ,\mathrm{n}$$

(11)

Upon normalizing the eigenvector, the weight vector $\mathrm{W}$ is derived, $\mathrm{W}=({\mathrm{W}}_1,{\mathrm{W}}_2,\dots ,{\mathrm{W}}_{\mathrm{n}}{)}^{\mathrm{T}}$.

• To ensure the logical consistency and completeness of the importance judgment content and to establish scientific indicator weights, we conducted consistency checks on the judgment matrices at each level. The specific formulas are as follows:

$$\mathrm{C}.\mathrm{I}.={\displaystyle \frac{{\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}-\mathrm{n}}{\mathrm{n}-1}}$$

(12)

$${\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}=\sum {\displaystyle \frac{(\mathrm{A}\mathrm{W}{)}_{\mathrm{i}}}{\mathrm{n}{\mathrm{W}}_{\mathrm{i}}}}$$

(13)

$$\mathrm{C}.\mathrm{R}.={\displaystyle \frac{\mathrm{C}.\mathrm{I}.}{\mathrm{R}.\mathrm{I}.}}$$

(14)

In the above formula,${\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}$ is the maximum eigenvalue of the judgment matrix, n is the order of the judgment matrix, and $\mathrm{R}.\mathrm{I}.$ is the predetermined value based on the order of the matrix.

If the random consistency ratio $\mathrm{C}.\mathrm{R}.<0.1$, the calculated hierarchical ranking weights are considered to be valid and reasonable. If not, the judgment matrix must be adjusted until it satisfies the consistency test.

3.3. Multi-Level Fuzzy Comprehensive Evaluation Model

Following the determination of each indicator’s weight in the second stage, we apply the multi-level fuzzy comprehensive evaluation method to assess the design factors influencing the health of outdoor spaces in architecture [67]. The multi-level fuzzy evaluation method, a systematic approach designed to handle fuzziness, multi-layered structures, and multiple criteria, has been widely adopted in various fields, including network security [68], urban systems [67], water resources [69], and power systems [70]. The following outlines the detailed process for constructing the multi-level fuzzy evaluation model:

• Determine the evaluation factors and comment set. The set of evaluation factors is a collection of various factors that can influence the evaluation results, denoted by U.
  The set of comments is a collection of evaluation results given by the evaluator to the evaluated object, denoted by V, and the results are scored on a scale of 1 to 4—that is, the set of comments can be represented as follows:

  $$\mathrm{V}=\left\{{\mathrm{V}}_1\mathrm{P}\mathrm{o}\mathrm{o}\mathrm{r},{\mathrm{V}}_2\mathrm{F}\mathrm{a}\mathrm{i}\mathrm{r},{\mathrm{V}}_3\mathrm{G}\mathrm{o}\mathrm{o}\mathrm{d},{\mathrm{V}}_4\mathrm{E}\mathrm{x}\mathrm{c}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{e}\mathrm{n}\mathrm{t},\right\}$$

  (15)

• To explain the degree of membership of indicator factors within the fuzzy set concerning health in architectural outdoor spaces, this study introduces the fundamental concept of membership degree. First proposed by Lotfi A. Zadeh in 1965, this concept was developed to overcome the limitations of traditional binary logic in dealing with the complexity of the real world [71]. The membership degree ranges from [0, 1], with higher values signifying a stronger association of the indicator with the fuzzy set.
• In this study, the evaluation index factors are divided into two categories: quantitative indicators and qualitative indicators. These types of indicators have different methods for calculating membership degrees, which will be detailed in the following sections.

• Quantitative Indicators

Quantitative indicators refer to indicators that can be precisely measured through numerical values. For the calculation of the membership degree of such indicators, mathematical functions are usually used to represent them. In this study, the membership degree of the quantitative indicators in the evaluation index system shows a semi-trapezoidal distribution, so a (semi) trapezoidal function is used to determine its membership degree, as shown in Figure 3. The detailed calculation formula for the membership degree is as follows:

$$\mathrm{R}\mathrm{a}(\mathrm{i})=\left\{\begin{array}{c}1,\mathrm{i}<\mathrm{a} \\ {\displaystyle \frac{\mathrm{b}-\mathrm{i}}{\mathrm{b}-\mathrm{a}}},\mathrm{a}\le \mathrm{i}\le \mathrm{b} \\ 0,\mathrm{i}>\mathrm{b} \end{array}\right.$$

(16)

$$\mathrm{R}\mathrm{b}(\mathrm{i})=\left\{\begin{array}{c}1-\mathrm{R}\mathrm{a}(\mathrm{i})\mathrm{,}\mathrm{a}<\mathrm{i}<\mathrm{b} \\ {\displaystyle \frac{\mathrm{c}-\mathrm{i}}{\mathrm{c}-\mathrm{b}}},\mathrm{b}\le \mathrm{i}\le \mathrm{c} \\ 0,\mathrm{i}<\mathrm{a} \mathrm{o}\mathrm{r} \mathrm{i}>\mathrm{c} \end{array}\right.$$

(17)

$$\mathrm{R}\mathrm{c}(\mathrm{i})=\left\{\begin{array}{c}1-\mathrm{R}\mathrm{b}(\mathrm{i}),\mathrm{b}<\mathrm{i}<\mathrm{c} \\ {\displaystyle \frac{\mathrm{d}-\mathrm{i}}{\mathrm{d}-\mathrm{c}}},\mathrm{c}\le \mathrm{i}\le \mathrm{d} \\ 0,\mathrm{i}<\mathrm{c} \mathrm{o}\mathrm{r} \mathrm{i}>\mathrm{d} \end{array}\right.$$

(18)

$$\mathrm{R}\mathrm{d}(\mathrm{i})=\left\{\begin{array}{c}1-\mathrm{R}\mathrm{c}(\mathrm{i}),\mathrm{c}<\mathrm{i}<\mathrm{d} \\ {\displaystyle \frac{\mathrm{e}-\mathrm{i}}{\mathrm{e}-\mathrm{d}}}, \mathrm{d}\le \mathrm{i}\le \mathrm{e} \\ 0, \mathrm{i}<\mathrm{d} \mathrm{o}\mathrm{r} \mathrm{i}>\mathrm{e} \end{array}\right.$$

(19)

In the above formula, $\mathrm{R}$ represents the membership degree of the indicator, $\mathrm{i}$ is the actual value of the $\mathrm{i}$-th indicator evaluation set for calculation, and $\mathrm{V}=\left\{\mathrm{a},\mathrm{b},\mathrm{c},\mathrm{d}\right\}$ is the evaluation grade standard.

2.

Qualitative Indicators

Qualitative indicators are metrics evaluated based on descriptions or classifications, and their membership degrees are usually determined by user evaluations. The specific calculation formula is as follows:

$${\mathrm{R}}_{\mathrm{i}}={\displaystyle \frac{\mathrm{X}}{\mathrm{M}}}$$

(20)

In the above formula, $\mathrm{R}$ represents the degree of membership of the indicator, $\mathrm{M}$ represents the total number of questionnaires for the evaluation object, and $\mathrm{X}$ represents the proportion of the evaluation object that has achieved the evaluation grade.

Based on the above, the actual values of the evaluation indicators can be calculated for the four evaluation levels, forming the membership degree set at $\mathrm{R}=\left\{{\mathrm{R}}_{\mathrm{a}},{\mathrm{R}}_{\mathrm{b}},{\mathrm{R}}_{\mathrm{c}},{\mathrm{R}}_{\mathrm{d}}\right\}$.

• Based on the previous step’s calculation results, you can construct the membership degree matrix ($\tilde{\mathrm{R}}$) for each basic indicator factor in the evaluation index system, as shown below:

  $$\tilde{R}=\left[\begin{array}{ccccc}{R}_{1a}& \cdots & R_{1j}& \cdots & R_{1n}\\ ⋮& \ddots & ⋮& & ⋮\\ R_{ia}& \cdots & R_{ij}& \cdots & R_{in}\\ ⋮& & ⋮& \ddots & ⋮\\ R_{ma}& \cdots & R_{mj}& \cdots & R_{mn}\end{array}\right]$$

  (21)

In the above formula, $\tilde{\mathrm{R}}$ represents the membership degree matrix, $\mathrm{m}$ is the number of rows in the matrix, each row represents an object, $\mathrm{n}$ is the number of columns in the matrix, and each column represents an evaluation criterion.

• Based on the weight vector obtained from the second phase operation and the fuzzy matrix obtained in the previous step, the fuzzy comprehensive evaluation conclusion vector can be obtained through weighted calculation. The algorithm formula is as follows:

  $$\mathrm{B}=\mathrm{W} \ast \tilde{\mathrm{R}}$$

  (22)
• Among them,

  $$\mathrm{W}=({\mathrm{w}}_{\mathrm{i}}{)}_{\mathrm{i}=1}^{\mathrm{n}}$$

  (23)

  $$\tilde{\mathrm{R}}=[{\overrightarrow{\mathrm{r}}}_1,{\overrightarrow{\mathrm{r}}}_2,\dots ,{\overrightarrow{\mathrm{r}}}_{\mathrm{n}}]$$

  (24)

  $${\overrightarrow{\mathrm{r}}}_{\mathrm{j}}=\left[\begin{array}{c}{\mathrm{R}}_{1\mathrm{j}}\\ ⋮\\ {\mathrm{R}}_{\mathrm{m}\mathrm{j}}\end{array}\right]$$

  (25)

In the above formula, $\mathrm{B}$ represents the fuzzy comprehensive evaluation conclusion vector, $\mathrm{W}$ is the weight vector, $\tilde{\mathrm{R}}$ is the fuzzy matrix, and ${\overrightarrow{\mathrm{r}}}_{\mathrm{j}}$ represents all the elements in the j-th column.

• Based on the above calculations, the fuzzy evaluation results for each element layer can be obtained. The calculation formula is as follows:

  $$\mathrm{E}=\mathrm{B} \ast \mathrm{H}$$

  (26)

In the above formula, $\mathrm{E}$ represents the comprehensive evaluation score of the factor index layer, $\mathrm{B}$ represents the weight vector of the criterion layer index relative to the overall goal layer, and $\mathrm{H}$ is the measurement scale (i.e., ordinal quantification scores of 1, 2, 3, 4).

• Calculate the fuzzy comprehensive evaluation values of all indicators at the element level in this manner, forming the fuzzy comprehensive evaluation for each element level.
• The calculation method for fuzzy evaluation at the criterion level is the same as that at the element level. After calculating the fuzzy evaluation vectors and comprehensive evaluation values for all indicators at the criterion level, the fuzzy comprehensive evaluation at the criterion level is formed.
• By performing a comprehensive analysis of the resulting data from the fuzzy comprehensive evaluations of the factor and criterion layers, we can compute the fuzzy comprehensive evaluation for the overall goal layer. The calculation process mirrors that of the fuzzy evaluations for the factor and criterion layers.
• Based on the data obtained from the above process and the established fuzzy rules, the comprehensive evaluation results of the target can be calculated. Finally, the target is evaluated based on the calculated results. This study categorizes the five evaluation results of the health assessment set for outdoor architectural spaces into good, better, average, and poor, with corresponding score ranges of (3, 4], (2, 3], (1, 2], and (0, 1], respectively. See

  Table 4. Comparative standards and ratings for health assessments of the building’s outdoor public spaces.

  Rank	Category	Interval	Description
  1	Excellent	3 ≤ E ≤ 4	Highly effective in promoting social interaction and enhancing the function of the building’s outdoor public spaces, with a clear positive impact.
  2	Good	2 ≤ E < 3	Effectively promotes social interaction and enhances the function of the building’s outdoor public spaces, with a noticeable positive impact.
  3	Fair	1 ≤ E < 2	Ineffective in promoting social interaction or enhancing the function of the building’s outdoor public spaces, with no clear benefit.
  4	Poor	E ≤ 1	Minimally effective in promoting social interaction and enhancing the function of the building’s outdoor public spaces, with minimal positive impact.

  for details.

4. Application of the Proposed Framework

4.1. Data Collection

In this study, we selected ten experts to gather their opinions on the evaluation indicators within the framework for assessing outdoor spaces in health-oriented architecture. Among them, four experts are university professors with a background in architecture, one expert is a university professor specializing in urban and rural planning, three experts are designers from architectural design institutes, and two experts are attending physicians from tertiary hospitals. The selection of experts was made to ensure a diverse representation of professionals from relevant fields, allowing for a comprehensive evaluation. The number of experts was chosen based on similar studies, such as Zhuanglin Ma et al. [62], who selected thirteen experts, and Rohit Gupta et al. [72], who selected ten experts, indicating that selecting ten experts is a reasonable choice.

To ensure the reliability and consistency of the experts’ assessments, we performed an inter-rater agreement test using Fleiss’ Kappa. This test evaluates the level of agreement among multiple experts on categorical ratings. The calculation resulted in a Fleiss’ Kappa value of 0.72, which indicates significant consistency (with a Kappa value in the range of 0.61–0.80, reflecting substantial agreement). This result suggests that the experts’ evaluations are highly reliable and provide a solid foundation for calculating the fuzzy weightings.

For the quantitative environmental indicators, data were collected using industry-standard tools and methods. Air quality was assessed using AQI data from local monitoring points, and field measurements of air humidity and noise levels were taken at regular intervals throughout the day using handheld sensors. Drinking water quality was tested in a laboratory based on site samples. Various spatial indicators, such as shaded area ratios, green visibility, and the mix of land uses, were determined using satellite imagery and geographic information system (GIS) analysis. Additionally, the site’s design features, such as rest facilities, fitness trails, and plant diversity, were measured through field surveys and satellite imagery. These measurements provided a comprehensive dataset to evaluate the health-oriented aspects of the outdoor space.

Detailed methods for each indicator, including measurement frequencies and specific tools used, are outlined in

Table A2. Quantitative index measurement method.

Quantitative Indicators	Measurement Method
C1: Air quality	The 22 national and municipal air quality monitoring points in Wuhan and their air quality index (AQI) data in the past 15 days (17 April~1 May 2024) were imported into Arc GIS (10.2) software for interpolation analysis (ordinary kriging method), and the project points were sampled and calculated
C2: Air humidity	Three days were randomly selected for field measurements using handheld humidity measuring instruments, from 6 a.m. to 11 p.m., every half hour, and finally averaged
C3: Drinking water quality	Drinking water is sampled from the site and sent to the laboratory to measure the water quality
C5: Shaded area ratio	Select the shaded areas in the venue at 9 o’clock, 12 o’clock, 15 o’clock, and 18 o’clock, count its proportion in the overall field, and finally take the average value of the four time periods
C6: Vertical greening area	Photographs of the four facades of the building were taken by a drone and imported into Adobe Photoshop (2021) to calculate the ratio of the number of green pixels to the number of pixels in the building
C7: Site noise level	3 days were randomly selected for on-site measurements using a handheld decibel meter, every half hour from 6 a.m. to 11 p.m., and finally averaged
C8: Green buffer ratio between site and urban roads	By downloading a satellite map, the ratio of the greened part to the overall boundary is calculated by marking the greened part of the site and the road boundary of the site
C9: Area of exercise and fitness spaces for the elderly	By downloading a satellite map, mark the area of the site and all outdoor areas for users, and separately mark the area for the elderly, and calculate the proportion of the area
C10: Area of playgrounds and sports facilities for children	By downloading a satellite map, mark the area of the site and the outdoor area for the use of users, and mark the area for children separately to calculate the proportion of the area
C11: Number of rest facilities	By downloading a satellite map, mark the site and all the facilities for users to rest, and count the number of them
C12: Length of fitness trails	By downloading a satellite map, mark the continuous fitness trails in the site for users to exercise and count their length
C13: Proportion of ground-level space elevated	Through field investigation, the area of the ground floor overhead area in the site is counted, and the ratio of it to the overall area of the first floor is calculated
C14: Mix of land uses	By downloading a satellite map, mark the outdoor areas with multi-use functions and calculate the area ratio to all outdoor areas
C15: Rooftop space utilization rate	By downloading a satellite map, mark the area where the roof of the building can be used and calculate the ratio of it to the overall roof area
C18: Slip resistance of pavements	By downloading a satellite map, mark the areas of the building’s outdoor space with non-slip paving, and calculate the ratio of its area to the overall outdoor space
C21: Coverage rate of security systems	Through field research, the monitoring screen in the site is obtained, and the area covered by the site is drawn in Adobe Photoshop (2021), and the ratio of the area to the overall area is counted
C22: Plant diversity	Plant species in the site were obtained through field research
C23: Greening rate	By downloading a satellite map, mark the greenery in the site and calculate its area ratio to the overall site
C24: Green visibility rate	The green view rate was selected to quantitatively analyze the green landscape of the three-dimensional space, the four corners of the theme building of the site were selected as the sampling points, the camera was placed vertically at a height of 160 cm, and the panoramic photos were taken, respectively
C29: Number of participants in activities promoting interaction and integration	The number of people involved in activities that promote interaction and integration is counted and averaged from 6 a.m. to 10 p.m. throughout the week
C30: Frequency of organized activities	Count the number of activities organized in the venue in a month

.

4.2. Study Area

The case study focuses on the Wuhan Citizen’s Home, located in the urban core of Wuhan, China. This site was selected due to its representation of a high-density urban environment, characteristic of many large cities, particularly in rapidly developing regions. Wuhan, a major city with diverse land use, provides a relevant context for evaluating the health-oriented design of outdoor spaces in urban settings.

The study area is situated in proximity to various urban zones (see Figure 4), including residential areas, commercial office buildings, industrial zones, and concentrated urban green spaces. These diverse functions within the surrounding area reflect typical characteristics of urban environments, making it a suitable representative of Wuhan’s urban fabric. The site itself features multiple functional zones (see Figure 4), such as seating areas, green spaces, squares, and parking areas, all designed to promote public health through accessibility and social interaction. The spatial configuration of the Citizen’s Home, combined with its proximity to key urban features, allows for an examination of how urban space organization influences health outcomes.

The outdoor space design of the Citizen’s Home is centered around the building, utilizing small, dispersed open spaces to provide citizens with health-oriented areas for daily activities. Located in the heart of Wuhan, the Citizen’s Home features highly open, diverse outdoor spaces that reflect the healthy design of small urban spaces. Unlike large centralized public spaces, these flexible small spaces meet the daily and diverse health needs of surrounding citizens. As such, the Citizen’s Home serves as a key case study for evaluating health-oriented small open spaces in urban centers, offering valuable insights for future public space design.

4.3. Phase 1: Identification of the Relevant Factors Using FDM

Drawing on existing research and design guidelines, we identified 53 design factors influencing the health of architectural outdoor spaces and developed a corresponding indicator selection questionnaire. This questionnaire was distributed to ten experts, who were asked to evaluate each factor using a fuzzy language scale.

Based on the collection and organization of expert opinions (as presented in

Table A1. Expert fuzzy evaluation of the importance of assessment indicators results.

Evaluation Criteria	$\mathbf{Cognitive} \mathbf{Conservatism} \mathbf{Bias} {\mathbf{C}}^{\mathbf{i}}$			$\mathbf{Optimistic} \mathbf{Cognitive} \mathbf{Bias} {\mathbf{O}}^{\mathbf{i}}$
	${\mathbf{C}}_{\mathbf{L}}^{\mathbf{i}}$	${\mathbf{C}}_{\mathbf{U}}^{\mathbf{i}}$	${\mathbf{C}}_{\mathbf{M}}^{\mathbf{i}}$	${\mathbf{O}}_{\mathbf{L}}^{\mathbf{i}}$	${\mathbf{O}}_{\mathbf{U}}^{\mathbf{i}}$	${\mathbf{O}}_{\mathbf{M}}^{\mathbf{i}}$
Comfort (A1)
Air (B1)
Air quality (C1)	7	9	8.14	9	10	9.86
Air humidity (C2)	4	9	5.86	8	10	8.29
Location and number of smoking areas (C3)	4	6	5.14	5	8	7
Intelligent air display system (C4)	1	5	2.14	4	7	4.43
Number of no-smoking signs (C5)	2	6	4	4	8	6.14
Water (B2)
Water quality for landscape use (C6)	3	7	5.14	6	8	7.29
Drinking water quality (C7)	6	9	8	8	10	9.71
Esthetic value of water features (C8)	3	7	5	5	8	7
Rainwater and wastewater discharge systems (C9)	2	8	6.29	6	10	8.57
Density of drinking water facilities (C10)	2	6	4.29	5	8	6.43
Illumination (B3)
Light pollution from nighttime illumination (C11)	1	5	3.71	3	7	5.73
Number of nighttime lighting fixtures (C12)	3	7	4.86	5	8	7
Thermal (B4)
Permeability rate of paved surfaces (C13)	4	7	5.57	5	9	7.86
Shaded area ratio (C14)	6	8	6.71	7	10	8.71
Vertical greening area (C15)	4	8	5.57	6	10	7.71
Sound(B5)
Site noise level (C16)	7	9	7.57	8	10	9.29
Green buffer ratio between site and urban roads (C17)	4	8	5.71	6	10	8.14
Diversity (A2)
Activity facilities (B6)
Area of exercise and fitness spaces for the elderly (C18)	5	8	6.14	7	10	8.57
Area of playgrounds and sports facilities for children (C19)	4	8	5.86	7	10	8.57
Number of rest facilities (C20)	4	8	5.86	6	10	7.86
Length and continuity of fitness trails (C21)	3	6	4.57	5	9	6.71
Site layout (B7)
Provision of basic cycling amenities (C22)	1	5	2.71	4	6	4.57
Length and continuity of bicycle lanes (C23)	2	5	3.29	4	6	5
Proportion of ground-level space elevated (C24)	3	7	5	6	9	7.14
Mix of land uses (C25)	4	8	5.86	6	10	7.86
Rooftop space utilization rate (C26)	5	7	6	7	10	8.43
Safety(A3)
Natural environment (B8)
Plant non-toxicity (C27)	5	8	7.43	8	10	9.43
Extreme climate contingency plan (C28)	4	8	6.71	8	10	9.14
Artificial environment (B9)
Safety of high-temperature light sources (C29)	2	8	4.57	5	9	7.71
Slip resistance of pavements (C30)	7	8	7.43	9	10	9.43
Drowning prevention facilities (C31)	2	5	3.57	3	9	6.43
Use of non-toxic materials (C32)	6	9	7.43	8	10	9.29
High-altitude falling object buffer zone (C33)	3	7	5.29	5	9	7.43
Pedestrian and vehicle segregation (C34)	5	8	6.71	6	8	8.71
Density of safety signage (C35)	2	5	3.43	4	7	5.43
Accessibility facilities (C36)	6	8	7.14	8	10	9.16
Coverage rate of security systems (C37)	5	8	6.29	7	10	8.14
Emergency response measures (C38)	5	8	6	7	9	8.29
Ecological (A4)
Natural landscape (B10)
Plant diversity (C39)	3	8	5.29	6	9	7.43
Greening rate (C40)	5	8	7	8	10	9.43
Green visibility rate (C41)	3	8	6	7	10	8.43
Spatial enclosure (C42)	3	7	4.86	6	8	6.71
Plant accessibility (C43)	1	5	4	3	10	6.71
Artificial landscape (B11)
Landscape color richness (C44)	1	5	2.29	2	7	5.43
Landscape interactivity (C45)	3	6	4.14	6	9	6.86
Number of structures (C46)	2	5	4	4	7	5.71
Landscape maintenance condition (C47)	5	8	6.29	7	9	7.86
Humanities (A5)
Culture (B12)
Display of local culture (C48)	4	8	5.57	6	10	7.71
Elderly and child-friendly care (C49)	6	8	6.71	8	10	9
Number of participants in activities promoting interaction and integration (C50)	4	6	5.17	6	9	7.33
Management (B13)
Frequency of organized activities (C51)	3	7	4.43	6	9	7
Frequency of plant maintenance (C52)	3	6	4.71	4	8	7.14
Intensity of management supervision (C53)	5	7	6	7	10	8.43

), we computed the design factors influencing the health of architectural outdoor spaces, with the final results shown in

Table 5. Assessment indicators screening table.

Evaluation Criteria	Gray Area Value ${\mathit{Z}}^{\mathit{i}}$	Consensus Value ${\mathit{G}}^{\mathit{i}}$	Gray Area Test Value ${\mathit{M}}^{\mathit{i}}$	Decision
Comfort (A1)
Air (B1)
Air quality (C1)	0	9.5	1.72	Accepted
Air humidity (C2)	1	7.075	2.43	Accepted
Location and number of smoking areas (C3)	1	6.07	1.86	Rejected
Intelligent air display system (C4)	1	3.285	2.29	Rejected
Number of no-smoking signs (C5)	2	5.07	2.14	Rejected
Water (B2)
Water quality for landscape use (C6)	1	6.215	2.15	Rejected
Drinking water quality (C7)	1	8.855	1.71	Accepted
Esthetic value of water features (C8)	2	6	2	Rejected
Rainwater and wastewater discharge systems (C9)	2	7.43	2.28	Accepted
Density of drinking water facilities (C10)	1	5.36	2.14	Rejected
Illumination (B3)
Light pollution from nighttime illumination (C11)	2	4.72	2.02	Rejected
Number of nighttime lighting fixtures (C12)	2	5.93	2.14	Rejected
Thermal (B4)
Permeability rate of paved surfaces (C13)	2	6	2.29	Rejected
Shaded area ratio (C14)	1	7.71	2	Accepted
Vertical greening area (C15)	2	6.64	2.14	Accepted
Sound (B5)
Site noise level (C16)	1	8.43	1.72	Accepted
Green buffer ratio between site and urban roads (C17)	2	6.925	2.43	Accepted
Diversity (A2)
Activity facilities (B6)
Area of exercise and fitness spaces for the elderly (C18)	1	7.355	2.43	Accepted
Area of playgrounds and sports facilities for children (C19)	1	7.215	2.71	Accepted
Number of rest facilities (C20)	2	6.86	2	Accepted
Length and continuity of fitness trails (C21)	1	7	2.14	Accepted
Site layout (B7)
Provision of basic cycling amenities (C22)	1	3.64	1.86	Rejected
Length and continuity of bicycle lanes (C23)	1	4.145	1.71	Rejected
Proportion of ground-level space elevated (C24)	1	6.07	2.14	Accepted
Mix of land uses (C25)	2	6.86	2	Accepted
Rooftop space utilization rate (C26)	0	7.215	2.43	Accepted
Safety (A3)
Natural environment (B8)
Plant non-toxicity (C27)	0	8.43	0	Accepted
Extreme climate contingency plan (C28)	0	7.925	0	Accepted
Artificial environment (B9)
Safety of high-temperature light sources (C29)	3	6.14	3.14	Rejected
Slip resistance of pavements (C30)	−1	8.43	2	Accepted
Drowning prevention facilities (C31)	2	5	2.86	Rejected
Use of non-toxic materials (C32)	1	8.36	1.86	Accepted
High-altitude falling object buffer zone (C33)	2	6.28	2.14	Rejected
Pedestrian and vehicle segregation (C34)	2	5.45	2	Rejected
Density of safety signage (C35)	1	4.43	2	Rejected
Accessibility facilities (C36)	0	6.8	2.02	Accepted
Coverage rate of security systems (C37)	1	7.215	1.85	Accepted
Emergency response measures (C38)	1	4.54	2.29	Rejected
Ecological (A4)
Natural landscape (B10)
Plant diversity (C39)	2	6.36	2.14	Accepted
Greening rate (C40)	0	8.215	2.43	Accepted
Green visibility rate (C41)	1	7.215	2.43	Accepted
The degree of empty depression (C42)	1	5.785	1.85	Rejected
Plant accessibility (C43)	2	5.355	2.71	Rejected
Artificial landscape (B11)
Landscape color richness (C44)	3	3.86	3.14	Rejected
Landscape interactivity (C45)	0	7.5	2.72	Accepted
Number of structures (C46)	1	4.855	1.71	Rejected
Landscape maintenance condition (C47)	1	7.075	1.57	Accepted
Humanities (A5)
Culture (B12)
Display of local culture (C48)	2	6.64	2.14	Accepted
Elderly and child-friendly care (C49)	0	7.855	2.29	Accepted
Number of participants in activities promoting interaction and integration (C50)	0	7.5	2.16	Accepted
Management (B13)
Frequency of organized activities (C51)	1	7.5	2.57	Accepted
Frequency of plant maintenance (C52)	2	5.925	2.43	Rejected
Intensity of management supervision (C53)	0	8.5	2.43	Accepted

. All indicators satisfied the condition ${\mathrm{Z}}^{\mathrm{i}}\le {\mathrm{M}}^{\mathrm{i}}$, confirming consistency across the experts’ opinions. The threshold value ($\mathrm{S}=6.301$) for selecting design factors was determined based on the geometric mean of the expert ratings. Specifically, we calculated the geometric mean of the consensus values for all indicators to establish a statistically grounded threshold. This method ensures the threshold reflects a statistically aggregated consensus from the experts’ ratings, ensuring that it is not arbitrary.

Indicators with a consensus value above this threshold ($\mathrm{S}=6.301$) were retained, as they were deemed to have a significant impact on health outcomes in architectural outdoor spaces. Indicators below this threshold were excluded due to their lower relevance or impact as per expert judgment. This approach is consistent with previous studies that have used similar threshold methodologies in fuzzy Delphi and multi-criteria decision-making models, ensuring that the selected indicators are statistically significant and relevant for the study’s objectives.

As shown in Table 5, 22 indicators were excluded, and 31 evaluation indicators were retained. The excluded indicators were primarily due to the following reasons: mismatch with the applicable context of architectural outdoor spaces, high implementation and maintenance costs, challenges in data collection, and insufficient direct relevance to health promotion. Excluding these indicators ensures a more focused analysis of factors that directly affect residents’ health, comfort, and social interaction.

4.4. Phase 2: Calculate the Weight Using AHP

In this stage, we employed the Analytic Hierarchy Process (AHP) to calculate the weights of the selected indicators. We constructed a pairwise comparison matrix based on expert opinions and asked the ten experts involved in the first stage to assign quantitative ratings using the fuzzy language scale. Using the expert ratings, we built a comparison judgment matrix and applied the root method to calculate the approximate values of the matrix.

To ensure consistency in the expert ratings, we conducted consistency checks on the judgment matrices at each level. As detailed in Section 3 (Research methodology), we performed the Consistency Ratio (C.R.) test to evaluate the logical consistency of the expert comparisons. According to Saaty’s method, the C.R. should be below 0.1 for the pairwise comparisons to be considered consistent. In this study, all C.R. values were found to be below 0.1, indicating satisfactory consistency in the expert assessments. For additional assurance of consistency, we also calculated the Consistency Index (C.I.), as outlined in the Research Methodology section, and checked that it was within acceptable limits. The C.I. was used as an alternative consistency check to ensure the reliability of the pairwise comparison matrix. In cases where C.R. values exceeded 0.1 (although no such instances occurred in this study), we would have revisited the pairwise comparisons with the experts. Experts would have been prompted to revise their assessments, and the matrix would have been adjusted accordingly to improve consistency.

After conducting the consistency checks, all C.R. values remained below 0.1, and the CI values confirmed the logical consistency of the expert judgments. The final calculated weights and the results of the consistency checks are shown in

Table 6. Weight of health assessment indicators.

Primary Indicator	Weight	Secondary Indicator	Weight	Tertiary Indicator	Weight
Comfort (A1)	17.67%	Air (B1)	4.94%	Air quality (C1)	2.86%
				Air humidity (C2)	2.08%
		$\mathrm{C}.\mathrm{R}.=0.0774<0.1,{\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}=6.4875$
		Water (B2)	4.85%	Drinking water quality (C3)	3.39%
				Rainwater and wastewater discharge systems (C4)	1.45%
		$\mathrm{C}.\mathrm{R}.=0.0579<0.1,{\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}=4.5198$
		Thermal (B3)	4.59%	Shaded area ratio (C5)	2.57%
				Vertical greening area (C6)	2.02%
		$\mathrm{C}.\mathrm{R}.=0.0348<0.1,{\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}=5.1852$
		Sound (B4)	5.20%	Site noise level (C7)	2.65%
				Green buffer ratio between site and urban roads (C8)	2.55%
		$\mathrm{C}.\mathrm{R}.=0.0218<0.1,{\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}=2.0589$
Diversity (A2)	19.59%	Activity facilities (B5)	12.94%	Area of exercise and fitness spaces for the elderly (C9)	3.83%
				Area of playgrounds and sports facilities for children (C10)	3.72%
				Number of rest facilities (C11)	2.83%
				Length and continuity of fitness trails (C12)	2.56%
		$\mathrm{C}.\mathrm{R}.=0.0514<0.1,{\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}=3.0585$
		Site layout (B6)	8.36%	Proportion of ground-level space elevated (C13)	2.24%
				Mix of land uses (C14)	3.21%
				Rooftop space utilization rate (C15)	2.91%
		$\mathrm{C}.\mathrm{R}.=0.0049<0.1,{\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}=3.0000$
Safety (A3)	27.88%	Natural environment (B7)	9.04%	Plant non-toxicity (C16)	4.78%
				Extreme climate contingency plan (C17)	4.27%
		$\mathrm{C}.\mathrm{R}.=0.0172<0.1,{\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}=3.0427$
		Artificial environment (B8)	16.33%	Slip resistance of pavements (C18)	4.51%
				Use of non-toxic materials (C19)	4.39%
				Accessibility facilities (C20)	3.77%
				Coverage rate of security systems (C21)	3.66%
		$\mathrm{C}.\mathrm{R}.=0.0415<0.1,{\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}=8.1484$
Ecological (A4)	17.29%	Natural landscape (B9)	9.97%	Plant diversity (C22)	3.64%
				Greening rate (C23)	3.40%
				Green visibility rate (C24)	2.93%
		$\mathrm{C}.\mathrm{R}.=0.0121<0.1,{\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}=2.9172$
		Artificial landscape (B10)	7.32%	Landscape interactivity (C25)	2.73%
				Landscape maintenance condition (C26)	4.59%
		$\mathrm{C}.\mathrm{R}.=0.0279<0.1,{\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}=2.9853$
Humanities (A5)	17.57%	Culture (B11)	9.08%	Display of local culture (C27)	3.32%
				Elderly and child-friendly care (C28)	2.93%
				Number of participants in activities promoting interaction and integration (C29)	2.83%
		$\mathrm{C}.\mathrm{R}.=0.0751<0.1,{\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}=3.5128$
		Management (B12)	7.49%	Frequency of organized activities (C30)	3.25%
				Intensity of management supervision (C31)	4.24%
		$\mathrm{C}.\mathrm{R}.=0.0941<0.1,{\mathrm{\lambda }}_{\mathrm{m}\mathrm{a}\mathrm{x}}=2.0899$

.

4.5. Phase 3: Establishment of Multi-Level Fuzzy Comprehensive Evaluation Model

Based on the architectural outdoor space health evaluation indicators developed in the first and second stages, this study classifies the indicators into two categories, qualitative and quantitative, and calculates their respective membership degrees. The detailed evaluation criteria for each indicator are provided below.

4.5.1. Quantitative Indicators

The health evaluation framework for outdoor architectural spaces proposed in this study includes a total of 21 quantitative indicators, with specific evaluation standards detailed in

Table 7. Quantitative indicators.

Quantitative Indicators	Evaluation Criteria
	Poor (1 Point)	Fair (2 Points)	Good (3 Points)	Excellent (4 Points)
C1: Air quality (Unit: AQI)	>200	(100,200]	(50,100]	(0,50]
C2: Air humidity (Unit: relative humidity, percent)	[0,20) or (80,100]	[20,30) or (70,80]	[30,40) or (60,70]	[40,60]
C3: Drinking water quality (Unit: amount of residual chlorine, mg/L)	[0,0.1) or >1.0	[0.1,0.2) or (0.7,1]	[0.2,0.3) or (0.5,0.7]	[0.3,0.5]
C5: Shaded area ratio (Unit: percent)	[0,60)	[60,75)	[75,90)	[90,100]
C6: Vertical greening area (Unit: percent)	[0,10)	[10,30)	[30,50)	[50,100]
C7: Site noise level (Unit: noise level, dB)	>75	(63,75]	(50,63]	[0,50]
C8: Green buffer ratio between site and urban roads (Unit: percent)	[0,25)	[25,50)	[50,75)	[75,100]
C9: Area of exercise and fitness spaces for the elderly (Unit: the total area as a percentage of the total area of the community, percent)	[0,15)	[15,30)	[30,45)	≥45
C10: Area of playgrounds and sports facilities for children (Unit: the total area as a percentage of the total area of the community, percent)	[0,15)	[15,30)	[30,45)	≥45
C11: Number of rest facilities (Unit: sites per hectare)	[0,3)	[3,5)	[5,8)	≥8
C12: Length of fitness trails (Unit: meters)	[0,300)	[300,500)	[500,1000)	≥1000
C13: Proportion of ground-level space elevated (Unit: the proportion of the area of the overhead layer to the total area of the plot, percent)	[0,15)	[15,25)	[25,40)	[40,100]
C14: Mix of land uses (Unit: the total mix as u percentage of the total mix of the community, percent)	[0,25)	[25,50)	[50,75)	[75,100]
C15: Rooftop space utilization rate (Unit: percent)	[0,25)	[25,50)	[50,75)	[75,100]
C18: Slip resistance of pavements (Unit: the proportion of non-slip pavement in the total pavement, percent)	[0,50)	[50,65)	[65,80)	[80,100]
C21: Coverage rate of security systems (Unit: percent)	[0,50)	[50,70)	[70,90)	[90,100]
C22: Plant diversity (Unit: plant species, species)	[0,20)	[20,30)	[30,40)	>40
C23: Greening rate (Unit: relative rate, percent)	[0,20)	[20,30)	[30,40)	[40,100]
C24: Green visibility rate (Unit: the proportion of green vegetation in the field of view, percent)	[0,20)	[20,30)	[30,40)	[40,100]
C29: Number of participants in activities promoting interaction and integration (Unit: number)	[0,25)	[25,50)	[50,75)	[75,100]
C30: Frequency of organized activities (Unit: frequency, times/month)	[0,2)	[2,5)	[5,8)	>8

. These indicators cover various aspects such as air quality, water quality, noise levels, and greenery coverage, aiming to comprehensively assess the environmental quality of public spaces and their role in promoting health. Each indicator has clear evaluation standards set, and the actual performance of each indicator is reflected through graded scoring. This evaluation framework helps monitor and improve the existing public space environment and provides a scientific basis for future urban planning and design.

4.5.2. Qualitative Indicators

The outdoor space health evaluation framework for buildings proposed in this study includes a total of ten qualitative indicators, as detailed in

Table 8. Qualitative indicators.

Qualitative Indicators	Evaluation Criteria	Score
Rainwater and wastewater discharge systems (C4)	Highly efficient rainwater collection and utilization system that effectively controls runoff and reduces flooding risk.	4 points
	Basic rainwater collection and utilization present, controlling partial runoff.	3 points
	Existing but inefficient rainwater management systems, insufficient runoff control.	2 points
	Poor rainwater management, frequent occurrences of waterlogging or flooding.	1 point
Plant non-toxicity (C16)	Exclusive use of non-toxic plants, ensuring environmental and human safety.	4 points
	Mostly non-toxic plant species, with minor exceptions that may have slight toxicity.	3 points
	Some toxic plants used, but with warning signs.	2 points
	Extensive use of toxic plants without warnings.	1 point
Extreme climate contingency plan (C17)	Comprehensive contingency plans for extreme weather, with regular drills and effective responses.	4 points
	Basic measures in place, infrequent drills.	3 points
	Initial plans for extreme weather, average implementation effectiveness.	2 points
	Lack of contingency plans for extreme climates.	1 point
Use of non-toxic materials (C19)	Sole use of non-toxic, environmentally friendly materials.	4 points
	Primarily non-toxic materials used, with some traditional materials in minor areas.	3 points
	Partial use of non-toxic materials, still reliant on some hazardous materials.	2 points
	Widespread use of hazardous materials.	1 point
Accessibility facilities (C20)	Comprehensive accessibility design, facilitating easy use of facilities.	4 points
	Basic accessibility facilities, with minor areas of inaccessibility.	3 points
	Limited accessibility facilities, inconvenient for disabled individuals.	2 points
	Lack of accessibility facilities.	1 point
Landscape interactivity (C25)	Rich interactive elements that encourage public participation.	4 points
	Moderate interactivity, but limited in scope or type.	3 points
	Few interactive elements, primarily for viewing.	2 points
	No interactive elements, single-function.	1 point
Landscape maintenance condition (C26)	Well-maintained landscape with regular upkeep.	4 points
	Generally well-maintained with occasional neglect.	3 points
	Insufficient maintenance, signs of landscape degradation.	2 points
	Almost no maintenance, severe landscape degradation.	1 point
Display of local culture (C27)	In-depth display of local cultural characteristics, educational.	4 points
	Displays local culture to a certain extent, but lacks depth and breadth.	3 points
	Occasionally references local culture, no in-depth display.	2 points
	Neglects local culture, devoid of cultural elements.	1 point
Elderly and child-friendly care (C28)	Facilities well-designed to cater to the needs of the elderly and children.	4 points
	Partial consideration of the needs of the elderly and children, incomplete facilities.	3 points
	Occasional consideration of special needs, inadequate facility support.	2 points
	Neglect of the needs of the elderly and children.	1 point
Intensity of management supervision (C31)	Dedicated supervision and maintenance teams, regular inspections and upkeep.	4 points
	Basic supervision and maintenance, occasional oversights.	3 points
	Insufficient supervision and maintenance, affecting facility use.	2 points
	Almost no supervision and maintenance, poor facility condition.	1 point

. Each indicator is divided into four levels, with a score range of 1–4 points. These indicators include aspects such as rainwater and wastewater management and plant non-toxicity. The evaluation of qualitative indicators was mainly conducted using questionnaires, on-site surveys, and interviews, collecting users’ subjective perceptions of the building’s outdoor space and scoring them to convert these subjective indicators, which cannot be quantitatively assessed, into quantitative data. This study collected the subjective opinions of users of the outdoor space of the Wuhan Citizen’s Home building, with a total of 50 participants, including staff of the Citizen’s Home, visiting public, and nearby residents.

4.6. Fuzzy Comprehensive Evaluation

This study categorizes and statistically analyzes the membership degrees of quantitative and qualitative indicators. First, the comment set V for Wuhan Citizen Home is determined; second, on-site research is conducted to obtain various parameters within the site to determine the membership degrees of quantitative indicators. The methods of on-site research include, but are not limited to, the following: use a wet and dry bulb thermometer to measure air temperature and humidity; use a water quality analyzer to measure the pH value, dissolved oxygen, and turbidity of drinking water; use a fisheye camera to take images and calculate the green view index; and use a drone to take aerial images and calculate the greening coverage rate. By distributing questionnaires on-site, we collected users’ subjective opinions on qualitative indicators. A total of 50 questionnaires were distributed in this study, with 48 valid responses, resulting in a validity rate of 96%. We then compiled the membership degree sets for each indicator, constructed the membership degree matrix for the basic indicator factors in the evaluation index system, and calculated the measurement scale. Finally, based on the relative weights calculated using measurement scales and the Analytic Hierarchy Process, the comprehensive evaluation score for the Wuhan Citizen Home was obtained.

The final comprehensive evaluation score for the Wuhan Citizen’s Home is 2.505 points, categorized as good, indicating that the Wuhan Citizen’s Home effectively promotes social interaction and enhances the function of the building’s outdoor public spaces, with a noticeable positive impact.

Among the primary indicators, A3 Safety scored the highest at 2.972 points, while A5 Humanities scored the lowest at 1.897 points, as shown in Figure 5. In the A1 Comfort dimension, B2 Water scored the highest at 3.347 points, while B3 Thermal scored the lowest at 2.502 points. In the tertiary indicators, the C8 Green buffer ratio between the site and urban roads scored the highest at 3.9 points, while the C6 Vertical greening area scored the lowest at 1 point.

At the A2 Diversity level, B6 scored higher at 2.281 points, while B5 scored relatively lower at 1.903 points.

In the tertiary indicators, C14 Mix of land uses scored the highest at 3.82 points, while C15 Rooftop space utilization rate scored the lowest at 1.26 points.

In the A3 Safety dimension, B8 Artificial environment scored higher, with a score of 3.196, while B7 Natural environment scored lower, with a score of 2.566. In the tertiary indicators, the C21 Coverage rate of security systems scored the highest, with 4 points, while the C1 Extreme climate contingency plan scored the lowest, with 1.783 points.

Ecological dimension, B9 Natural Landscape scored higher at 2.854 points, while B1 Artificial Landscape scored lower at 1.948 points. In the tertiary indicators, the C23 Greening rate scored the highest at 3.371 points, while C25 Landscape Interactivity scored the lowest at 1.88 points.

At the A5 Humanities level, B1 Management scored higher, with a score of 2.234, while B11 Culture scored lower, with a score of 1.62. In the tertiary indicators, C31 Intensity of management supervision scored the highest at 2.566 points, while C27 Display of local culture scored the lowest at 1.53 points.

5. Discussion

The indicator weights in this study were determined using the Analytic Hierarchy Process (AHP). These weights reflect the relative importance of each design factor in promoting health within architectural outdoor spaces. Notably, there is a significant difference in the weights assigned to various categories. This difference stems from the specific context of the study and the expert consensus, rather than from empirical validation alone. The specific indicator weights are shown in Figure 6.

The higher weight assigned to the Safety category is attributed to the fundamental role that safety plays in ensuring the well-being of users in outdoor spaces. In high-density urban environments, where outdoor spaces are often shared by large numbers of people, safety is a primary concern. Factors such as slip resistance of pavements, pedestrian-vehicle segregation, and use of non-toxic materials were all considered essential for reducing risks and preventing accidents, thus directly contributing to the health of individuals. Experts agreed that inadequate safety measures could significantly undermine the potential health benefits of outdoor spaces, thereby justifying the higher weight given to this category.

In contrast, the Comfort category, while important, was regarded as secondary to safety in terms of immediate health impacts. Design factors such as air quality, thermal comfort, and noise control contribute to the physical and psychological well-being of users, but they do not carry the same level of urgency as safety-related factors. As such, Comfort received a lower weight relative to Safety.

The weights assigned to each category were based on the collective judgment of ten experts with expertise in architecture, urban planning, and healthcare. The use of the fuzzy Delphi method (FDM) in the initial phase, followed by AHP, allowed for a structured approach to synthesizing expert opinions. This process captured both the relative importance of each design factor and the specific context in which the outdoor spaces are situated, ensuring that the weightings align with the goals of health promotion in urban environments.

While expert consensus formed the foundation for these weight assignments, it is important to note that the weights are context-dependent. In this study, the focus on high-density urban environments, where safety concerns are paramount, naturally led to higher weights for safety-related factors. Further empirical validation of the weights could enhance the robustness of the model, but the current methodology provides a sound basis for evaluating the health impacts of architectural outdoor spaces based on expert-informed priorities.

Specifically, among the five primary indicators, four indicators fall into the “Good” category, with the A3 Safety indicator having the highest score of 2.97. The A5 Humanities indicator has the lowest membership degree, with only 1.9 points, falling into the “Fair” category. This indicates that Wuhan Citizen Home places more emphasis on ensuring the safety of residents and users from both natural and artificial environments, but neglects the construction and care of people’s spiritual well-being. In the secondary indicators of the evaluation framework proposed in this study, three indicators belong to the “Excellent” category, six indicators belong to the “Good” category, and three indicators belong to the “Fair” category. In the tertiary indicators of the evaluation framework proposed in this study, there are eleven indicators each in the “Good” and “Fair” categories and nine indicators in the “Excellent” category.

According to the above results, it can be seen that the Wuhan Citizen’s Home has already shown good performance in terms of infrastructure, but there is still room for improvement in quality enhancement and spiritual civilization construction. In terms of comfort, the vertical greening area on the facade of Wuhan Citizen’s Home is insufficient. To enhance the thermal comfort of the site, vertical greening can be planted on the building’s facade. This measure can reduce temperature fluctuations on the building’s surface, lower energy consumption inside the building, and improve air quality on the site, as plants absorb carbon dioxide and release oxygen through photosynthesis.

In terms of diversity support, there is a lack of sufficient space for elderly exercise and fitness facilities, no elevated ground floors in buildings, and low utilization of building rooftops. The facilities should include dedicated areas suitable for elderly exercise methods and low-intensity high-safety fitness equipment. In subsequent renovations, consider installing a raised ground floor. In modern architectural design, the raised ground floor is a commonly adopted design strategy that significantly enhances the comfort and health of the living environment. First, the elevated ground floor improves the building’s natural ventilation, helping to reduce indoor humidity and prevent microbial growth, thereby decreasing the risk of respiratory diseases among residents. Second, this design effectively isolates ground heat or cold, improving indoor temperature control and enhancing thermal comfort within the site, reducing dependence on energy, and lowering environmental pressure. At the community level, the design of open spaces in the ground-level stilted areas not only provides necessary recreational and social venues, but also promotes interaction and a healthy lifestyle within the community. Proper planning of these ground-level stilted areas can also increase the introduction of natural light, improve the indoor lighting environment, and have a positive impact on residents’ visual health and psychological well-being. Increasing the utilization of rooftop spaces in urban environments also has multiple health benefits. By implementing rooftop greening, urban air quality can be significantly improved, and noise pollution can be reduced while providing crucial ecological habitats for urban biodiversity. Additionally, rooftop gardens and green areas offer essential contact with nature, helping to alleviate residents’ psychological stress and improve mental health. Rooftop spaces, as venues for exercise and recreation, can encourage residents to engage in physical activities, thereby reducing the risk of chronic diseases and enhancing their quality of life.

In terms of venue safety, Wuhan Citizen Home’s extreme weather contingency plans are insufficient. In the context of climate change, extreme weather contingency plans play a crucial role in safeguarding public health. By planning ahead and implementing comprehensive preventive measures, these plans can significantly reduce the impact of heatwaves, floods, storms, and other extreme weather events on human health. In particular, by establishing heat relief centers and optimizing emergency shelters, extreme weather contingency plans help lower mortality and morbidity rates caused by high temperatures and natural disasters. Furthermore, these plans effectively address the increased risk of respiratory diseases caused by air pollution by enhancing air quality monitoring and public health education.

In terms of site ecology, Wuhan Citizen’s Home performs well in natural landscapes, but has significant room for improvement in artificial landscapes. Firstly, the current site has poor interactivity in artificial landscapes. The interactivity of artificial landscapes plays an important role in enhancing residents’ health, especially in urban environments. Specifically, designing highly interactive artificial landscapes can significantly increase residents’ physical activity levels, thereby reducing the risk of chronic diseases and improving overall quality of life. Moreover, highly interactive landscape design can also promote mental health by providing a relaxing and stress-relieving environment, thereby reducing the incidence of anxiety and depression. Additionally, due to poor management, there is insufficient maintenance of the landscape within the site, resulting in some artificial landscapes being damaged. The site management agency should increase maintenance efforts to ensure that the landscape within the site can be used normally.

In terms of the site’s cultural aspects, the Wuhan Citizen Home is not fully completed, especially in the cultural section, which lacks landscape structures that can showcase Wuhan’s Jingchu culture. Additionally, in creating an overall atmosphere, the Wuhan Citizen Home lacks care for the elderly and children, and there is also room for improvement in the frequency of organized activities. To enhance the cultural and health promotion aspects of the Wuhan Citizen’s Home, it is recommended to adopt a series of strategies in future venue updates to strengthen cultural displays, improve the environmental atmosphere, and enhance the effectiveness of event organization. Firstly, by constructing distinctive landscape structures and themed exhibitions that showcase Jingchu culture, cultural heritage can be strengthened and visitors’ educational experiences can be enhanced. Secondly, designing accessible facilities for all age groups and ability levels will help create an inclusive environment suitable for both the elderly and children. Increasing the diversity and frequency of health promotion activities, such as sports events and health education seminars, can effectively enhance public health awareness and participation.

The framework’s universality lies in its adaptability to diverse urban contexts beyond Wuhan, such as compact Asian or sprawling Western cities, due to its standardized methodology. It quantifies health impacts of small-scale spaces, offering a scalable tool for global urban design. However, its single-case focus limits generalizability, and expert-driven weights may miss site-specific nuances. Future work should expand to multi-site studies, incorporate user data (e.g., via sensors), and refine indicators for broader applicability, enhancing precision and practical utility in health-oriented planning.

6. Conclusions

In summary, this study proposes and applies a multi-level fuzzy comprehensive evaluation model to assess the healthiness of outdoor architectural spaces, using an outdoor space of a building in Wuhan as an example for verification. Practical evidence indicates that the multi-level fuzzy comprehensive evaluation model established in this study not only has theoretical significance, but is also applicable for evaluating the healthiness of outdoor architectural spaces.

The findings translate into actionable urban planning guidelines. Safety, scoring highest (2.97), underscores the need for policies mandating slip-resistant surfaces, pedestrian-vehicle segregation, and robust extreme weather plans in high-density cities. These measures reduce risks and climate-related health threats like heatwaves, vital for crowded urban environments. Similarly, ecology gaps highlight the value of requiring green facades and rooftops in building codes, improving air quality, mitigating heat islands, and enhancing mental health—strategies scalable to Asian megacities with limited land.

This study uses the Wuhan Citizens’ Home as a case to validate the multi-level fuzzy comprehensive evaluation model. The results demonstrate that the model effectively assesses the health performance of architectural outdoor spaces, with the Wuhan Citizens’ Home achieving a comprehensive evaluation membership degree of 2.505, categorized as “Good”. The evaluation highlights strong performance in comfort, diversity, safety, and ecology, especially regarding noise levels, security coverage, plant non-toxicity, and functional integration. However, there is room for improvement in the “humanity” aspect, particularly in reflecting local culture and enhancing age-friendly design. Additionally, the frequency of activities promoting social and mental well-being is insufficient and requires attention.

From a practical standpoint, this evaluation framework is expected to be incorporated into the standard procedures of urban planning and design, becoming a core tool for evaluating new or renovated projects. With its standardized evaluation process and indicators, it can be applied across diverse urban environments, irrespective of their scale or type, driving the development of more health-focused urban projects.

This study employs the Fuzzy Delphi method and Analytic Hierarchy Process (AHP) to establish a multi-level fuzzy comprehensive evaluation model for assessing the health performance of urban architectural outdoor spaces. The model was applied to the Wuhan Citizens’ Home, uncovering several deficiencies in the space’s health performance. However, this study does have certain limitations. Due to time constraints, the evaluation model was applied to a single case study. Although multiple sites were compared and selected during the process, to avoid the uniqueness of the research subject, future research should aim to expand the sample size, improve data quality, and optimize the evaluation model to increase the accuracy and practical applicability of the evaluation framework.