A Demand-Oriented Study on Residential Pilotis Satisfaction in Hefei Using the KANO-IPA Model

Abstract

Under high-density urban development, Residential Pilotis have been widely constructed in Chinese cities as a critical measure to mitigate public space shortages. However, a mismatch between spatial supply and residents’ needs remains prevalent. This study develops a resident satisfaction evaluation framework comprising 23 indicators across four dimensions: Spatial Usability and Sociality, Landscape and Visual Experience, Physical Environment Comfort, and Governance and Operational Maintenance. Using the Integrated KANO-IPA Model, 553 questionnaires from Hefei were analyzed to classify the quality attributes and improvement priorities of the indicators. Results suggest a structural supply–demand mismatch, with the Governance and Operational Maintenance dimension emerging as a particularly prominent area of concern. Satisfaction with Must-be and One-dimensional attributes, especially cleanliness and facility maintenance, age-friendly design, and resting facilities, all of which are highly valued by residents, is generally low. Conversely, landscape-related attributes receive higher satisfaction and have a lower priority for improvement. Based on these findings, a phased optimization strategy is proposed, encompassing short-term priority improvements, medium-term gradual enhancements, and long-term maintenance or flexible adjustments. This research provides an operable methodological framework for supply–demand diagnosis and optimization in similar spatial contexts.

1. Introduction

Residential Pilotis refer to open spaces located at the ground level of the building, supported by structural columns and without enclosing walls [1]. In the context of high-density urban development, the shortage of public spaces has become a global challenge. The United Nations New Urban Agenda emphasizes that by 2030, all people should have access to safe, inclusive, and accessible green public spaces [2]. This goal is particularly challenging for Asian cities, where land scarcity and population concentration exert dual pressures on urban development, compelling cities to explore new spatial solutions for public use [3]. As a form of community public space, Residential Pilotis have emerged as an important approach to alleviate the shortage of public space [4,5].

Residential Pilotis are characterized by their semi-outdoor and semi-public nature [4,6,7,8,9]. Located at the interface between indoor and outdoor environments, and between private and public domains, they naturally attract people to pause, linger, and engage in informal social interactions with neighbors [7,10]. Their open spatial form not only contributes to regulating the microclimate and improving thermal comfort, but also renders them more vulnerable to external climatic influences [11,12,13]. Existing studies have often treated Residential Pilotis merely as a subordinate component of residential public spaces [8,14]. Current research primarily focuses on physical environments and basic facilities [5,11,12], while the relationship between residents’ actual needs and spatial supply remains underexplored and demands further attention [5]. Therefore, the design of Residential Pilotis should fully consider both their spatial characteristics and residents’ differentiated needs, avoiding the mechanical application of traditional planning models.

In China, the development of Residential Pilotis was initially constrained by strict planning policies [15]. Around 2015, cities such as Hefei, Shenzhen, and Guangzhou introduced planning requirements mandating the construction of pilotis, marking their transition from marginal to mainstream Spatial Form [16,17,18]. Under such policy-driven expansion, the spatial supply of pilotis—primarily guided by a top-down approach—has been susceptible to information asymmetry between supply and demand, often resulting in a supply–demand mismatch [5]. Moreover, within the real-estate developer–dominated construction model, developers may prioritize visual design elements that enhance marketing and price premiums, while neglecting residents’ actual usage needs [19,20,21], thereby exacerbating the mismatch. Undersupply can lead to declines in residents’ quality of life and well-being, while oversupply can cause inefficient resource allocation and waste [22,23]. Only by accurately understanding users’ needs and achieving demand-driven spatial provision can cities effectively mitigate the supply–demand mismatch, enhance resource efficiency, and ensure sustainable spatial development [24].

Satisfaction serves as a key indicator for assessing the alignment between spatial supply and user demand [24,25,26]. Originating from consumer behavior theory, “satisfaction” is defined as the perceived gap between expectations and actual experience [27]. In fields such as urban planning [28], public administration [29], residential environment studies [30], and environmental sustainability [31], satisfaction research has evolved into a multidimensional evaluation system encompassing physical, social, and managerial dimensions [25], making it an important tool for evaluating the degree of match between spatial supply and demand [24,32,33]. The level of residents’ satisfaction not only directly affects their willingness and frequency of use but also indirectly influences the long-term sustainability of public spaces [32]. Thus, satisfaction analysis is an effective tool for diagnosing potential supply–demand mismatches [24,32].

In prior satisfaction studies, the KANO-IPA Model has been widely applied to public spaces such as urban parks, green spaces, and neighborhoods. The KANO Model distinguishes quality attributes of user needs, identifying their nonlinear impacts on satisfaction and clarifying how different attributes influence user perceptions [34]. The Importance–Performance Analysis (IPA) simultaneously evaluates the importance and satisfaction of each indicator, providing an intuitive framework to identify items that require urgent improvement or long-term maintenance, enabling an intuitive identification of key items that require either priority improvement or long-term maintenance, thereby providing a scientific basis for optimizing public spaces and residential environments [35]. The integration of KANO and IPA models enables both qualitative categorization and quantitative prioritization—revealing nonlinear relationships between user needs and satisfaction, while offering data-driven insights into improvement strategies—thus enhancing the comprehensiveness and practical value of satisfaction research.

Taking Hefei, China, as a case study, this research investigates residential satisfaction as an entry point to explore the alignment between spatial supply and residents’ needs in Residential Pilotis. It aims to identify key improvement indicators and provide targeted strategies for the optimization of spatial design, operational management, and policy formulation. The ultimate goal is to improve resource allocation efficiency from the supply side and mitigate the supply–demand mismatch. The specific objectives are as follows:

(1)

To identify demand indicators for evaluating residents’ satisfaction with Residential Pilotis and construct a comprehensive satisfaction evaluation framework;

(2)

To apply the KANO-IPA Model to classify the quality attributes of each indicator and, through the analysis of satisfaction and importance, reveal the current supply–demand alignment and corresponding decision categories;

(3)

To propose phased and prioritized optimization strategies based on decision types and improvement priorities, enhancing the future alignment between spatial supply and residents’ needs in Residential Pilotis.

2. Materials and Methods

2.1. Study Area and Research Subject

This study selects Hefei, the capital city of Anhui Province, China, as the case study area for two primary reasons.

First, Hefei was among the earliest cities to introduce planning policies mandating the construction of Residential Pilotis. As a pioneer in implementing such policies, Hefei’s practical experience provides valuable insights into the initial effects and potential issues of policy execution—particularly regarding the discrepancies between spatial supply and residents’ actual needs. These insights offer reference value for other cities adopting similar policy frameworks.

Second, Hefei’s climate is characterized by hot summers and cold winters, a typical pattern in the middle and lower reaches of the Yangtze River. Unlike tropical regions where pilotis are more prevalent and primarily serve year-round ventilation, the distinct seasonal fluctuations in Hefei provide a more complex climatic context to evaluate the year-round adaptability of such semi-outdoor spaces [36]. The semi-outdoor nature of Residential Pilotis makes their environmental quality highly sensitive to seasonal variations. Studying them under such climatic conditions helps to better understand the relationship between Physical Environment Comfort and user satisfaction, thereby enhancing the generalizability and applicability of the research findings.

Typical spatial configurations and functional types of Residential Pilotis observed in this area are shown in Figure 1. The geographic location and administrative districts of the study area are presented in Figure 2.

Figure 1. Visual examples. (a) Piloti Entrance; (b) Fitness and Exercise Space; (c) Children’s Play Space; (d) Social Interaction Space.

Figure 2. Location of the Study Area.

To ensure the homogeneity of the research subjects, this study operationally defines ‘Residential Pilotis’ as column-supported, wall-free public spaces located at the ground level of residential buildings. These spaces, constructed under Hefei’s 2016 planning regulations, exhibit semi-outdoor and semi-public characteristics, and remain fully accessible to all residents [1,16].

2.2. Research Framework

This study first establishes a preliminary evaluation framework through an extensive literature review, which is then refined using Exploratory Factor Analysis (EFA) to develop the final assessment structure.

Subsequently, a resident satisfaction questionnaire was designed based on the KANO-IPA Model and administered in Hefei for data collection. The collected data are analyzed as follows:

The KANO Model is applied to classify the quality attributes of residents’ demand indicators.

The Importance–Performance Analysis (IPA) is employed to examine the alignment between supply and demand across existing spatial indicators, thereby identifying the improvement priorities and decision types.

An Improvement Priority Index is calculated to determine the specific sequence of improvement actions.

Finally, based on the integrated results derived from the KANO-IPA, this study proposes phased and targeted optimization strategies to guide spatial enhancement and management. The overall research framework is illustrated in Figure 3.

Figure 3. Study framework.

2.3. Construction and Refinement of the Evaluation Framework

2.3.1. Preliminary Construction of the Evaluation Framework

Maslow’s Hierarchy of Needs Theory categorizes human needs into five levels: physiological, safety, belongingness, esteem, and self-actualization [37]. Carr et al. (1992) identified three key dimensions for evaluating public spaces—needs, rights, and meanings—and further elaborated five fundamental human needs within public spaces: comfort, relaxation, passive engagement, active engagement, and discovery [38]. Carmona (2008) incorporated Management as an independent dimension and defined four essential aspects of the Governance and Operational Maintenance dimension—coordination, regulation, maintenance, and investment [39].

Building upon the complementarity of these theoretical perspectives, this study establishes several principles for constructing a resident demand-oriented evaluation framework:

(1)

Guided by Maslow’s hierarchy of needs and Carr et al.’s framework, residents’ needs for Residential Pilotis are restructured into basic needs (e.g., spatial functionality and physical environmental comfort supporting daily activities) and advanced needs (e.g., participation in community governance and social interaction). This ensures that the framework addresses not only the physical supply of space but also its capacity to satisfy higher-level psychological and social needs.

(2)

Abstract needs, anchored in spatial functions and activity facilities as carriers and utilizing the physical environment as a prerequisite, are operationalized into perceptible indicators. This allows resident satisfaction to effectively measure the supply–demand match within the spatial provision.

(3)

The Management dimension is explicitly incorporated into the framework and subdivided into concrete demand indicators. Residents’ satisfaction with these indicators reflects their psychological expectations regarding the space’s long-term capacity to continually meet evolving needs.

Following these principles, a preliminary resident satisfaction evaluation framework (Table 1) was developed, comprising four dimensions. These dimensions are categorized into basic needs and improvement needs. Basic needs include Comfort and Relaxation and Space Function and Inclusiveness, which address spatial functionality and facilities supporting residents of different ages, as well as physical comfort in semi-outdoor environments. Improvement needs include Social Interaction and Community Identity and Governance and Operation, which concern participation in community governance, social interaction, and sense of community belonging.

Table 1. Initial Evaluation Framework.

Target Layer	Principle	Criterion Layer	Code	Alternative Layer	References
Resident Needs	Basic Needs (Activity Carriers and Basic Conditions)	Comfort and Relaxation	1	Degree of Openness	[11,40,41,42]
			2	Summer Ventilation and Winter Wind Protection	[11,40,41]
			3	Natural Daylighting and Artificial Lighting	[42,43]
			4	Thermal Comfort	[6,40,44]
			5	Acoustic Comfort	[45]
			6	Visual Aesthetics and Coherence	[46,47]
			7	Greenery Configuration	[42,43]
			8	Landscape Features	[45,48]
			9	External Landscape View	[49]
		Space Function and Inclusiveness	10	Resting Facilities	[20]
			11	Children’s Play Facilities	[20]
			12	Fitness and Exercise Facilities	[50,51]
			13	Age-Friendly Design	[5,52,53,54]
			14	Spatial Flexibility	[2,55]
			15	Accessibility	[56]
			16	Visibility	[49,56]
	Improvement Needs (Community Identity and Participatory Governance)	Social Interaction and Community Identity	17	Informal Social Interaction	[10,57]
			18	collective space	[58,59]
			19	Design to Promote Social Interaction	[38,60]
			20	Design to Promote Community Identity	[57,61]
		Governance and Operation	21	Management Policies and Usage Regulations	[2,39,62]
			22	Cleanliness and Facility Maintenance	[39,49,54]
			23	Resident Feedback Channels	[39,49]
			24	Resident Participation in Decision-Making	[2,39]
			25	Public Disclosure of Fund Utilization	[2,39,49]

Residential Pilotis constitute a distinctive type of semi-outdoor and semi-public space, characterized by partial exposure to outdoor environmental conditions [6,11,12], a transitional position between private and public domains [7,8,10], and complexity in long-term governance and operation [39]. These characteristics distinguish Pilotis from conventional public spaces such as parks or plazas. Therefore, the selected demand indicators should be comprehensive yet avoid redundancy, while reflecting the unique spatial attributes of Pilotis and capturing residents’ differentiated needs.

The preliminary indicators in this study are derived from three sources: (1) domestic and international design guidelines for residential public spaces, particularly Chinese design guidelines and planning policies for Residential Pilotis; (2) academic literature on traditional residential public spaces as well as semi-outdoor and semi-public environments; (3) field investigations and informal interviews with residents conducted in Hefei. This study aims to extract the key elements that residents are concerned about in Residential Pilotis, transform them into assessable demand indicators, and incorporate them into the satisfaction evaluation framework. By filtering the high-frequency terms from sources (1) and (2), extracting common spatial elements, and supplementing with indicators identified from source (3), a preliminary set of 25 demand indicators was established (Table 1).

2.3.2. Refinement of the Evaluation Framework

To enhance the applicability of the evaluation framework within the Chinese context, this study employs Exploratory Factor Analysis (EFA) to refine the initial set of indicators. Based on the analysis results—specifically factor loadings and communalities—unsuitable indicators were removed, and the underlying dimensional structure of the framework was identified. EFA is particularly effective for uncovering latent constructs behind observed variables and is thus well-suited to studies requiring empirical validation of theoretical frameworks.

A total of 150 professionals were invited to participate as experts (Table 2). All participants had a minimum of three years of relevant work experience in fields such as urban planning, architecture, real estate development, and property management. Using a five-point Likert scale, the experts evaluated the importance of the 25 preliminary indicators included in the initial framework.

Table 2. Characteristics of professionals for factor analysis.

Category	Item	Frequency	Percentage
Gender	Male	68	46.67%
	Female	82	53.33%
Work Experience	Under 5 years	10	10.00%
	5–10 years	52	6.67%
	10–15 years	48	34.66%
	15–20 years	34	32.00%
	Over 20 years	6	4.00%
Field	Architecture	40	26.67%
	Planning	35	23.33%
	Interior Design	30	20.00%
	Real Estate Dev	25	16.67%
	Property Services	20	13.33%
No. of answered questionnaires		150

Data analysis was conducted using the Principal Axis Factoring (PAF) method for factor extraction, followed by oblique rotation to achieve optimal factor interpretability. The dataset was tested for sampling adequacy and suitability using the Kaiser-Meyer-Olkin (KMO) Test and Bartlett’s Test of Sphericity. The KMO value was 0.876, indicating strong sampling adequacy, while Bartlett’s test was statistically significant (p = 0.000), confirming that the data were appropriate for factor analysis.

Based on the criterion of eigenvalues greater than 1, four common factors were extracted. The results showed that the indicators “Visibility” and “Collective Space” had communalities below 0.4, suggesting that their variance was not effectively explained by the extracted factors; hence, these two indicators were removed (Table 3). After excluding them and re-running the analysis, the four-factor structure remained stable, accounting for 85.9% of the total variance, demonstrating a strong explanatory power of the refined model (Table 4).

Table 3. Factor Loadings After Rotation and Communalities.

Demand Indicator			Factor Loading				Communality
Criterion Layer	Code	Alternative Layer	Factor 1	Factor 2	Factor 3	Factor 4
Comfort and Relaxation	1	Degree of Openness	0.036	0.067	0.968	0.002	0.945
	2	Summer Ventilation and Winter Wind Protection	0.026	0.012	0.971	0.011	0.945
	3	Natural Daylighting and Artificial Lighting	0.014	0.045	0.934	0.019	0.875
	4	Thermal Comfort	0.098	0.009	0.937	0.055	0.884
	5	Acoustic Comfort	0.034	0.042	0.916	0.020	0.844
	6	Visual Aesthetics and Coherence	0.006	0.052	0.035	0.948	0.920
	7	Greenery Configuration	0.082	0.015	0.014	0.962	0.932
	8	Landscape Features	0.071	0.030	0.022	0.962	0.923
	9	External Landscape View	0.010	0.032	0.047	0.971	0.934
Space Function and Inclusiveness	10	Resting Facilities	0.982	0.013	0.020	0.010	0.165
	11	Children’s Play Facilities	0.972	0.001	0.032	0.011	0.147
	12	Fitness and Exercise Facilities	0.978	0.012	0.008	0.003	0.957
	13	Age-Friendly Design	0.979	0.002	0.015	0.054	0.960
	14	Spatial Flexibility	0.975	0.040	0.034	0.002	0.954
	15	Accessibility	0.971	0.023	0.040	0.013	0.946
	16	Visibility	0.150	0.084	0.054	0.035	0.035
Social Interaction and Community Identity	17	Informal Social Interaction Spaces	0.980	0.002	0.040	0.022	0.964
	18	Collective Space	0.042	0.247	0.080	0.044	0.067
	19	Design to Promote Social Interaction	0.981	0.031	0.005	0.006	0.963
	20	Place Identity and Community Belonging Design	0.963	0.003	0.051	0.017	0.929
Governance and Operation	21	Cleanliness and Facility Maintenance	0.045	0.961	0.056	0.025	0.924
	22	Management Policies and Usage Regulations	0.048	0.977	0.018	0.002	0.957
	23	Resident Feedback Channels	0.058	0.943	0.044	0.047	0.885
	24	Resident Participation in Decision-Making	0.035	0.952	0.006	0.031	0.918
	25	Public Disclosure of Fund Utilization	0.037	0.940	0.065	0.045	0.901

Table 4. Total Variance Explained.

Factor	Initial Eigenvalue			Extraction Sums of Squared Loadings			Rotation Sums of Squared Loadings A
	Total	% of Variance	Cumulative %	Total	% of Variance	Cumulative %
1	8.684	34.736	34.736	8.635	34.540	34.540	8.632
2	5.015	20.059	54.795	4.922	19.688	54.228	4.714
3	4.611	18.444	73.239	4.512	18.047	72.275	4.507
4	3.488	13.952	87.191	3.407	13.627	85.901	3.815

The results of the factor analysis led to a refinement of the preliminary evaluation framework. The original dimension of Comfort and Relaxation was differentiated into two distinct dimensions: one consisting of visual-aesthetic indicators (e.g., greenery configuration, landscape features, external view, and visual harmony and aesthetics), and another comprising physical performance indicators (e.g., summer ventilation, natural and artificial lighting, and acoustic comfort). The clustering patterns of the remaining indicators were largely consistent with those in the preliminary framework.

Based on the analysis of the factor loading matrix and the conceptual characteristics of each factor, the four dimensions were renamed as follows:

Factor 1: Spatial Usability and Sociality, mainly encompassing indicators related to spatial functionality and social interaction;

Factor 2: Governance and Operational Maintenance, including indicators pertaining to spatial management, regulation, and maintenance;

Factor 3: Physical Environment Comfort, covering indicators of environmental and physical performance;

Factor 4: Landscape and Visual Experience, aggregating indicators related to visual aesthetics and landscape perception.

Consequently, a refined evaluation framework was developed, consisting of four dimensions and twenty-three indicators (Table 5).

Table 5. Criterion Layer and Alternative Layer Mapping.

Criterion Layer	Code	Alternative Layer
Spatial Usability and Sociality	S1	Resting Facilities
	S2	Children’s Play Facilities
	S3	Fitness and Exercise Facilities
	S4	Age-Friendly Design
	S5	Spatial Flexibility
	S6	Accessibility
	S7	Design to Promote Social Interaction
	S8	Informal Social Interaction Spaces
	S9	Place Identity and Community Belonging Design
Landscape and Visual Experience	L1	Greenery Configuration
	L2	Landscape Features
	L3	External Landscape View
	L4	Visual Aesthetics and Coherence
Physical Environment Comfort	E1	Thermal Comfort
	E2	Acoustic Comfort
	E3	Natural Daylighting and Artificial Lighting
	E4	Degree of Openness
	E5	Summer Ventilation and Winter Wind Protection
Governance and Operational Maintenance	G1	Cleanliness and Facility Maintenance
	G2	Public Disclosure of Fund Utilization
	G3	Management Policies and Usage Regulations
	G4	Resident Feedback Channels
	G5	Resident Participation in Decision-Making

2.4. Evaluation Model and Questionnaire Design

2.4.1. KANO Model

Proposed by Noriaki Kano in 1984, the KANO Model is a theoretical tool used to identify and analyze the nonlinear relationship between product or service attributes and user satisfaction. It reveals how the presence or absence of specific features or services exerts differentiated impacts on user satisfaction.

The model classifies product or service quality attributes into five categories: Must-be (M), One-dimensional (O), Attractive (A), Indifferent (I), Reverse (R). These five categories and their corresponding relationships with satisfaction are summarized in Table 6.

Table 6. Explanation of the Five Categories for Service Quality Attributes in the Kano Model.

Attribute Category	Explanation
Must-be Attributes (M)	Basic attributes that users take for granted. Their absence causes strong dissatisfaction, but their presence does not significantly increase satisfaction.
One-Dimensional Attributes (O)	One-dimensional attributes that have a linear relationship with satisfaction; the better they are fulfilled, the higher the satisfaction.
Attractive attributes (A)	Excitement attributes that delight users when present but do not cause dissatisfaction when absent.
Indifference Attributes (I)	Attributes that have little to no impact on user satisfaction, regardless of their presence or absence.
Reverse Attributes (R)	Attributes that cause dissatisfaction when present and satisfaction when absent.

The traditional KANO Model primarily focuses on qualitative classification methods. However, when users’ perceptions or preferences are ambiguous, the model often struggles to capture the fuzziness and uncertainty inherent in user evaluations. To address this limitation, the Analytical KANO Model introduces a quantitative mechanism that transforms users’ subjective perceptions into measurable vector data by calculating importance and satisfaction indices.

This enhanced approach not only provides a more accurate representation of users’ psychological uncertainty but also improves analytical precision. As a result, it allows for a more effective identification of the nonlinear relationships between specific demand indicators and user satisfaction.

2.4.2. Importance-Performance Analysis (IPA) Model

The IPA model is a comprehensive analytical approach that incorporates two dimensions: importance and performance [63]. Typically, the horizontal axis represents the level of perceived importance by users, while the vertical axis represents the level of perceived satisfaction. The IPA is divided into four quadrants (Figure 4). Based on the distribution of indicators, one can determine the decision type for spatial demand metrics.

Figure 4. Schematic diagram of IPA.

Demands in the first quadrant exhibit both high user satisfaction and high importance, whereas demands in the second quadrant show high satisfaction but low importance, indicating that the current configuration meets or even exceeds user expectations. The decision type for demand indicators in these two quadrants is “Maintain current supply.” In the third quadrant, demands have both low satisfaction and low importance. In the fourth quadrant, demands have low satisfaction but high importance, suggesting that these are aspects that users generally consider important but are dissatisfied with. The decision type for demand indicators in these two quadrants is “Improve current supply.”

2.4.3. Questionnaire Design and Data Collection

This study employed an analytical KANO model integrated with IPA to design the questionnaire. Questions reflecting both positive and negative directions of the KANO model were combined into a single set, with asymmetric scoring of options based on Xu et al.’s study. Importance ratings for each demand indicator were assigned using a [0, 1] interval scale [64]. The questionnaire was developed according to the 23 indicators in the final evaluation framework (Table 5); the specific questionnaire format and scoring are provided in Appendix A.

This study distributed online questionnaires in the four main administrative districts of Hefei via the Questionnaire Star platform. Questionnaires were promoted through social media and disseminated with the assistance of neighborhood committees and property management companies. The questionnaire included a screening item to identify whether residents resided in communities equipped with Residential Pilotis; only those selecting “Yes” proceeded to subsequent sections. The survey collected respondents’ demographic variables, including residential location, length of residence, and household composition, as well as basic information on their community, such as name and year of construction. To ensure data reliability, the questionnaire included attention-check items, and responses with abnormal completion times were excluded.

The survey was conducted from March to April 2024 over eight weeks. A total of 753 responses were collected, of which 553 were valid after data cleaning, yielding an effective rate of 73.44%. The sample size met the requirements for statistical analysis. Demographic information is presented in Table 7. Respondents were distributed across major districts of Hefei and resided in communities of varying construction periods. Notably, 84.4% lived in communities completed after 2019 (with sales starting in 2016), consistent with the widespread adoption of Piloti following the policy implementation in 2016.

Table 7. Demographic Characteristics of Residents.

Variable	Category	Frequency	Percentage
Gender	Male	229	45.8%
	Female	271	54.2%
Age	18–30 years	143	28.6%
	31–40 years	156	31.2%
	41–50 years	112	22.4%
	51–60 years	64	12.8%
	Above 60 years	25	5.0%
Education Level	Bachelor’s degree	276	55.2%
	Bachelor’s and Associate degrees	132	26.4%
	Master’s degree or above	92	18.4%
Household Structure	Single/couple	214	42.8%
	Family with children	158	31.6%
	Three-generation household	128	25.6%
Communities age	Within 1 year	112	22.4%
	1–3 years	179	35.8%
	3–5 years	131	26.2%
	More than 5 years	78	15.6%
Residential District	Luyang District	136	24.6%
	Yaohai District	109	19.8%
	Baohe District	162	29.2%
	Shushan District	146	26.4%

This study obtained informed consent from all participants through an online questionnaire platform, with all data collected anonymously. As a non-interventional social study, this research involved only non-sensitive demographic data and evaluations of public service facility usage experience, without collecting medical, financial, or other private information. In accordance with Article 32 of the Measures for Ethical Review of Life Science and Medical Research Involving Human Beings (National Health Commission Order No. 4 [2023]), this study qualifies for exemption from ethical review [65].

2.5. Data Analysis

2.5.1. Classification of Demand Indicator Attributes

x denotes the reverse-worded satisfaction score given by resident j for demand indicator $f_i$, and $y_{ij}$ denotes the positively worded satisfaction score; $b_{ij}$ denotes the importance rating. By applying a weighted average to the scores of both positively and negatively worded items, the weighted per capita reverse-worded satisfaction (${\overline{x}}_i$) and positively worded satisfaction (${\overline{y}}_i$) for demand indicator $f_i$ can be obtained. The calculation formulas are as follows:

$${\overline{x}}_i={\displaystyle \frac{1}{J}}\sum _{j=1}^J{x}_{ij}{b}_{ij};{\overline{y}}_i={\displaystyle \frac{1}{J}}\sum _{j=1}^J{y}_{ij}{b}_{ij}$$

(1)

This study follows the guidelines for establishing an analytical KANO model to plot the KANO coordinates [64]. Using ${\overline{x}}_i$ as the horizontal axis and ${\overline{y}}_i$ as the vertical axis, the values are transformed into a vector ${\overrightarrow{r}}_i$, The magnitude of the vector $r_i=\left|{\overrightarrow{r}}_i\right|$ represents the importance index, while the angle ${\alpha }_i$ represents the satisfaction index. The calculation formulas are as follows:

$$r_i=\sqrt{{{\overline{x}}_i}^2+{{\overline{y}}_i}^2};{\alpha }_i=\arctan \left({\overline{y}}_i/{\overline{x}}_i\right)$$

(2)

According to the classification criteria of the KANO model, the quality attribute corresponding to each demand indicator $f_i$ can be determined. Following the approach proposed by Xu et al., the threshold values for quality attribute classification are set as $r_0=0.5$, ${\alpha }_L=\mathsf{\pi }/6$, ${\alpha }_H=\mathsf{\pi }/3$. When $r_i>0.5$ and ${\alpha }_i>\mathsf{\pi }/3$, the indicator is classified as an Attractive Attribute; When $r_i>0.5$ and ${\mathsf{\pi }/6<\alpha }_i<\mathsf{\pi }/3$, it is classified as a One-dimensional Attribute; When $r_i>0.5$ and ${\alpha }_i<\mathsf{\pi }/6$, it is classified as a Must-be Attribute; When $r_i<0.5$, it is classified as an Indifferent Attribute.

Thus, the position of each demand indicator $f_i$ in the coordinate system—determined by its importance index $r_i$ and satisfaction index ${\alpha }_i$ indicates its corresponding quality attribute.

While Xu et al. originally proposed optimizing thresholds based on manufacturing cost data, such data is typically unavailable for residential public spaces [64]. Consequently, the threshold settings adopted in this study are consistent with recent applications in the built environment [32,66].

2.5.2. Classification and Quantitative Ranking of Improvement Priorities for Demand Indicators

Based on the IPA model, a coordinate system is constructed using the importance index $r_i$ as the horizontal axis and the satisfaction index ${\alpha }_i$ as the vertical axis, where $0\le r_i\le \sqrt{2}$, $0\le {\alpha }_i\le \pi /2.$The coordinate system is divided into four quadrants using the average satisfaction $\overline{\alpha }$ and average importance $\overline{r}$. The quadrant in which each demand indicator is located represents its priority level (Figure 5). Among these, Quadrants I and II constitute the “Maintain current supply” region, while Quadrants III and IV constitute the “Improve current supply” region.

Figure 5. IPA Model Quadrant Diagram.

Since the four-quadrant IPA diagram can only provide an initial qualitative assessment of spatial demand indicators and cannot achieve quantitative ranking, a priority index is introduced as a quantitative criterion for ordering improvement priorities. This index allows for a more accurate determination of the ranking of demand indicators. The priority index ${\rho }_i$ for each demand indicator is calculated as follows:

$${\rho }_i={\displaystyle \frac{2\sqrt{2}}{3}}\left(1-{\displaystyle \frac{{\alpha }_i}{\pi }}\right)\left|r_i\right|$$

(3)

A higher value of the priority index ${\rho }_i$ indicates that the corresponding demand indicator requires greater attention in decision-making. By applying this calculation to all demand indicators, a quantitative ranking can be obtained, providing a more accurate basis for subsequent decision-making.

3. Results

3.1. Reliability and Validity

Prior to analysis, reliability and validity tests were conducted using SPSS 27.0 to assess the quality of the residents’ questionnaire data. The Cronbach’s α values for the positive questions, negative questions, and importance ratings were 0.833, 0.893, and 0.900, respectively, indicating good reliability (Appendix B).

Validity was assessed using the Kaiser-Meyer-Olkin (KMO) measure and Bartlett’s Test of Sphericity. The KMO values for positive questions, negative questions, and importance ratings were 0.876, 0.932, and 0.947, respectively. Bartlett’s Test of Sphericity was significant at p < 0.001, indicating that the data were suitable for subsequent analysis (Appendix C).

3.2. KANO Model Analysis Results

Based on the analytical KANO model, the 23 demand indicators were classified into four quality attributes (Figure 6).

Figure 6. Attribute classification results for spatial demand indicators.

Two indicators, “Summer Ventilation and Winter Wind Protection (E5)” and “Cleanliness and Facility Maintenance (G1),” were classified as Must-be attributes (M). Fourteen indicators, including “Resting Facilities (S1),” “External Landscape Views (L3),” “Accessibility (S6),” “Natural and Artificial Lighting (E3),” “Informal Social Interaction Spaces (S8),” “Thermal Comfort (E1),” “Acoustic Comfort (E2),” “Management Policies and Usage Regulations (G3),” “Resident Feedback Channels (G4),” “Resident Participation in Decision-Making (G5),” “Age-Friendly Design (S4),” “Public Disclosure of Fund Utilization. (G2),” “Fitness and Exercise Facilities (S3),” and “Visual Aesthetics and Coherence (L4),” were classified as One-dimensional attributes (O). Three indicators, “Greenery Configuration (L1),” “Landscape Features (L2),” and “Children’s Play Facilities (S2),” were classified as Attractive attributes (A). Four indicators, including “Design Promoting Social Interaction (S7),” “Place Identity and Community Belonging Design (S9),” “Degree of Openness (E4),” and “Spatial Flexibility (S5),” were classified as Indifferent attributes (I) (Figure 6).

3.3. IPA Model Analysis Results

In this study, the IPA coordinate system was divided into four quadrants using the mean importance index (0.617) and mean satisfaction index (0.772) as thresholds, namely: Keeping on, Over-provisioning, Lower value, and Prioritize upgrading. The distribution of demand indicators with the four quality attributes across these quadrants is shown in Figure 7.

Figure 7. IPA diagram.

In the IPA diagram (Figure 7), the decision type for Quadrants I and II is “Maintain current supply.” Quadrant I includes six demand indicators: S3, L2, L3, L4, G2, and G5. These indicators exhibit relatively high importance and high satisfaction, indicating that the current provision largely meets residents’ needs. Among them, L2 is classified as an Attractive attribute, while the remaining indicators are One-dimensional attributes.

Quadrant II contains five indicators: L1, S2, S5, S9, and E4. These indicators have high satisfaction but relatively low importance, suggesting the current provision has met residents’ needs. Among these, L1 and S2 are Attractive attributes, whereas the others are Indifferent attributes.

The decision type for Quadrants III and IV is “Improve current supply.” Quadrant IV shows higher importance than Quadrant III. Quadrant III includes five indicators: E1, E2, E5, S6, and S7. Except for E5, the remaining indicators have both low satisfaction and low importance, suggesting limited resident attention and the potential for targeted improvements.

Quadrant IV contains seven indicators: S1, S4, S8, E3, G1, G3, and G4. These indicators have low satisfaction but high importance, indicating that the current provision of these high-importance aspects does not meet residents’ needs, resulting in overall low satisfaction. In particular, G1, as a Must-be attribute, shows a significant drop in satisfaction if under-provided, highlighting the need for prioritized attention.

3.4. Prioritization of Improvement for Demand Indicators

Establishing the improvement priority of demand indicators helps planners allocate resources efficiently under limited conditions, focusing on key indicators that significantly affect residents’ satisfaction. This approach allows for the optimization of spatial provision while maximizing resident satisfaction.

The priority index ${\rho }_i$ for each demand indicator was calculated using Equation (3). The improvement priorities were then determined by combining the KANO quality attributes of the indicators with their positions in the IPA quadrants. The ranking follows these principles:

Decision type: “Improve current supply” > “Maintain current supply”;

IPA quadrant: Prioritize upgrading > Lower value > Keeping on > Over-provisioning;

Within the same category, indicators are ranked in descending order of the priority index.

The improvement priorities of the 23 demand indicators are presented in Table 8 and Table 9. The average priority indices for indicators classified as “Improve current supply” and “Maintain current supply” were 0.486 and 0.388, respectively. The priority indices of both categories exhibit a decreasing trend, indicating a gradual decline in their impact on resident satisfaction or the urgency of improvement for Residential Pilotis. This result aligns with the decision logic of the IPA model, validating the accuracy of the improvement priority ranking. Within each category, the priority indices of spatial demand indicators show significant differentiation, suggesting considerable variation in the importance of these needs. Therefore, optimization efforts should focus on the most critical demands.

Table 8. Priority ranking of indicators in “Improve current supply”.

Demand Indicator (Alternative Layer)	Code	Quality Attribute	IPA Quadrant	$\mathbf{Priority}\mathbf{Index}({\mathit{\rho }}_{\mathit{i}})$	Rank
Age-Friendly Design	S4	O	Prioritize upgrading	0.559	1
Resting Facilities	S1	O	Prioritize upgrading	0.556	2
Resident Feedback Channels	G4	O	Prioritize upgrading	0.546	3
Cleanliness and Facility Maintenance	G1	M	Prioritize upgrading	0.534	4
Management Policies and Usage Regulations	G3	O	Prioritize upgrading	0.526	5
Natural Daylighting and Artificial Lighting	E3	O	Prioritize upgrading	0.516	6
Informal Social Interaction Spaces	S8	O	Prioritize upgrading	0.496	7
Summer Ventilation and Winter Wind Protection	E5	M	Prioritize upgrading	0.488	8
Accessibility	S6	O	Lower value	0.458	9
Acoustic Comfort	E2	O	Lower value	0.452	10
Thermal Comfort	E1	O	Lower value	0.389	11
Design to Promote Social Interaction	S7	I	Lower value	0.314	12

Table 9. Priority ranking of indicators in “Maintain current supply”.

Demand Indicator (Alternative Layer)	Code	Quality Attribute	IPA Quadrant	$\mathbf{Priority}\mathbf{Index}({\mathit{\rho }}_{\mathit{i}})$	Rank
Resident Participation in Decision-Making	G5	O	Keeping on	0.537	13
Public Disclosure of Fund Utilization	G2	O	Keeping on	0.457	14
External Landscape View	L3	O	Keeping on	0.456	15
Visual Aesthetics and Coherence	L4	O	Keeping on	0.432	16
Fitness and Exercise Facilities	S3	O	Keeping on	0.417	17
Landscape Features	L2	A	Keeping on	0.381	18
Children’s Play Facilities	S2	A	Over-provisioning	0.354	19
Greenery Configuration	L1	A	Over-provisioning	0.350	20
Place Identity and Community Belonging Design	S9	I	Over-provisioning	0.346	21
Degree of Openness	E4	I	Over-provisioning	0.296	22
Spatial Flexibility	S5	I	Over-provisioning	0.241	23

Among the indicators classified as “Improve current supply,” “Age-Friendly Design (S4)” has the highest improvement priority. Indicators within the Governance and Operational Maintenance dimension generally exhibit high priority indices, highlighting their importance in future improvement efforts. Significant differences are observed across evaluation dimensions: the Spatial Usability and Sociality dimension shows a polarization, with basic functional needs being urgent while social interaction-related indicators have low priority; the Physical Environment Comfort dimension is overall weak, with most indicators falling into the “Prioritize upgrading” quadrant; the Landscape and Visual Experience dimension performs the best, with sufficient provision across indicators.

The priority indices reveal an imbalance in resource allocation: Must-be attributes (average 0.526) exhibit higher improvement priorities, whereas Attractive attributes (average 0.362) and Indifferent attributes (average 0.299) show higher satisfaction levels and lower urgency for improvement. Issues within the Governance and Operational Maintenance dimension are particularly prominent. Subsequent improvements should prioritize basic needs, optimize high-priority One-dimensional attributes, and optimize resource allocation to achieve a balance between supply and demand.

Notably, among indicators classified as “Maintain current supply,” some have priority indices higher than certain indicators in the “Improve current supply” category. These high-priority One-dimensional attributes, due to their high importance, can significantly influence overall satisfaction if their provision levels change, and therefore require focused attention in future resource allocation.

4. Discussion

The results reveal the structural characteristics of the supply–demand mismatch in Residential Pilotis: while Must-be and high-priority One-dimensional attributes are generally under-supplied, satisfaction with Attractive and Indifferent attributes remains relatively high. Specifically, demand indicators categorized under the ‘Improve current supply’ decision type and located in the ‘Prioritize upgrading’ quadrant span three dimensions: Spatial Usability and Sociality, Governance and Operational Maintenance, and Physical Environment Comfort. These indicators generally exhibit low satisfaction but high importance, reflecting significant supply-side shortcomings. In contrast, within the ‘Maintain current supply’ decision type, some indicators are Attractive attributes with high satisfaction, while others are identified as Indifferent attributes. This suggests that fluctuations in the provision of these indicators are unlikely to significantly impact resident satisfaction, indicating that their current supply may exceed the importance residents place on them. This section will discuss this phenomenon from the perspective of the spatial and supply characteristics of Residential Pilotis and propose targeted optimization strategies.

4.1. Satisfaction Analysis Based on Supply–Demand Characteristics

The spatial characteristics of Residential Pilotis as “semi-outdoor and semi-public” spaces [6,7,8,9], together with their policy-driven production mechanisms [10], collectively shape unique supply and demand features that distinguish them from traditional urban public spaces such as parks or green areas [10]. Analyzing these characteristics helps to understand the phenomenon of supply–demand mismatch in Piloti and provides an explanation for the observed satisfaction patterns across different indicators.

From the perspective of residents’ demand characteristics. Multiple studies have indicated that, due to their spatial ambiguity as semi-outdoor and semi-public spaces, Residential Pilotis can spontaneously adapt to residents’ temporary activity needs and informal social interactions, such as elderly people supervising children or casual neighbor encounters [5,13,67]. However, the results of this study reveal that Age-Friendly Design (S4) and Resting Facilities (S1) rank among the top three in improvement priority (Table 8), both exhibiting high importance and low satisfaction. This suggests that residents’ needs for lingering and short pauses are currently underserved, and the existing spaces provide insufficient support for intergenerational activities, highlighting the need to enhance spatial activity support and inclusiveness. These findings align with research by Cao et al. They highlighted that since the residential pilotis are situated along residents’ daily circulation paths, the demand for stay points and age-inclusive facilities is particularly pronounced [5,68]. Ultimately, the availability of age-inclusive activity support and adequate resting facilities significantly affects residents’ willingness to use the space and their overall satisfaction [53,58,69].

Among the One-dimensional attributes, the Informal Social Interaction Spaces (S8) is rated as highly important, whereas the Design Promoting Social Interaction (S7) is classified as an indifferent attribute. This indicates that a pilotis where interactions occur naturally during daily activities is more satisfying to residents than a deliberately arranged formal social space [70].

Moreover, the semi-outdoor and semi-public spatial characteristics of Residential Pilotis amplify residents’ demand for indicators under the Governance and Operational Maintenance dimension [10,71]. The results indicate that residents’ needs are primarily reflected in two aspects: basic provision and management mechanism development.

First, Cleanliness and Facility Maintenance (G1), as a Must-be attribute, ranks fourth in improvement priority. Due to the semi-outdoor characteristics of pilotis, they may face greater maintenance challenges compared to enclosed public spaces. This finding corroborates the concerns of the United Nations Human Settlements Programme (UN-Habitat) regarding the “build-first, maintain-later” tendency in rapidly urbanizing areas [2]. Poor cleanliness and facility maintenance reduce space utilization and resident satisfaction [24,69], thereby undermining the sustainability of the space [21,67]. Secondly, the majority of indicators within the Governance and Operational Maintenance dimension exhibit high improvement priorities, indicating that residents’ expectations for long-term management capacity have not been fully met. This finding aligns with the research of Lui and Xu et al. in Hong Kong and other Chinese cities [29,49], suggesting that improving governance mechanisms is of universal importance for enhancing satisfaction with residential public spaces.

Although the semi-outdoor spatial characteristics of Residential Pilotis provide a greater sense of shelter compared to fully outdoor spaces, they cannot completely isolate users from climatic conditions like indoor spaces, and they exhibit certain disadvantages in terms of daylighting. The results indicate that Summer Ventilation and Winter Wind Protection (E5) and Natural and Artificial Lighting (E3) are ranked among the top eight in improvement priority. Additionally, indicators with relatively lower importance, such as Thermal Comfort (E1) and Acoustic Comfort (E2), also exhibit suboptimal satisfaction levels. These findings suggest that current Piloti designs inadequately address the environmental challenges associated with semi-outdoor spaces: aspects highly valued by residents, such as ventilation and lighting, are underprovided, while basic expectations for thermal and acoustic comfort are not sufficiently met.

From the perspective of spatial provision, the current spatial provision has effectively met residents’ needs in certain aspects. Indicators such as Greenery Configuration (L1) and Landscape Features (L2), classified as Attractive attributes, achieved high satisfaction, with improvement priority indices of 0.350 and 0.381, respectively—both below the average—indicating that the level of spatial provision meets or approaches residents’ expected levels. Notably, the “low importance” of these landscape indicators is relative rather than absolute. When basic needs such as Cleanliness and Facility Maintenance (G1) and Physical Environment Comfort (E3, E5) remain inadequately met, residents tend to prioritize these deficiencies, thereby relatively lowering the stated importance of higher-level aesthetic attributes. In contrast, functional facilities like S1 and S4, as well as physical environment aspects such as E3 and E5, exhibit high importance but low satisfaction, reflecting insufficient provision. Similar patterns have been observed in the Yang et al. study in the Yangtze River Delta region, China [45]. This study identifies a distribution pattern where landscape indicators (e.g., L1, L2) exhibit high satisfaction, whereas functional indicators (e.g., S1, S4, S6) show relatively low satisfaction. This finding aligns with Francis’ observation regarding the phenomenon of “art and aesthetics priority” in public space design [72]. Such a distribution is consistent with the developer-led resource allocation model described in existing literature—specifically, the tendency to prioritize visual aesthetics and marketing premiums while neglecting residents’ actual usage needs [7,38,39].

4.2. Optimization Strategies Based on Improvement Priorities

To address the supply–demand mismatch in Residential Pilotis and optimize resource allocation to enhance resident satisfaction, this study proposes phased and targeted optimization strategies. These strategies are informed by the improvement areas and priority rankings of spatial demand indicators identified within the “Improve current supply” and “Maintain current supply” categories.

4.2.1. Short-Term Priority Improvement Strategies

Short-term priority improvement strategies primarily target demand indicators in the “Improve current supply” category that fall within the “Prioritize upgrading” quadrant. These indicators exhibit high importance but low satisfaction and are critical for enhancing resident satisfaction, requiring prompt intervention. This study proposes strategies based on existing research findings and practical case studies.

Age-Friendly Design (S4) and Resting Facilities (S1) rank first and second in the improvement priority order and form the foundational infrastructure supporting residents’ spatial activities. Informal Social Interaction Spaces (S8) ranks seventh, with satisfaction influenced by residents’ social habits and the spatial characteristics of the Piloti. Therefore, designers and developers should incorporate age-friendly principles during the design stage, providing layered and multifunctional resting facilities [68]. For example, spatial arrangements can integrate adult resting or social areas with children’s activity zones, and universal design principles can enhance the inclusiveness of resting facilities [68,73]. Such a design enables the space to accommodate the diverse needs of all age groups, promoting intergenerational interaction and joint activities, while also creating “stay points” that facilitate temporary, informal social interactions [53,68].

Natural and Artificial Lighting (E3) and Summer Ventilation and Winter Wind Protection (E5) significantly affect the basic Physical Environment Comfort of the space, ranking sixth and eighth in the improvement priority order, respectively. Accordingly, strategies should address the semi-outdoor characteristics of Piloti by optimizing the microclimate to enhance physical comfort and improve the practical usability of the space [40].

For example, in the early design stage, local climatic conditions—such as the hot summers and cold winters in Hefei—should inform the optimization of building orientation, Piloti layout [74], and the openness of functional zones [12]. Adjustable architectural window systems can be employed to facilitate natural ventilation in summer and effective wind protection in winter [75,76,77]. To address insufficient lighting, natural daylighting should be maximized, complemented by layered artificial lighting, including task and ambient lighting, to create a safe, comfortable, and pleasant nighttime environment. These measures can effectively extend the usable hours of the space [78].

Resident Feedback Channels (G4), Cleanliness and Facility Maintenance (G1), and Management Policies and Usage Regulations (G3) rank third, fourth, and fifth in the improvement priority order, respectively. To address these weak points in governance and operations, property service providers should establish lifecycle management and maintenance systems to prevent deterioration of spatial quality due to prolonged use and insufficient upkeep [38,39]. Clear Management Policies and Usage Regulations can reduce conflicts and enhance the sustainability of facilities [38,39]. Additionally, regularized participation channels—such as residents’ councils or digital feedback platforms—can transform residents from passive recipients into co-managers of the space [79]. The United Nations Human Settlements Programme emphasizes that resident participation in management decisions is critical for achieving sustainable governance. Similarly, Carmona et al. note that sustainable governance of public spaces requires not only institutionalized management frameworks, including clear rules and defined maintenance responsibilities, but also democratic participation mechanisms that provide residents with meaningful influence over spatial decision-making [38,39].

4.2.2. Medium-Term Gradual Improvement Strategies

Medium-term improvement strategies target demand indicators in the “Improve current supply” category that fall within the “Lower value” quadrant. Although these indicators exhibit relatively low importance and satisfaction, most are classified as One-dimensional attributes, indicating potential for improvement. Therefore, strategies focus on long-term planning and continuous optimization, where moderate resource investment can enhance provision quality and achieve notable gains in resident satisfaction.

Accessibility (S6) ranks ninth in the priority order. Accessibility encompasses not only physical distance but also psychological perception and social inclusiveness, and improving it can enhance space utilization [38]. Accordingly, entrance points and signage systems should be optimized, barriers such as steps and narrow passages removed, and fully age-inclusive, barrier-free circulation ensured [2]. Visual permeability can be enhanced through interface design to avoid feelings of enclosure and exclusion [19].

Regarding Thermal Comfort (E1) and Acoustic Comfort (E2), which rank 11th and 10th, respectively, in terms of improvement priority, medium-to-long-term passive optimizations can be adopted in consideration of Hefei’s distinct seasonal characteristics: shading structures to mitigate summer heat; vegetation as light and sound buffers, with deciduous trees providing summer shade and winter sunlight penetration [53]; vertical green walls or low hedges to improve local microclimate and reduce noise [78]; and functional zoning to separate quiet resting areas from active zones [78].

Finally, although Design Promoting Social Interaction (S7) is classified as an indifferent attribute, moderate improvements can help stimulate spatial vitality. Based on the preceding analysis, Residential Pilotis should avoid deliberately constructed formal social spaces and instead create natural gathering points through mixed-use functional layouts [70].

4.2.3. Long-Term Maintenance or Flexible Adjustment Strategies

Long-term maintenance or flexible adjustment strategies target demand indicators in the “Maintain current supply” category, including those in the “Keeping on” or “Over-provisioning” quadrants.

For the Landscape and Visual Experience dimension, a tiered improvement strategy should be established. External Landscape Views (L3) and Visual Aesthetics and Coherence (L4), as One-dimensional attributes, can be maintained through regular evaluation to ensure continued alignment with residents’ expectations [80]. In contrast, Greenery Configuration (L1) and Landscape Features (L2) are currently over-provisioned, and resources allocated to these indicators can be adjusted accordingly. For instance, low-maintenance ecological practices, such as selecting native plant species, reducing pruning frequency, and minimizing capital and operational expenditures, can be adopted [48,81].

For governance-related indicators, such as Public Disclosure of Fund Utilization. (G2) and Resident Participation in Decision-Making (G5), although these indicators have achieved high satisfaction, their provision levels are near the threshold, and fluctuations could significantly impact satisfaction. Carmona et al. emphasize that sustainable governance of public spaces relies on ongoing coordination mechanisms and resource investment [39]. Therefore, these governance mechanisms should be institutionalized, with regular information disclosure, participatory decision-making channels, and periodic evaluation to adapt to evolving resident needs [2,82,83].

The study also indicates that Spatial Flexibility (S5) is classified as an Indifferent attribute. This may be because Piloti, as open spaces, inherently possess basic flexibility. Thus, deliberately enhancing flexibility is unlikely to significantly affect satisfaction. Stevens et al. note that adaptability in public spaces primarily arises from users’ spontaneous adjustments, and excessive functional planning may even disrupt residents [84]. Accordingly, design should maintain openness, with careful adjustments to planning strategies and governance mechanisms based on periodic evaluation, directing resources toward indicators of greater resident concern [85].

5. Conclusions

This study adopts a resident demand-centered approach, using resident satisfaction as an entry point to examine the supply–demand mismatch in Residential Pilotis spaces and to propose improvement strategies. First, through literature review and factor analysis, a resident satisfaction assessment framework was developed and validated, comprising four dimensions—Spatial Usability and Sociality, Landscape and Visual Experience, Physical Environment Comfort, and Governance and Operational Maintenance—with a total of 23 indicators. Second, the Integrated KANO-IPA Model was applied to analyze 553 valid questionnaires from Hefei, identifying the quality attribute category of each demand indicator and clarifying decision types and improvement priorities using a priority index. Finally, based on the characteristics of supply and demand, phased optimization strategies were proposed, including short-term priority improvements, medium-term gradual enhancements, and long-term maintenance or flexible adjustments.

The study reveals structural characteristics of the supply–demand mismatch in Residential Pilotis. From the demand perspective, the semi-outdoor and semi-public spatial characteristics of Piloti intensify residents’ needs for facility maintenance, environmental comfort, and participatory governance; however, current provision inadequately addresses these challenges. From the supply perspective, the study identifies a degree of imbalance in resource allocation. Existing spaces underperform in terms of resident satisfaction regarding Must-be and high-priority One-dimensional attributes (e.g., cleanliness and maintenance, resting facilities, and age-friendly design), necessitating an urgent improvement of the current provision. In contrast, the supply for Attractive attributes (e.g., landscape features and greenery arrangements) has already met residents’ expectations. Under the current development and construction model, if developers prioritize visual aesthetics while relatively neglecting actual usage needs, it may further exacerbate the supply–demand mismatch. Notably, the Governance and Operational Maintenance dimension performs weakly overall, reflecting a certain degree of “build-first, manage-later” tendencies in current space production. Such supply–demand mismatches not only reduce resident satisfaction but also lead to inefficient allocation of public resources.

This study developed an operational resident satisfaction assessment framework for Residential Pilotis and introduced the Integrated KANO-IPA Model into this domain. By combining demand classification with priority quantification, the study systematically identifies critical points of the supply–demand mismatch and proposes phased improvement strategies. The framework provides an empirical decision-making tool for Piloti design: the assessment framework can diagnose the degree of supply–demand alignment, the priority ranking can guide optimized resource allocation, and the phased strategies support incremental improvements. Although the study is based on empirical data from Hefei, the evaluation methodology and analytical framework offer a reference for other high-density cities facing similar challenges.

This study has several limitations. In terms of spatial scope, the sample is concentrated in Hefei, so the applicability of findings to different climatic regions, stages of urban development, or older communities remains to be tested. Temporally, data were collected at a single point in spring, limiting the capture and analysis of seasonal variations and dynamic evolution of resident needs. Methodologically, the online questionnaire method may result in underrepresentation of groups who do not frequently use the internet. This might create a disparity between the demographic information of the respondents and the actual users of the Residential Pilotis, and lacks validation from objective data such as behavioral observations of residents. Additionally, while this study benefits from a large sample size that captures general trends across Hefei, we did not strictly control for community-level heterogeneity such as community age, construction cost, property management quality, or specific spatial layouts. Future research could expand in four directions: (1) broadening the range of case cities, particularly including different climatic zones and stages of urban development; (2) adopting longitudinal study designs to explore seasonal fluctuations and long-term evolution of resident preferences; (3) integrating multiple methods, such as space syntax analysis and post-occupancy evaluation (POE), to construct a comprehensive framework combining subjective and objective assessments; and (4) employing multi-level modeling approaches to quantitatively analyze how community-level heterogeneity factors moderate resident satisfaction.