Healing-Oriented Street Space Model: A Multidisciplinary Multi-Stakeholder Approach for High-Density Cities

Abstract

In the 21st century, rapid urban development during global urbanization has led to high-density environments. These settings have become a significant cause of stress-related health problems for residents. Healing street design plays an important role in helping address mental health challenges caused by this process. Current research often focuses on healing elements and methods from only a single field. As a result, it lacks the integration of multidisciplinary and multi-stakeholder perspectives. To address this gap, this paper formed a Delphi expert panel with multidisciplinary scholars, urban managers, and practicing designers. The panel developed a quantitative evaluation model. This model covers four core dimensions: Safety (0.3210), Attractiveness (0.1080), Friendliness (0.2155), and Comfort (0.3553). It also includes eleven healing elements, such as Pedestrian Right-of-Way (0.4131), Night Lighting (0.3209), Visual Landscape (0.759), Street Furniture (0.4000), and Street Scale (0.3274). Using this model, the healing potential of Jingliu Road in Zhengzhou was assessed. The analysis identified the overall healing potential, core healing dimensions, and shortcomings of the street. This finding provides a clear direction for future healing-oriented street design. This paper builds a healing system for pedestrian spaces in high-density urban streets in China. It thus offers an evidence-based scientific tool for environmental design. Healing environments have expanded from less accessible spaces, such as squares and parks, to interactive and accessible streets. This transition enhances urban spaces’ capacity to address residents’ mental health concerns and promotes public health. Additionally, this paper offers specific recommendations for planners and policymakers to prioritize healing elements in urban renewal projects.

1. Introduction

The 21st-century urbanization has transformed global living patterns, with high-density urban environments emerging as a double-edged sword, simultaneously driving economic vitality while exacerbating unresolved public health challenges. Despite extensive research on urban development, the association between high-density settings and residents’ stress-related health outcomes remains underexplored, particularly in rapidly urbanizing regions like China. Yet, this rapid urban expansion has exacerbated environmental pollution, traffic congestion, and social isolation, which in turn pose significant threats to residents’ physical and mental well-being [1,2,3]. Specifically, high-density urban environments characterized by spatial compression, noise pollution, and limited green space directly contribute to increased risks of emotional disorders, anxiety [4,5], and even severe psychological conditions [6].

Moreover, the effects of urban living on mental health are long-term and intricately connected to the built environment. A study published in The Lancet Psychiatry indicates that urbanization is associated with an increased risk of mental health issues [7], particularly among young people [8,9]. Narrowed living spaces and highly crowded public places act as specific stressors [10], leading to phenomena such as “nature deficit disorder” due to a lack of exposure to natural environments [11]. Consequently, addressing these mental health challenges requires a paradigm shift towards enhancing the restorative and healing functions of urban environments.

1.1. From “Healthy Streets” to “Healing Streets”

The concept of healing streets is derived from healthy streets. Healthy streets encompass three aspects: promoting physical health, mental health, and public health. Key strategies include optimizing the physical environment, facilitating physical activity, reducing environmental cognitive load, enhancing residents’ sense of security and belonging, fostering social interaction and equity, and leveraging the health-supporting functions of diverse street components [12]. Since 2010, renewal projects centered around the concept of healthy streets have been promoted and applied as policies in countries such as the US, Canada, and the UK. The healthy streets approach primarily focuses on physiological health through walkability and physical safety. In contrast, healing streets not only meet the physical space requirements of healthy streets but also highlight the mechanism of environmental psychology in transitioning from “physical comfort” to “psychological healing”. In high-density urban streets, micro-scale continuous landscapes and water features help reduce cortisol levels [13,14], while visual elements and interactive installations enhance social engagement and community belonging. These streets are designed to promote mental health through evidence-based spatial interventions, offering residents opportunities for psychological relaxation, emotional connection, emotional calmness, and a sense of hope and vitality [15,16,17,18]. As the most frequently used public space in high-density cities, streets provide unique daily healing experiences for residents.

1.2. Theoretical Foundations and Policy Context

The design of healing environments has deep historical roots, from ancient temples, gardens, and natural sanctuaries to Florence Nightingale’s advocacy for fresh air and greenery in 19th-century healthcare architecture [19,20,21]. In modern urban design, these practices are grounded in rigorous scientific frameworks, specifically Stress Recovery Theory (SRT) [22] and Attention Restoration Theory (ART) [23]. SRT posits that visual engagement with natural features facilitates recovery from physiological stress, while ART suggests that environments characterized by “soft fascination” help replenish depleted cognitive resources.

While these theories were initially developed for parks and healthcare landscapes, their systematic translation into everyday street design requires further development. Streets offer only fragmented, micro-scale contact with nature compared to parks, and their ubiquity positions them as critical intervention points. Studies have demonstrated that even limited exposure to street greenery, bird sounds, and human-scale proportions can significantly reduce stress [13,14,24,25,26,27,28,29,30]. However, translating these psychological principles into systematic street design criteria remains a challenge.

This need aligns with contemporary policy frameworks. The United Nations Sustainable Development Goals (SDGs), particularly Goal 3 (Good Health and Well-being) and Goal 11 (Sustainable Cities and Communities), emphasize the need for inclusive urban environments that promote both physical and mental health [31,32]. Similarly, China’s “Healthy China 2030” initiative prioritizes the development of walkable, health-promoting public spaces [33]. These policy commitments create an urgent need for evidence-based tools to guide healing street design under resource constraints.

1.3. Research Gaps and Study Objectives

Despite growing policy support and theoretical foundations, challenges remain in effectively translating these principles into healing street design practice.

First, systematic frameworks that turn psychological theories into measurable design parameters remain underdeveloped. Existing tools, like the Healthy Streets Indicators, offer useful guidance but often lack precise thresholds—for example, optimal greenery density or width-to-height ratios for safety perception. As a result, the current design often relies on subjective judgment, limiting consistent use of empirical evidence.

Second, a few assessment approaches use weighting mechanisms to prioritize interventions when resources are limited. In densely populated urban renewal areas, planners must make hard choices. Should they prioritize pavement continuity or focus on visual landscaping? Without a scientific prioritization system that shows the order of healing needs, decision making is often fragmented or ineffective. Unweighted checklists do not account for the relative impact of design elements.

This study aims to address existing gaps by developing a weighted assessment model that integrates principles of environmental psychology into healing street design for high-density cities. The model uses Delphi expert consultations to identify quantitative “healing elements” and the Analytic Hierarchy Process (AHP) to weight their importance. The model features a dual-validation mechanism that combines “healing potential” and “environmental quality” to verify therapeutic effects. “Healing elements” identified via Delphi consultations represent the objective healing potential of street space, while resident questionnaires provide feedback on subjective environmental quality. This two-way verification prevents the model from being just an “environmental quality checklist” and enables proactive improvement by optimizing potential elements. The model is used for both pre-intervention evaluation and post-occupancy assessment. Empirical validation is conducted in Zhengzhou, China. Since the 2019 walkability-focused renewal of Jingliu Road in Zhengzhou, no systematic evaluation of its impact has been conducted. The model can assess the quality of street renewal and its healing potential, helping upgrade spaces from “walk-friendly” to healing-oriented.

2. Materials and Methods

This research integrates the Delphi method with the AHP system. Through a multi-stage, progressive design of “forming a Delphi expert group–screening healing indicators through multiple rounds of consultation–quantifying weights and building a model-validating with an empirical questionnaire”, it forms a complete closed loop from theoretical construction to practical application, as shown in Figure 1. This approach ensures the scientific nature of indicator screening, strengthens the rigor of the model, and provides systematic methodological support for evaluating the street pedestrian environment.

2.1. The Process of Expert Selection

This research ensures scientific validity and practical relevance by including diverse professional perspectives. The expert panel includes researchers, design practice experts, and policy management experts from multiple disciplines.

Multidisciplinary researchers should focus on urban environments, environmental remediation, and environmental psychology, drawing on interdisciplinary backgrounds. Preference is given to scholars who have conducted in-depth research in high-density urban environments and public health, as this enables robust analysis of how street spaces impact human physical and mental health from perspectives such as architecture and environmental science.

Design practice experts should have practical experience in high-density urban projects. Areas include urban planning, architectural design, landscape design, and urban renewal. Priority is given to practice duration, project scale, and implementation effects. Experts must systematically understand innovative design and practical implementation. Emphasis should be on balancing humanistic needs and space efficiency in complex urban environments.

Policy management experts are administrators with experience in urban planning, policy formulation, and implementation. They must understand urban logic and the implementation of governance policies. Selection emphasizes broad urban planning management, diverse policy project involvement, and skill in interpreting public policy. Experts must recommend practical ways to implement street healing environments, including policy design, regulatory enforcement, and resource allocation.

These three types of experts form a closed-loop of “theoretical support–practical implementation–policy coordination” to ensure the academic depth of the research conclusions. By combining design experience with policy perspectives, this research enhances the model’s practical application value and provides comprehensive guidance for optimizing the healing environment of urban streets.

To ensure practical, targeted input, this study used the Expert Familiarity Coefficient (Cs) to select expert panels. As shown in

Table 1. Five-level scoring system for expert familiarity.

Very Familiar	Familiar	Generally Familiar	Slightly Familiar	Unfamiliar
1	0.8	0.6	0.4	0.2

, Cs is ranked in five levels [34]: “very familiar” (1), “familiar” (0.8), “generally familiar” (0.6), “slightly familiar” (0.4), and “unfamiliar” (0.2).

Eighteen experts were invited to participate in the study. Among them, five experts selected the “very familiar” option, seven experts selected familiar, two selected generally familiar, two selected slightly familiar, and two selected unfamiliar. After excluding those who were “slightly familiar” and “unfamiliar,” fourteen experts were selected for the Delphi consultation.

Meanwhile, this study utilizes the expert participation rate and Expert Authority Coefficient to validate and refine the expert team. The expert participation rate is a crucial indicator that clearly influences the quality of research outcomes. High participation not only reflects experts’ interest but also directly impacts the reliability of collected data by ensuring surveys are completed thoroughly, thus mitigating missing or biased data.

Experts received the distribution plan for the questionnaire before participating.

Table 2. Expert participation rate of three rounds.

	Number of Questionnaires	Recycling Quantity	Expert Participation Rate
First Round	14	14	100%
Second Round	14	14	100%
Third Round	14	14	100%

shows a 100% expert participation rate.

This paper uses the Expert Authority Coefficient ($Cr$) to verify the scientificity and accuracy of expert data. The formula for $\mathrm{C}\mathrm{r}$ is

$$Cr=(Ca+Cs)/2$$

(1)

where $Ca$ represents the coefficient of Judgment Basis based on theoretical analysis, practical experience, peer understanding, and intuition (values are shown in

Table 3. Judgment Basis and scoring system.

Judgment Basis	Influence Degree
	Big	Middle	Small
Theoretical Analysis	0.5	0.4	0.3
Practical Experience	0.3	0.2	0.1
Peer Understanding	0.1	0.1	0.1
Intuition	0.1	0.1	0.1

).

Utilizing statistical methods, the calculated $Cr$ value exceeding 0.7 indicates a significant level of expert consensus, which contributes to the reliability of the research findings, as detailed in

Table 4. Expert Authority Coefficient (Cr).

$\mathit{C}\mathit{a}$	$\mathit{C}\mathit{s}$	$\mathit{C}\mathit{r}$
0.907	0.606	0.756

.

2.2. The Process of Establishing a Healing Model

An expert team comprising multidisciplinary and multi-stakeholder members was established, and three rounds of Delphi consultation combined with AHP were employed to develop a context-specific restoration framework tailored to high-density urban streets in China.

The research process followed a systematic workflow: a comprehensive review of the existing literature on restorative environments was conducted, followed by on-site investigations on high-density streets in Zhengzhou. This approach served as the basis for an initial framework of healing elements. Three rounds of expert consultations were carried out to refine and validate the healing factors and dimensions, ensuring alignment with local urban contexts and expert consensus. AHP was applied to construct a pairwise comparison matrix, enabling the quantification of relative weights for each therapeutic element. The detailed workflow for this process is summarized visually in Figure 2.

Step 1: A comprehensive literature review on healing environments was conducted using CNKI, Scopus, and WOS. To ensure contextual relevance, researchers also used findings from on-site observations of high-density urban streets. Four dimensions and thirteen environmental elements were identified and classified, as shown in

Table 5. Healing element framework and references.

Dimensions	References	Elements	References
Safety	Jacobs, J. (1961) [35] Huo, D., Chen, F. & Chen, B. (2024) [36]	Pedestrian Right-of-Way	Chen, J., Zhang, Z., Long, Y. (2020) [37] Qi, Y.T. (2022) [17] Sheng, K., Liu, L., Zhou, X. (2024) [38]
		Proper Lighting	Zhang, N.W., Dun, S.T. (2023) [39]
		Convenience Shops and Cameras	Jacobs, J. (1961) [35]
		Community Identity Element	Chen, J., Zhang, Z., Long, Y. (2020) [37]
Attractiveness	Kaplan, S. (1995) [23] Ren, R. (2022) [40]	Transparency of the Architecture	Gehl, J. (2010) [41]
		Interactive Facilities	Chen, J., Zhang, Z., Long, Y. (2020) [37] Shen, Z.M. (2023) [42]
		Visual Art	Kaplan, S. (1995) [23] Zhang, Z., Peng, M., Li, Y. (2022) [43]
Friendliness	Wu, Z., Lei, H. (2022) [44]	Barrier-Free Paving	Tan, B. (2013) [45]
		Rest Facilities	Hunter, R. F., et al. (2019) [46]
		Humanity Care Facilities	Huo, D., Chen, F., Chen, B. (2024) [36]
Comfort	Yang, B., Wang, Q. (2018) [47] Chen, J., Zhang, Z., Long, Y. (2020) [37]	A Sense of Belonging	Shen, Z.M. (2023) [42]
		Street Greening	Ulrich, R.S. (1983) [22] Wu. R, Pan, Z.L., et al. (2021) [48] Wang, R.Y., et al. (2019) [49] Zhang et al. (2024) [50]
		Space Scale	Chen, J., Zhang, Z., Long, Y. (2020) [37] Gehl, J. (1987) [51]

.

Step 2: Conduct the first round of Delphi consultation. This includes the following: (1) Indicator screening: Experts evaluate the healing importance of four healing dimensions and thirteen factors, and may supplement or delete factors. This study used a five-point Likert scale (1: very unimportant, 2: unimportant, 3: neutral, 4: somewhat important, 5: very important). (2) Data analysis and questionnaire modification: SPSS Statistics 27 calculated the coefficient of variation (CV) and Kendall’s coefficient of concordance (Kendall’s W). (3) Revise the second-round questionnaire based on expert feedback and quantized data.

CV was used to quantify the dispersion of expert evaluation for a specific indicator. The formula for CV is

$$CV=\sigma /\mu \times 100\%$$

(2)

where σ denotes the standard deviation and μ denotes the mean.

A CV value near 0.1 shows experts’ ratings are highly consistent and concentrated, producing reliable results.

Kendall’s W evaluated expert consistency across indicators, supporting robust assessments. A value above 0.7 indicates high consistency, meaning expert consultation can end. Over three Delphi consultation rounds, Kendall’s W increased progressively and exceeded 0.7, as shown in

Table 6. The comparison of Kendall’s W in Delphi consultation with three rounds.

	N	Kendall’s W
First-round	14	0.51
Second-round	14	0.638
Three-round	14	0.714

. Analysis of CV and Kendall’s W objectively shows the degree of expert opinion concentration and consistency, strengthening research reliability and validity.

Step 3: In the second round, experts evaluated the 35 opinions from the first-round questionnaire as either “adopt” or “do not adopt” (see Figure 2 for the process). They reviewed the statistical results—including the mean, standard deviation, and CV—and rescored only those factors with CV above 0.1 to ensure reliability. The second-round questionnaire was developed from first-round results, aiming to stimulate critical thinking and promote consensus on controversial issues.

Step 4: The third-round questionnaire was designed, drawing on the findings from the second round. Schematic diagrams were used to visually display environmental factors, and the literature was used to provide background explanations of their origins. The questionnaire included the following components: (1) a restatement of the purpose of the round and the refined scoring guidelines; (2) the results of the second-round questionnaire, including mean values, CV, and Kendall’s W, reflecting the variability and consensus among responses; and (3) a targeted revision of items showing relatively high variability in the second round. The results of the third-round questionnaire indicated that all CV values were less than 0.1, and Kendall’s W was 0.714. When Kendall’s W exceeded 0.7, it indicated a high level of consistency among expert opinions, signaling the conclusion of the Delphi consultation. Through the three rounds of Delphi consultation, four core dimensions (Safety, Attractiveness, Friendliness, Comfort) and 11 healing factors were identified.

Step 5: Building on the results from Step 4 and the Delphi method findings, this study developed a three-level structure of healing factors for the street pedestrian environment, as shown in Figure 3.

For the target layer and criterion layers A1–A4, a judgment matrix was used to compare each pair of healing elements, establishing a pairwise comparison matrix as seen in

Table 7. The judgment matrix and weight value of the criterion layer A1–A4.

	A1	A2	A3	A4	Feature Vector	Weighting Values
A1	1	2	2	1	1.414	0.3210
A2	0.5	1	0.308	0.333	0.476	0.1080
A3	0.5	3.25	1	0.5	0.949	0.2155
A4	1	3	2	1	1.565	0.3553

. Using this method, pairwise comparison matrices were established for A11–A13, A21–A23, A31–A32, and A41–A42. The pairwise comparison matrix was normalized, and the weight vector of each element was computed using the eigenvector method.

To ensure logical consistency in the matrix, the CR value is calculated. A CR value less than 0.1 indicates satisfactory consistency. The weights obtained are assigned to each healing element, forming a prioritized ranking to support further research and decision making. The consistency index (CI) and consistency ratio (CR) were calculated for the judgment matrices at both the criterion layer and factor layer.

As shown in

Table 8. Consistency test results of the criterion layer.

The Biggest Characteristic Root	CI	RI	CR	Consistency Test
4.137	0.046	0.882	0.052	Pass

, the criterion layer’s maximum eigenvalue (λ_max) was 4.137, yielding a CI of 0.046. With a random index (RI) of 0.882 for a 4 × 4 matrix, the resulting CR value was 0.052—well below the 0.1 threshold—which confirms the consistency of the criterion layer matrix and validates the expert input.

At the factor layer, the consistency of each dimension’s sub-indicators was evaluated. For the Safety dimension, the CI and CR were 0.001, indicating high consistency. For the Comfort dimension, the CI was 0.027, and the CR was 0.051, meeting the required threshold. The Attractiveness and Friendliness dimensions both exhibited perfect consistency, as shown in

Table 9. Consistency test results of the factor layer.

	CI	RI	CR	Consistency Test
A1	0.001	0.525	0.001	Pass
A2	0.031	0.525	0.059	Pass
A3	0	0	0	Pass
A4	0.027	0.525	0.051	Pass

.

2.3. Empirical Research Design

Based on the healing evaluation model outlined in this paper, an empirical study of the healing potential assessment of Jingliu Road in Zhengzhou City was conducted. This model can verify the healing potential of the updated street projects and identify the advantages and shortcomings of the street environment in influencing residents’ healing experiences. Meanwhile, it provides evidence-based guidance for street-healing-oriented design.

Jingliu Road is situated in Jinshui District, Zhengzhou, Henan Province. Located in the core area of Zhengzhou’s main urban district, it extends from Huanghe Road (north) to Jinshui Road (south). The street lies within Shengwei Community, which is a densely populated residential neighborhood with a well-connected transportation network. Due to its strategic location and urban setting, Jingliu Road serves as a critical link between residential zones and diverse urban public service facilities, embodying the typical spatial structure of high-density urban streets in China. The base map in Figure 4 is a vector block map created by the author on the Mapbox platform (https://www.mapbox.com, accessed on 22 February 2026). Based on the layout of streets and buildings in the study area, it is generated by adjusting the display styles of roads, building blocks, and green spaces. This map aims to intuitively display the locations of the sample streets and the division method for the four street sections, providing a basic reference for subsequent analysis of the current situation.

Since 2019, the street has undergone phased renovations, resulting in four distinct zones. The primary focus was on Section 2–3, where, in 2019, a 1.1-m-wide “urban running track” was added to the west sidewalk. In 2023, this area was further upgraded with a rainbow fitness path, children’s play facilities, and leisure seats. Simultaneously, illegal buildings were demolished, sidewalk space was expanded, and a pocket park and reading promenade were constructed. Section 3–4 also completed its child-friendly renovation in 2023, adding similar amenities and benefiting from environmental remediation. French plane trees now line both sides of Jingliu Road, creating a shaded, slow-moving space. This study conducts segmented healing potential assessments to integrate evaluations across street sections, better capture the street’s overall healing potential, and identify key healing elements. This approach provides a reusable design logic and evidence-based decision support for healing-oriented urban street design.

Questionnaire data were collected on-site, enabling direct resident interaction and ensuring feedback grounded in real-life experiences. The questionnaire assessed satisfaction with eleven environmental elements. Meanwhile, on-site observations collected descriptive data on the street environment to help validate the questionnaire results.

The questionnaire followed three core principles: (1) Scale Design: A 5-point Likert scale (1 = Strongly Disagree, 5 = Strongly Agree) quantified respondents’ opinions; (2) Variables and Items: Eleven core variables, each measured by at least two items, helped reduce bias from misunderstanding or emotions; (3) Recovery and Screening: Of 260 questionnaires distributed, 231 were returned. Excluding incomplete or inconsistent responses, 210 valid questionnaires yielded an 80.8% response rate.

In this study, the questionnaire’s reliability was evaluated using Cronbach’s alpha. With a sample size of 210 and 29 items, the computed alpha was 0.904, as shown in

Table 10. Overall Cronbach’s alpha reliability coefficient of the questionnaire.

Sample Capacity	Items	Cronbach.α
210	29	0.904

. Cronbach’s alpha measures internal consistency, with values from 0 to 1. Generally, coefficients above 0.70 are acceptable; those exceeding 0.90 indicate excellent consistency.

Data analysis was conducted using Microsoft Excel, focusing on: (1) comparing healing effects across four street sections to identify performance disparities and assess Jingliu Road’s overall healing potential; (2) systematically identifying key healing elements and weak links to provide evidence-based directions for future improvement.

3. Results

3.1. A Healing Framework for High-Density Urban Streets

In this study, we imported the expert evaluation data into SPSS to compute the mean, standard deviation, and CV for each factor, as shown in

Table 11. The CV value of the first round. CVs above 0.2 are marked with an asterisk (*).

Dimensions	Factors	Mean	Standard Deviation	CV
Safety	Pedestrian Right-of-Way	4.6429	0.63332	0.136406125
	Proper Lighting	4.5	0.65044	0.144542222
	Convenience Shops and Cameras	3.7857	0.69929	0.18471881
	Community Identity Element	3.7143	0.82542	0.222227607 *
Attractiveness	Transparency of the Architecture	3.8571	0.77033	0.199717404
	Interactive Facilities	3.7143	0.61125	0.164566675
	Visual Art	4.3571	0.63332	0.145353561
Friendliness	Barrier-Free Paving	4.2143	0.69929	0.165932658
	Rest Facilities	3.7143	0.61125	0.164566675
	Humanity Care Facilities	2.9286	0.61573	0.210247217 *
Comfort	A Sense of Belonging	4.6429	0.49725	0.107099011
	Street Greening	4.8571	0.36314	0.074764777
	Space Scale	4.6429	0.49725	0.107099011

. Typically, the CV value exceeding 0.2 indicates a lack of consensus among experts, leading to the removal of the index. Indexes with a CV above this threshold are marked with an asterisk (*). This study retained factors with values between 0.1 and 0.2 and continued optimization in the second round of expert consultation.

The first round of Delphi consultation collected 35 valid expert responses. Feedback centered on four core dimensions: proposing new indicators, clarifying hierarchical relationships between indicators, resolving conceptual ambiguities, and revising indicator nomenclature. Of these, 30 responses had an adoption rate exceeding 75%. The remaining five were not adopted, as they conflicted with this study’s theoretical framework.

Based on statistical results and qualitative insights from the first round, targeted adjustments were finalized without altering the fundamental structural framework. Specifically, the indicators “Community Identity Element” and “A Sense of Belonging” were removed. Conversely, a new indicator, “Sanitation Infrastructure and Management,” was added to explicitly address the role of hygiene facilities in alleviating pedestrians’ discomfort and anxiety, thereby enriching the coverage of environmental health dimensions. Additionally, three indicators were renamed for precision and clarity: “Proper Lighting” was revised to “Night Lighting,” emphasizing the importance of illumination for pedestrian safety at night; “Transparency of the Architecture” became “Diverse Street Interface,” highlighting the interactive and esthetic qualities of varied building facades; and “Visual Art” was updated to “Visual Landscape,” to better encompass the full range of visual elements in street spaces.

To enhance expert understanding and ensure the accuracy of subsequent evaluations, detailed explanatory notes were added to specific indicators. The indicator “Convenience Shops and Cameras” was clarified to mean a moderate presence of convenience stores with visible surveillance. The previous indicator, “Barrier-Free Paving,” was combined with suggestions on guided Special Pavement Design after “A Sense of Belonging” was removed, resulting in a new indicator called “Pavement Design.” This unified indicator now addresses accessibility features, tactile guidance, child- and intergenerational-friendly zoning, and material and color considerations. Furthermore, the indicators “Rest Facilities” and “Humanity Care Facilities” were merged to create a single indicator, “Street Furniture,” reflecting the diverse and supportive role of street furniture in pedestrian experience.

These updates to the indicator framework and their outcomes following the second round of expert consultation are detailed in

Table 12. The CV value of the second round.

Dimensions	Factors	Mean	Standard Deviation	CV
Safety	Pedestrian Right-of-Way	4.7143	0.46881	0.099444244
	Night Lighting	4.5714	0.64621	0.141359321
	Convenience Shops and Cameras	3.3571	0.49725	0.148118912
Attractiveness	Diverse Street Interface	3.6429	0.49725	0.136498394
	Interactive Facilities	3.3571	0.49725	0.148118912
	Visual Landscape	4.7857	0.42582	0.088977579
Friendliness	Sanitation Infrastructure and Management	4.6429	0.49725	0.107099011
	Street Furniture	3.7143	0.46881	0.126217591
Comfort	Pavement Design	4.6429	0.49725	0.107099011
	Street Greening	4.8571	0.36314	0.074764777
	Street Scale	4.6429	0.49725	0.107099011

.

The second Delphi round improved expert consensus. All factors had CVs between 0.10 and 0.15, showing moderate dispersion. Kendall’s W rose from 0.51 to 0.638, showing a substantial increase in consistency over the first round. These results confirmed that the adjusted indicator framework gained wider expert recognition.

To refine the consultation process and improve evaluation accuracy, we revised the questionnaire. Schematic diagrams clarified environmental factor hierarchies and simplified complex indicator relationships. We also added concise literature-based explanations for each factor to clarify their origins and relevance to street healing. These changes reduce ambiguity and ensure that experts evaluate indicators with a shared understanding.

The revised questionnaire, incorporating these optimizations, served as the basis for the third round of Delphi consultation. Detailed results of the third round are presented in

Table 13. The CV value of the third round.

Dimensions	Factors	Mean	Standard Deviation	CV
Safety	Pedestrian Right-of-Way	4.7143	0.46881	0.099444244
	Night Lighting	4.8571	0.36314	0.074764777
	Convenience Shops and Cameras	3.0714	0.26726	0.087015693
Attractiveness	Diverse Street Interface	3.8571	0.36314	0.094148454
	Interactive Facilities	3.1429	0.36314	0.11554297
	Visual Landscape	4.8571	0.36314	0.074764777
Friendliness	Sanitation Infrastructure and Management	4.6429	0.49725	0.107099011
	Street Furniture	3.8571	0.36314	0.094148454
Comfort	Pavement Design	4.7143	0.46881	0.099444244
	Street Greening	4.8571	0.36314	0.074764777
	Street Scale	4.7143	0.46881	0.099444244

.

The third round of Delphi consultation marked the end of the expert consensus-building process. The CVs for all factors were close to 0.1, indicating minimal dispersion in expert ratings and strong agreement on indicator importance. At the same time, Kendall’s W increased from 0.51 in the first round to 0.714, exceeding the critical 0.7 threshold for scientific validity in Delphi studies. This rise in consensus metrics confirmed that expert opinions had converged, meeting the criteria for ending the consultation process.

AHP was used to allocate weights to the criterion and factor layers. Pairwise comparison matrices assessed the relative significance of the indicator. The standard AHP procedure calculated the weights. The characteristic vector, corresponding to the maximum eigenvalue of the judgment matrix, was found using the eigenvalue method. Final weights were obtained by normalizing this vector.

Table 14. The weight value of the criterion layer and factor layer.

Criterion Layer	Weight Value	Feature Vector	Factor Layer	Weight Value
A1	0.3210	1.414	A11	0.4131
			A12	0.3209
			A13	0.2659
A2	0.1080	0.476	A21	0.1482
			A22	0.0925
			A23	0.759
A3	0.2155	0.949	A31	0.6000
			A32	0.4000
A4	0.3553	1.565	A41	0.4126
			A42	0.2599
			A43	0.3274

details the weight values.

3.2. Healing Potential of “Jingliu Road” Street

This study adopted a hierarchical weighted average method to systematically quantify the healing Potential of “Jingliu Road” Street and its four sections. The assessment framework was built on four primary dimensions (A1–A4) and eleven secondary healing factors (A11–A13, A21–A23, A31–A32, A41–A43), with weights derived from expert consensus (Table 11). For the four dimensions, scores of its secondary factors were weighted and summed to generate dimension-level scores:

$$A_i={\sum }_{j=1}^n(S_{ij}\times W_{ij})\hspace{1em}(i=1, 2, 3, 4;\hspace{0.17em}n=11)$$

(3)

where S_ij is the score of the j-th secondary factor under the i-th primary dimension, and W_ij is the weight of the j-th secondary factor under the i-th primary dimension.

To derive the comprehensive healing score for each section, a weighted summation approach was employed using the dimension-specific weights (A1: 0.321, A2: 0.1080, A3: 0.2155, A4: 0.3553) as coefficients. The formula for the section-level score is Score_section = A1 × 0.321 + A2 × 0.1080 + A3 × 0.2155 + A4 × 0.3553. The healing potential index for Jingliu Road was determined by averaging the comprehensive scores across the four street sections. This aggregated index reflects the street’s holistic healing potential, integrating both section-level variations and dimension-specific contributions. Detailed values are shown in

Table 15. The weighted scores of Jingliu Road in the four dimensions of Safety, Attractiveness, Friendliness, and Comfort.

Street Sections	Safety	Weight Value	Attractiveness	Weight Value	Friendliness	Weight Value	Comfort	Weight Value	Healing Capacity
1–2	3.016795	0.3210	2.859397	0.1080	2.958	0.2155	2.877264	0.3553	2.9369
2–3	3.561524		3.591519		3.56		3.655454		3.5971
3–4	3.560637		3.428064		3.368		3.832934		3.6008
4–5	3.555209		3.089892		3.124		3.482192		3.3854

.

The healing potential of Jingliu Road is at a “good” level, but there is still clear room for improvement. The four street sections reveal the advantages, shortcomings, and development potential of “Jingliu Road” Street. In addition to the quantitative results, the image in Figure 5 validates the evaluation results.

The 3–4 street section received the highest comprehensive healing score (3.6008), with balanced performance in four dimensions. Its indicators of Comfort (3.832934), Safety (3.560637), and Friendliness (3.368) all rank among the top. According to on-site observations, the sidewalk in this section is flat, allowing pedestrians to flow freely. This street section has a high degree of greenery and shade, forming a pleasant tree canopy layer, which enhances the perceptual experience of the green environment. There is a wide variety of community-supporting functions along the street, including community bookstores, Chinese and Western restaurants, hotels, fruit shops, pharmacies, and community service centers. Both quantitative data and on-site observations indicate that this section has the best overall healing effect.

Following the previous section, the 2–3 section scores 3.5971 and has the highest Safety score (3.561524) among all samples. On-site research found that the street’s road surface is flat and continuous, and the sidewalk is designed with functional zoning, including an exclusive pedestrian space.

The 4–5 section scored 3.3854. Quantitative data shows that it has relatively high scores in Safety and Comfort, but its Attractiveness (3.089892) and Friendliness (3.124) are relatively low. Combined with on-site research, it was confirmed that there are problems, such as monotonous spatial design and insufficient interactive elements, in this street section.

The 1–2 section has the lowest score (2.9369) and fails to reach the “neutral” threshold. Its Attractiveness (2.859397) performs weakly, and it lacks Friendliness (2.958) and Comfort (2.877264). On-site research shows that the minimum width of the sidewalk on this section is 2.4 m, and the spread of tree roots has caused uneven pavement in some sections, affecting the smooth passage of pedestrians. It has obvious deficiencies in spatial charm, environmental quality, and social support, and it is the key target for future renovation and improvement.

The evaluation of four street sections along “Jingliu Road” by urban residents reveals the key healing elements and improvement directions that influence the walking healing experience. The results are as follows:

Section 1–2 is a street segment with aging residential areas. Figure 6 shows results from this study’s scoring system: a score of 3 means average healing function, and 4 means satisfactory healing function. Street Greening receives the top score of 3.62. Night Lighting is next at 3.43. Diverse Street Interface (3.31), Sanitation Infrastructure (3.27), and Convenience Shops and Cameras (3.26) also score relatively high. Special Pavement Design (2.4), Street Furniture (2.49), Interactive Facilities (2.47), and Pedestrian Right-of-Way (2.54) score lower. Street Scale (2.89) and Visual Landscape (2.82) fall in the middle range. These data help direct and guide future improvements to the section’s healing functions.

Section 2–3 performs better in the street system. Figure 6 shows scores of various healing elements. Street Scale scores highest, with 3.87, highlighting its key healing role. Special Pavement Design and Street Furniture follow at 3.71 each, showing strong healing effects. Diverse Street Interface (3.62), Pedestrian Right-of-Way (3.65), Visual Landscape (3.59), Interactive Facilities (3.57), and Convenience Shops and Cameras (3.56) score between 3.56 and 3.65, suggesting similar contributions. Street Greening (3.31), Sanitation Infrastructure and Management (3.46), and Night Lighting (3.45) show average performance, but still have potential to improve. These variations show how each healing element contributes to the street-healing environment.

Section 3–4 demonstrates a robust performance within the street system. Notably, Street Scale attains the highest healing score of 3.87, standing out prominently and indicating its significant role in the healing potential. Following closely, Street Greening (3.84) and Special Pavement Design (3.80) suggest considerable advantages in terms of healing effects. Elements such as Street Furniture (3.71), Night Lighting (3.61), Diverse Street Interface (3.57), Convenience Shops and Cameras (3.55), and Pedestrian Right-of-Way (3.53) form a cluster with scores ranging from 3.53 to 3.71. This similarity implies they contribute comparably to the street’s healing function. In contrast, Visual Landscape (3.41) and Interactive Facilities (3.36), while indicating average performance, still offer potential for improvement to meet higher evaluation standards. Lastly, the score of environmental Sanitation Infrastructure and Management is 3.14. Although it exceeded the passing line, it remains at an average level and is the element that most needs improvement at present. Overall, each healing element of this section of the road meets the basic healing requirements.

Section 4–5, a street section dominated by government office buildings, exhibits a polarized performance in healing function. Street Greening attains the highest score of 4.2, indicating a satisfactory healing function. Street Scale scores 4.0, also indicating satisfactory performance. Elements such as Sanitation Infrastructure And Management (3.6), Convenience Shops and Cameras (3.65), and Pedestrian Right-of-Way (3.67) register scores above 3.5, indicating performance between average and satisfactory. In contrast, Special Pavement Design (2.62), Street Furniture (2.41), and Interactive Facilities (2.34) have scores significantly below 3, indicating below-average healing effects. These objective data, combined with on-site investigations that identified issues in design Friendliness, Comfort, and Attractiveness, provide a clear direction for subsequent improvements to the section’s healing functions.

4. Discussion

The weighted assessment model in this study reveals a clear “hierarchy of healing needs” for high-density urban streets. It underscores the critical role of systematic weighting in prioritizing design interventions under resource constraints. Comfort (35.5%) and Safety (32.1%) received higher weights than Friendliness (21.6%) and Attractiveness (10.8%). Maslow’s theory supports this logic by showing that, in stressful urban settings, residents focus first on basic physiological and safety needs. Emotional needs at the social level follow, then esthetic and knowledge-seeking needs [52]. This finding aligns with environmental psychology and risk perception. If safety hazards are present, residents focus on survival threats and cannot relax [22,30,53]. This is why the “Safety” dimension outweighs “Attractiveness” in this study. For urban renewal in high-density districts, resource allocation should first improve key infrastructure, such as pavement continuity and lighting, before funding decorative elements. Without a solid “Safety–Comfort” foundation, even attractive streetscapes may fail to support psychological recovery. This weighted approach quantifies resource allocation in ways unweighted tools cannot.

The key findings are as follows: (1) the “Visual Landscape” was highly weighted (0.759) within the “Attractiveness” dimension, and (2) “Street Greening” had a vital role in the case study scores. These results indicate that even fragmented natural elements on urban streets help reduce stress, supporting SRT and ART in street design. “Soft fascination” from street trees and vegetation is highly effective at restoring attention in dense urban areas [19,20]. Accordingly, increasing the “Green View Ratio” is not only an ecological indicator but also a vital public health strategy. The framework translates psychological theories into actionable design parameters, bridging theory and practice in urban design.

The diagnostic value of this weighted framework is clear from the “deficit trap” seen in Section 1–2. This area has mature street trees and high greening potential, yet received the lowest healing score (2.94). The assessment shows that poor “Pavement Design” (due to root-induced damage) and the lack of “Interactive Facilities” cause this low score. This highlights that planting trees alone is not enough if infrastructure reduces walking comfort. The framework’s ability to find such “qualitative mismatches” is a key improvement over simple binary assessments (like “has greenery” vs. “lacks greenery”). It quantifies the importance of eleven indicators in four areas, so planners can tell the difference between “green streets” and truly “healing streets.” This changes street design from subjective art to evidence-based practice, making sure interventions focus on what most harms the environment’s restorative value.

The core of this research lies in constructing a universal framework adapted to the characteristics of high-density urban streets. However, a key limitation emerged during the research process: the purpose of this research, using Delphi-AHP, is to provide an operational decision-making tool for high-density urban planning [34,54]. However, the quantitative method fails to adequately capture the cultural embeddedness of qualitative experiences. Although the research has balanced subjective and objective criteria through the consensus of multidisciplinary experts, regional cultural differences in qualitative experiences, such as “A Sense of Belonging” and “Comfort,” remain difficult to fully capture with quantitative indicators. This limitation means that the adaptability of the evaluation results in cities with significant cultural context differences needs further verification. It also suggests that quantitative indicators are the “starting point of standardization” rather than “absolute truth”. The research recommends that future users should adjust the weights in combination with local cultural contexts to avoid misinterpreting the “universal healing hierarchy”.

5. Conclusions

This study developed a weighted assessment framework integrating multidisciplinary and multi-stakeholder perspectives to operationalize healing street design for high-density cities. Providing a quantitative approach to prioritize interventions under resource constraints, this research yields the following key conclusions:

(1)

Through Delphi consultations and AHP analysis, the framework established a hierarchical weighting system comprising four dimensions—Comfort (35.5%), Safety (32.1%), Friendliness (21.6%), and Attractiveness (10.8%)—and eleven healing indicators. Empirical validation in Jingliu Road, Zhengzhou, demonstrated effective diagnostic capability, with overall healing scores ranging from 2.94 (Section 1–2) to 3.60 (Section 3–4). The weighted system successfully identified critical deficiencies that traditional unweighted assessments would overlook, enabling evidence-based prioritization.

(2)

The framework supports urban renewal by offering a quantitative tool for planning and evaluation. Unlike qualitative lists, it assigns weight to healing elements, guiding efficient resource use. Integrating these indicators into municipal guidelines can help create psychologically restorative streets, aligning with SDGs 3 and 11.

(3)

Future research should broaden investigations to cities or countries with varied cultural, social, and economic contexts. Comparative analysis will strengthen the evaluation framework’s generalizability and identify context-specific adaptation strategies. Building on this, policy support and community co-construction can be integrated. Evaluation tools should be embedded in local policy, and street intervention plans should be co-created with residents to form a sustainable “design–implementation–maintenance” loop, ensuring interventions’ long-term efficacy. Mixed-methods research should also be used to combine objective physiological data—such as electroencephalogram and heart rate variability—with subjective perceptions for assessing design impacts on mental health and urban livability.