A Study on the Perception Evaluation of Public Spaces in Urban Historic Waterfront Areas Based on AHP–Cloud Modelling: The Case of the Xiaoqinhuai Riverside Area in Yangzhou

Abstract

With the acceleration of global urbanisation, the pace of evolution in urban waterfront areas has intensified, consequently hastening the renewal rate of their constituent public spaces. Compared to the macro-level planning and regulation of traditional port and coastal waterfronts, balancing the historical preservation of urban heritage waterfront public spaces with contemporary demands has emerged as a critical issue in urban regeneration. This study examines the historical waterfront area of the Xiaoqinhuai River in Yangzhou, establishing a public space perception evaluation framework encompassing five dimensions: spatial structure, landscape elements, environmental perception, socio-cultural context, and facility systems. This framework comprises 33 secondary indicators. The perception assessment system was developed through a literature review, field research, and expert interviews, refined using the Delphi method, and weighted via the Analytic Hierarchy Process (AHP). Finally, cloud modelling was employed to evaluate perceptions among residents and visitors. Findings indicate that spatial structure and socio-cultural dimensions received high perception ratings, highlighting historical layout and cultural identity as strengths of the Xiaoqinhuai Riverfront public space, while significant shortcomings were noted in terms of landscape elements, environmental perception, and facilities. These deficiencies manifest primarily in limited vegetation diversity, inadequate hard paving and surface materials, insufficient landscape node design, poor thermal comfort, suboptimal air quality and olfactory perception, uncomfortable resting facilities, limited activity diversity, and inadequate slip-resistant surfaces. Further analysis reveals perceptual differences between residents and visitors: the former prioritise daily living needs, while the latter emphasise cultural experiences and recreational facilities. Based on these findings, this paper proposes targeted optimisation strategies emphasising the continuity of historical context and enhancement of spatial inclusivity. It recommends improving public space quality through multi-dimensional measures including environmental perception enhancement, landscape system restructuring, and the tiered provision of facilities. This research offers an actionable theoretical framework and practical pathway for the protective renewal, public space reconstruction, and optimisation of contemporary urban historic waterfront areas, demonstrating broad transferability and applicability.

1. Introduction

Water is the origin of human life, and waterfront areas serve as vital conduits for the rapid advancement of human civilisation. In the present era of accelerated societal development and rapid urbanisation, waterfront zones have become integral components of the modern urban spatial system, with their public spaces forming a crucial part of the city’s overall public space framework [1]. Urban waterfronts fulfil multiple functions, including transportation, commercial exchange, cultural memory, and landscape experience [2]. Driven by urbanisation, these areas have progressively transformed from traditional productive spaces into vital venues for public life. Globally, from London’s Docklands [3] to New York’s Hudson River Waterfront [4] and from Barcelona’s Olympic Seafront [5] to Seoul’s Cheonggyecheon Stream [6], the continuous optimisation and enhancement of urban waterfronts have become pivotal issues in international urban development. This transformation not only drives economic revitalisation and landscape improvement but also fosters social dynamism and ecological restoration [7], maintaining enduring attention in global academic research and planning practice.

Breen and Rigby define urban waterfronts as settlements at the water’s edge of varying scales. Water may take the form of rivers, lakes, oceans, bays, streams, or canals, encompassing scales ranging from brooks to major international ports [8]. Waterfronts manifest in numerous forms, such as coastal, harbour, riverside, and lakeside areas [9]. Consequently, the attributes and typology of urban waterfronts frequently depend on a city’s location and its interconnected water systems. Coastal cities, for instance, possess coastal waterfronts, harbour waterfronts, and industrial waterfronts [10]. Cities developed adjacent to major inland water systems possess extensive riverfront and lakeside waterfront areas; many inland cities also feature medium- to small-scale urban waterfronts based on their internal water systems.

In recent years, the urbanisation processes and urban renewal methodologies of developing nations, exemplified by China, have undergone a profound shift from ‘incremental expansion’ towards ‘optimisation of existing stock’ [11]. Within this transition, society increasingly prioritises the quality of existing urban spaces [12] over blind and endless expansion. The optimisation and enhancement of urban waterfront areas have garnered significant attention within this process [13]. These waterfront zones serve as focal points for showcasing a city’s spirit and character, attracting substantial resident and tourist engagement [14]. Such waterfront public spaces play a pivotal role within the urban spatial system. They not only function as ecological transitional lands but also serve as vital connecting points between nature and urban populations [15]. These areas provide ideal venues for diverse groups to engage in transportation, trade, recreation, and tourism activities [16], while also offering crucial respite spaces to alleviate the pressures of high-density urban living [17]. Concurrently, historic waterfront public spaces within the core heritage districts of certain cultural towns merit particular attention and study. Compared to ordinary urban public spaces, these historic waterfront areas possess a more complex dual nature: firstly, they constitute vital components of the city’s historical fabric and cultural heritage, bearing local memory and identity; secondly, they are often situated within the historic urban core, facing multiple pressures including intensive land use, public space regeneration, and tourism development. Finally, and most crucially, within contemporary urban contexts, these historic waterfronts simultaneously function as habitats for indigenous communities and increasingly popular tourist destinations. Unlike static heritage zones, such waterfronts are frequently in a state of “renewal process”—continuously undergoing the integration of modern functions and spatial transformations to accommodate prevailing cultural tourism development trends, while simultaneously preserving historical-cultural elements and sustaining indigenous living environments. This dynamic state not only lends unique research significance but also generates tensions between conservation and development, residents and visitors, and publicness and commercialisation. Particularly pertinent is how public spaces within historic urban waterfronts can simultaneously satisfy the needs of two core user groups—indigenous residents and tourists—whose usage objectives and perceptual priorities diverge markedly during the renewal process. Consequently, conducting in-depth research into the public spaces of urban historic waterfront areas undergoing renewal, exploring their relationship with residents and tourists, and identifying strategies and methods for these spaces to meet the practical needs of their core users holds significant theoretical value and practical relevance.

2. Literature Review

2.1. Review of Research Related to Urban Waterfront Areas

Waterfront areas, as the interface between cities and water bodies, constitute vital spaces for human production and daily life. With the decline of industrialisation and port cities in earlier times, these zones gradually became focal points for urban renewal and regeneration [18], attracting sustained research across disciplines including architecture, urban and rural planning, sociology, economics, management studies, and history. Academic circles have conducted foundational research on urban waterfronts from macro perspectives such as conceptualisation [1], origins [19], classification [2], historical evolution and developmental history [20]. This foundational research has provided a robust basis for subsequent scholars to explore further aspects such as the evolution of urban waterfront planning and reconstruction [21], development utilisation systems and models [7], comprehensive planning regulations [10], and assessments of economic benefits alongside the complexity of usage contexts [13], gradually forming a systematic body of empirical knowledge. Entering the 1990s, the concept of sustainable development gained widespread acceptance. Scholars began to question the sole reliance on economically driven redevelopment models for waterfront areas. They shifted focus to exploring the fundamental attributes of urban waterfronts from the perspectives of stakeholders and core user groups. Research examined how foundational characteristics such as accessibility [22], connectivity and comfort [23], publicness [24], inclusivity [25], and functionality [26]. Additionally, the social context composition and cultural ecosystem operation of waterfront areas [27] have garnered increasing attention. Since the 21st century, more nuanced research topics have emerged, including public space [25], climate adaptation [28], pathways to carbon neutrality [29], the impact of private developments on waterfront publicness [18], localised design [30], and the integration of public participation systems into waterfront planning [31]. Overall, research on urban waterfronts has shifted from focusing on spatial materiality, land planning and utilisation, urban spatial shaping, and promoting economic benefits towards examining the relationship between space and urban public life. Waterfronts are now regarded as vital spatial vehicles for addressing environmental challenges, enhancing urban resilience, and rebuilding social connections. Research exhibits a trend towards refinement and diversification, progressing from macro-level to meso-level and multidimensional studies, and shifting from single-discipline to interdisciplinary approaches. While substantial theoretical and practical experience has been accumulated, research focus often centres on large-scale ports, bays, or river basins, with insufficient attention given to small-scale, everyday waterfront fragments. This presents scope for more focused subsequent research.

2.2. Review of Research on Public Spaces in Urban Waterfront Areas

In research concerning urban waterfront areas, public spaces have progressively emerged as a significant focus within academic circles due to their social, ecological, and cultural functions. Waterfront public spaces represent not only the crystallisation of interactions between natural and human elements, but also serve as vital venues for diverse groups to share, interact, and experience. Characterised by heightened openness, ecological significance, and symbolic value, they play a unique role in urban spatial planning and sustainable development agendas.

Firstly, in terms of research themes and subjects, studies primarily focus on the vitality [14,32,33,34] and quality assessment [35] of urban waterfront public spaces, accessibility evaluation [36,37], park waterfront environments [38,39], landscape elements of waterfront public spaces [40,41], public space climate [28,42,43,44,45,46,47], and the relationship and synergistic mechanisms of blue-green spaces [48,49]. Among these, public green spaces within waterfront areas, multidimensional vitality assessments, and thermal environmental adaptability are key focal points.

Secondly, from the perspective of public space users, research primarily explores user behaviour and perceptions. Scholars have systematically evaluated diverse perceptions of urban waterfronts based on user auditory [50,51] and olfactory [52] feedback. Emotional responses [53,54,55,56] and mental health assessments [57,58] are also key research focuses, with integrated physiological and psychological evaluation systems for waterfront users becoming central considerations in planning and design. User satisfaction serves as a direct indicator of urban waterfront quality. Consequently, research has examined user perceptions of vegetation and surface material quality [39], overall imageability [59], landscape and green space quality [60], coverage of public sports facilities along the waterfront [61], adequacy of lighting infrastructure [62], functional comprehensiveness [37], residential suitability [63], and climatic amenity [64]. This has fostered the development of comprehensive and systematic multi-dimensional evaluation frameworks, with particular attention directed towards specific user groups such as the elderly [65] and children [66].

Finally, the application of theory and technology in urban waterfront public space research warrants attention. Scholars employ mature methodologies including Geographic Information Systems (GIS) such as ArcGIS [65], spatial syntax [22], multi-source data analysis [35], social media analysis [67], Street View imagery [50], and remote sensing spatiotemporal image analysis [29]. Concurrently, they have integrated emerging technologies such as machine learning and artificial neural networks [36], digital twin technology and virtual scenario construction [68], thermal environment metrics [47], and mobile phone signalling tracking [69], thereby establishing a rich technical repository. Research methodologies concerning the behaviour, perceptions, and needs of users in waterfront public spaces have evolved from relying solely on questionnaire surveys and behavioural observation techniques. They now incorporate more objective and scientific approaches combining instruments such as electroencephalography (EEG) [58], electrocardiography (ECG) [53], and eye-tracking devices [70], alongside mathematical analysis models including the Analytic Hierarchy Process (AHP), entropy weight method [71], and structural equation modelling (SEM) [72].

Overall, research on public spaces within urban waterfront areas has evolved from early evaluations of spatial form and landscape elements to encompass assessments of user behavioural perceptions, emotional responses, and needs. This progression has established a relatively comprehensive research framework. However, existing studies predominantly focus on isolated spatial elements and singular metrics, lacking a holistic, multidimensional framework for evaluating overall value. Concurrently, research on diverse user groups remains fragmented, with insufficient systematic quantitative comparisons of user differences. Consequently, a comprehensive and scientifically rigorous integrated analytical model has yet to be developed. These shortcomings provide fresh entry points for the present study. Moreover, among the numerous investigations into waterfront public spaces, research on urban historic waterfronts—characterised by their historical and cultural attributes alongside dynamic renewal processes—remains relatively underdeveloped. Public space studies in this domain urgently require deeper exploration.

2.3. Review of Research on Public Spaces in Historic Urban Waterfront Areas

Urban historic waterfront public spaces represent a distinctive spatial typology where heritage and contemporary regeneration converge. They bear rich historical and cultural significance alongside local memory, serving as vital conduits for showcasing a city’s cultural depth, street fabric, and symbolic identity. Simultaneously, these areas undergo continuous renewal and functional transformation, frequently becoming sites where tourism development overlaps with residents’ daily lives. They thus exhibit the dual characteristics of possessing ‘heritage attributes’ while undergoing ‘dynamic renewal’.

However, compared to research on modern urban waterfronts, studies concerning the public spaces of historic urban waterfronts remain relatively limited, with existing work predominantly focused on the preservation of historical and cultural heritage and tourism development. Relevant scholars have explored historical and cultural transformations [73,74,75], the preservation, development [59], and sharing [76] of cultural heritage and historical legacies [21,75], emphasising spatial landscape restoration, historical authenticity, and cultural transmission. Yet, in-depth discussions on the holistic value and comprehensive evaluation of public spaces remain scarce. Concurrently, tourism within urban historic waterfront districts has drawn scholarly attention. Research has examined tourism sustainability [77], public space vitality [34,78], visual accessibility of historic heritage [79], and thermal environmental comfort [42] within these areas. At the user level, while some studies acknowledge differences between residents’ and tourists’ needs, these primarily rely on qualitative descriptions or case studies, lacking systematic quantitative comparative frameworks. Such disparities often generate conflicts between publicness and commercialisation, yet remain unsupported by effective theoretical models or methodological tools. Methodologically, research predominantly favours single approaches, with insufficient interdisciplinary integration and a lack of unified analytical frameworks capable of simultaneously addressing heritage conservation, safeguarding publicness, and coordinating diverse group requirements.

Overall, research into public spaces within historic urban waterfront areas is currently in its nascent and transitional phase. Their unique dual heritage and regeneration attributes determine their research value, yet existing studies remain deficient in terms of systematic approaches, analysis of user diversity, and methodological integration. This gap provides ample scope for the present research and creates an entry point for this case study.

In summary, drawing upon the background analysis and literature review, this study endeavours to address the following three questions:

(1)

What factors influence the quality of public spaces within historic urban waterfront areas?

(2)

What is the overall perceived evaluation of these influencing factors by the primary users of urban historic waterfront public spaces—residents and visitors?

(3)

Based on the primary users’ comprehensive evaluation of these factors, how should the public spaces of urban historic waterfronts be optimised and enhanced in the future?

3. Study Area

Yangzhou City, situated in central Jiangsu Province, boasts a history spanning 2500 years. It is one of the 24 cities designated as the first batch of National Historical and Cultural Cities under the Law of the People’s Republic of China on the Protection of Cultural Relics [80], and among the first cities to receive the titles of ‘China’s Outstanding Tourist City’ and ‘National Model City for Tourism Standardisation’. As a quintessential Jiangnan water city, the ancient city of Yangzhou features a dense network of rivers and crisscrossing waterways, forming a spatial pattern where ‘waterways string the city together, and streets and alleys follow the water’. The historical waterfront area of the Xiaoqinhuai River occupies the heart of Yangzhou’s ancient city (Figure 1). Spanning approximately 1.98 km, the Xiaoqinhuai is the sole surviving inner-city waterway within the ancient city walls. Originally measuring around 25 m in width, the river now stands approximately 12 m wide. This study focuses on the northern section of the Xiaoqinhuai River, spanning approximately 700 m from Wenchang Middle Road to Ganquan Road (Figure 2). Four bridges currently span the waterway, flanked by ageing residential buildings and established neighbourhoods. The area also incorporates renovated guesthouses, cultural spaces, and boutique hotels, creating a diverse mixed-use environment blending residential living with tourism. The historic waterfront area of the Xiaoqinhuai River lies between the Renfengli and Wanzi Street historic cultural quarters, serving as a buffer zone between these two core ancient city districts. It possesses both residential attributes and cultural tourism potential. In 2023, the Xiaoqinhuai River conservation and renewal project was designated among Jiangsu Province’s inaugural urban renewal pilot schemes, with its northern section planned as a tourism corridor [81]. On 4 January 2025, the Xiaoqinhuai district officially opened as a provincial urban renewal pilot project. Guided by the principles of ‘small-scale, incremental, and micro-renewal’, five demonstration nodes underwent conservation and revitalisation. Along the waterfront, historic buildings were restored, repurposed, or converted for cultural, exhibition, and commercial uses [82]. This approach enabled traditional riverside living to coexist and intertwine with modern urban vitality, forging a novel historical waterfront spatial pattern characterised by ‘water linking vistas and culture enhancing tourism’. The Xiaoqinhuai River Historic Waterfront District not only reflects the spatial evolution logic of Yangzhou’s ancient city under the concept of ‘protective regeneration,’ but also provides a typical case study for exploring how to balance residents’ living needs and tourist experiences in the renewal process of historic waterfront public spaces.

Figure 1. Geographical location of the study area (self-drawn by the author).

Figure 2. Research area starting point and endpoint (self-drawn by the author).

4. Materials and Methods

4.1. Research Framework

This research takes the historic waterfront area of Yangzhou’s Xiaoqinhuai River as its case study, focusing on the differing perceptions of residents and visitors towards public spaces in historic urban waterfront areas during the renewal process. It establishes a systematic public space evaluation framework and formulates enhancement and optimisation strategies specifically for public spaces in historic urban waterfront areas. The research comprised three phases: Phase One established a scientifically sound, multi-dimensional evaluation framework for historic waterfront public spaces through a systematic literature review, field investigations, group interviews, and the Delphi expert consultation method. The second stage employed the Analytic Hierarchy Process (AHP) for computational analysis to derive indicator weights. The third stage utilised a cloud model based on these weights to conduct quantitative analysis and fuzzy processing of residents’ and visitors’ perceived satisfaction, identifying perceptions and disparities in public space quality among different user groups. Finally, integrating the model outcomes, optimisation and enhancement strategies for historic waterfront public spaces were proposed. This established a framework of ‘indicator system construction–weight determination and integration–perceptual evaluation and optimisation strategies’, systematically synthesising comprehensive evaluation methods to provide a quantifiable and replicable analytical pathway for researching the quality of historic waterfront public spaces (Figure 3).

Figure 3. Research Framework Diagram (self-drawn by the author).

4.2. Preliminary Establishment of the Evaluation Indicator System

Urban historic waterfront public spaces constitute a complex multi-factor integrated system, influenced by geographical location, waterfront landscapes, historical and cultural context, user perceptions, and supporting facilities, each presenting distinct dominant factors. Consequently, the selected spatial demand evaluation indicators must be comprehensively considered to avoid duplication, while accounting for the unique characteristics of historic sites to ensure no critical indicators are omitted.

To this end, this study first conducted a literature review, examined exemplary design cases from historic waterfront areas globally, and analysed relevant policy documents. Through integration and categorisation, 40 objective evaluation indicators were preliminarily selected (Table 1). Concurrently, to ensure the relevance and locality of the predefined indicators to the study site, continuous field research and surveys were conducted along the Xiaoqinhuai River waterfront from 20 to 25 August 2025, gathering baseline condition data and on-site photographic evidence. The research team divided into pairs to conduct behavioural observations of public space users along both banks of the Xiaoqinhuai River, alongside semi-structured interviews based on the pre-selected indicators. Pre-designed interview outlines guided respondents in supplementing the indicators (Table 2), with all interviews documented via audio recording and written transcripts. All 23 interviewees (Appendix A Table A1) provided informed consent for participation and shared basic personal details within acceptable limits. This approach mitigated sample bias, ensuring the rationality and balance of the interview cohort. Respondents comprised 10 residents and 13 visitors, including 12 males and 11 females aged between 22 and 76 years, achieving an ideal demographic equilibrium (Table 3). The interview content was organised, summarised, and filtered to derive four indicators: ‘River Crossing Accessibility’, ‘Comprehensive Anti-Slip Measures’, ‘Spatial Accessibility’, and ‘Shared Sports Spaces and Facilities’. These complement the previously selected objective indicators.

Table 1. Source literature for indicators.

Number	Indicator	Source Literature	Number	Indicator	Source Literature
B1	Spatial Connectivity	Amerian (2021) [83]	B21	Air Quality and Olfactory Perception	Han (2025) [52]
B2	Spatial Hierarchy and Sequence	Chen (2024) [15]	B22	Climate Adaptability	Xi (2025) [28]
B3	Spatial Legibility	Ding (2023) [34]	B23	Ecological Friendliness	Durán Vian (2021) [2]
B4	Functional Variety	Yu (2025) [50]	B24	Cleanliness and Maintenance Status	Gallego (2014) [84]
B5	Spatial Safety	Zhang (2024) [85]	B25	Historical and Cultural Expression	Li (2025) [27]
B6	Accessibility and Connectivity	Luo (2023) [22]	B26	Sense of Place and Local Identity	Han (2025) [59]
B7	Integrity of Historical Layout	Zhu (2024) [76]	B27	Resident Cultural Affiliation	Durán Vian (2018) [31]
B8	Scale Appropriateness	Yin (2025) [38]	B28	Visitor Cultural Experience	Chen (2021) [86]
B9	Waterfront Quality	Li (2024) [29]	B29	Community Connectivity	Hagerman (2007) [19]
B10	Vegetation Diversity	Lyu (2025) [39]	B30	Social Inclusion	Attia (2018) [25]
B11	Paving and Material Quality	Li (2023) [40]	B31	Public Accessibility and Spatial Equity	Yuan (2024) [20]
B12	Landscape Nodes	Huang (2025) [36]	B32	Cultural Events and Festivities	Huang (2024) [87]
B13	View Corridor and Visual Axis	Huang (2024) [87]	B33	Sanitary Facility Adequacy	Li (2024) [53]
B14	Historic Landscape Conservation	Ye (2025) [74]	B34	Recreational Facility Comfort	Fan (2023) [33]
B15	Green Infrastructure Integration	Shi (2025) [72]	B35	Accessibility Facility Completeness	Attia (2018) [25]
B16	Colour and Aesthetic Harmony	Liu (2022) [88]	B36	Wayfinding and Information Systems	Zhao (2023) [89]
B17	Landscape Aesthetic Experience	Zhou (2023) [41]	B37	Activity Facility Diversity	Mansournia (2021) [66]
B18	Thermal Comfort	Xi (2023) [46]	B38	Safety and Emergency Facilities	Gong (2019) [63]
B19	Lighting Quality	Han (2025) [52]	B39	Smart and Digital Management Infrastructure	Cai (2025) [58]
B20	Acoustic Quality	Zhu (2023) [51]	B40	Management, Maintenance, and Operational Mechanisms	Wang (2024) [71]

Table 2. Interview Outline.

Interview Focus	Question Number	Question Content
Exploring the spatial layout, scale, connectivity and safety of waterfront areas	1	Do you find the spatial layout and pedestrian routes along the Xiao Qinhuai Riverside area clear and convenient?
	2	Did you perceive a sense of spatial layering or variation in key nodes during your visit?
	3	How would you rate the safety provisions in the riverside area (night-time lighting, protective facilities, etc.)?
Understanding the harmony between natural and cultural landscape elements and the resulting visual experience	4	How would you rate the water quality and greenery within the riverside area?
	5	Do you find the landscape design (vegetation, colour palette, architectural character) harmonious and appealing?
	6	Were there any particular landscapes or elements that left a strong impression?
Understanding subjective perceptions of comfort, environmental quality and maintenance conditions	7	Did you feel comfortable during your activities in the waterfront area (climate, lighting, noise, air quality, etc.)?
	8	How would you rate the cleanliness and maintenance of the waterfront area?
	9	Do you feel environmental changes (weather, seasons) affect your willingness to linger or engage in activities here?
Exploring Local Cultural Identity and Experiential Perception	10	Do you believe the Xiao Qinhuai River embodies Yangzhou’s historical and cultural character?
	11	Do you sense local memory or cultural atmosphere within this space?
	12	Have you participated in or observed cultural events or festivals along the waterfront? Were they appealing to you?
Assess the completeness and ease of use of public facilities	13	Do you consider the waterfront’s recreational, lighting, signage, and sanitary facilities adequate?
	14	Do you find the layout of public amenities well-planned and user-friendly?
	15	What additional services or facilities do you believe should be further improved?
Collection of Basic Personal Information; Personal Information Collection	16	What is your identity, age, occupation, and level of education?
Guiding respondents to provide supplementary information	17	Beyond the above questions and focus areas, what other aspects do you consider important or lacking?

Table 3. Profile of Respondents.

Total Number of Respondents: 23
Project	Category	Quantity	Percentage
Population category	Tourist	13	56.52%
	Resident	10	43.47%
Gender	Male	12	52.17%
	Female	11	47.82%
Age	18–30	4	17.39%
	31–40	5	21.73%
	41–50	5	21.73%
	51–60	5	21.73%
	>60	4	17.39%

The selection of predefined evaluation indicators in this study is grounded in the characteristics of urban historic waterfront public spaces. These indicators primarily derive from three sources: (1) exemplary design cases, design guidelines, and standardised policies for historic waterfront areas globally; (2) relevant literature on the optimisation and design of public spaces within urban historic waterfront zones; (3) field research, behavioural observations, and guided semi-structured interviews with space users conducted in the sample study area of the Xiaoqinhuai River historic waterfront public space. This study aims to extract elements defining urban historic waterfront public spaces, translate these into assessable metrics, and integrate them into a comprehensive perception evaluation framework for such spaces. By screening high-frequency elements from sources (1) and (2), extracting common factors, and combining supplementary indicators from source (3), 44 preset indicators were ultimately identified. These were organised into five dimensions: ‘Spatial Structure Dimension’, ‘Landscape Elements Dimension’, ‘Environmental Perception Dimension’, ‘Socio-Cultural Dimension’, and ‘Facility Systems Dimension’ (Table 4).

Table 4. Preset Evaluation Indicator System.

First-Level Dimension	Secondary Indicators	Indicator Explanation
B1 Spatial Structure Dimension	C1 Spatial Connectivity	Accessibility and continuity of waterfront promenades, alleyway systems and internal urban roads
	C2 Spatial layering	The interplay of spatial openings and closures with the rhythmic quality of the pedestrian experience
	C3 Spatial recognisability	Clarity of Landmark Elements and Path Guidance
	C4 Spatial diversity	Functional integration and spatial combination in multi-scenario environments
	C5 Spatial Safety	Line-of-sight safety, night-time illumination, protective installations and sense of boundaries
	C6 Spatial accessibility	Internal road accessibility, public transport convenience, and connectivity with external spaces
	C7 Spatial Integrity and Continuity of Historical Layout	Continuity of the historical street fabric and waterfront spatial layout
	C8 Spatial scale appropriateness	Matching of spatial scale to human activity
	C9 River-crossing accessibility	The ease of access between both riverbanks
	C10 Spatial accessibility	The speed of locating the main entrance to the waterfront area
B2 Landscape Elements Dimension	C11 Water Landscape Quality	Water body cleanliness, accessibility and visual appeal
	C12 Vegetation landscape diversity	Native plant species, ecological function and seasonal variation
	C13 Hard surfacing and surface materials	Textural quality, slip resistance and historical context harmony of paving materials
	C14 Landscape Node Design	Aesthetic and functional design of waterside platforms, viewing decks and plazas
	C15 Landscape Viewing Areas and Corridors	Continuity of sightlines and organisation of key visual nodes
	C16 Preservation of historical landscape elements	Conservation and adaptive reuse of heritage elements including historic wharves, quays, bridges and stone inscriptions
	C17 Green Infrastructure Integration	Integration of rain gardens, ecological revetments and vegetated buffer zones
	C18 Landscape Colour and Character Coordination	Colour coordination between architecture and landscape, ensuring overall aesthetic coherence
B3 Environmental Perception Dimension	C19 Landscape Aesthetic Experience	Spatial aesthetics, imagery appeal and aesthetic enjoyment
	C20 Thermal environmental comfort	Microclimate suitability, shade provision, ventilation and pedestrian comfort
	C21 Lighting environment quality	Balance between night-time illumination and safety lighting levels
	C22 Acoustic Environment Quality	Noise control, natural soundscapes and tranquil ambience experience
	C23 Air Quality and Olfactory Perception	Air cleanliness, odour perception and water body odour management
	B24 Climate resilience	Climate-responsive design adaptations (ventilation, water circulation, etc.)
	B25 Ecological Friendliness	Biodiversity, ecological corridors and sustainable material application
	B26 Space Cleaning and Maintenance Status	Spatial environmental cleanliness and routine maintenance frequency
B4 Socio-Cultural Dimension	B27 Historical and Cultural Expression	Cultural symbol display, historical narrative and local tradition continuity
	B28 Place Memory and Sense of Place	Uniqueness of historical sites and sense of emotional belonging
	B29 Local cultural identity	Acceptance and recognition of spatial transformations
	B30 Immersion in local culture	Immersive experiences of local culture and traditional crafts
	B31 Community Interaction	Frequency of residents’ daily social interactions and community activities
	B32 Social inclusion	Accessibility for diverse groups (children, elderly, disabled persons)
	B33 Publicness and Spatial Equity	Whether spaces are monopolised by commercial facilities and equality of public access
	B34 Cultural Activities and Festivals	Richness of festive events, exhibitions and cultural happenings
B5 Facility System Dimension	B35 Sanitation facilities completeness	Public toilet facilities, litter bins, seating, drinking fountains and other amenities
	B36 Comfort of Resting Facilities	Seating comfort, sunshade installations and waterside leisure facilities
	B37 Accessibility facilities completeness	Accessible ramps, tactile paving and public transport connectivity
	B38 Wayfinding and Information System	Signage clarity, multilingual guidance and information accessibility
	B39 Diversity of recreational facilities	Diversity of recreational, fitness and children’s play facilities
	B40 Safety and Emergency System	Firefighting equipment, surveillance systems, lifebuoys and other safety provisions
	B41 Shared Sports Facilities and Spaces	Dedicated public fitness spaces and supporting facilities
	B42 Completeness of anti-slip facilities	Surface slip resistance for pedestrian use in damp conditions
	B43 Smart and Digital Management Facilities	Smart amenities including intelligent lighting, guide applications, and environmental monitoring systems
	B44 Management, Maintenance and Operational Mechanisms	Facility maintenance frequency, cleaning management, and spatial governance efficiency

4.3. Delphi Method Integration and Optimisation

4.3.1. Survey Design and Expert Consultation

Following the preliminary extraction of indicators through systematic literature review, field research, and semi-structured interviews, it remains necessary to further validate the scientific rigour, representativeness, and operational feasibility of each indicator. This will enhance the assessment framework’s precision and practical guidance value. Refining the framework through the rational judgement of an expert panel enhances the scientific rigour, universality, and adaptability of the evaluation system, ensuring it draws upon both prior academic experience and local context while possessing methodological reliability. To this end, this paper employs the Delphi Method for expert consultation.

The Delphi Method is a qualitative research technique [83]. It constitutes a structured consensus-building technique that acquires and refines expert opinions through iterative feedback [85]. This approach relies on a structured panel of experts who respond to questionnaires across multiple rounds; through repeated feedback and revision, expert opinions gradually converge to form statistically significant collective judgements [87,88]. The Delphi Method has been extensively applied in constructing evaluation indicator systems across diverse fields [84,86,89].

The selection of panel members is pivotal to the Delphi method. A scientifically sound Delphi panel should comprise between 10 and 50 experts who are familiar with the research topic and possess at least five years of professional experience in the relevant field [90]. In this study, the expert panel comprised 20 members across three distinct categories (Appendix A Table A2). The first group (Group A) included six government administrators engaged in or involved with urban planning and waterfront development and utilisation. The second group (Group B) comprised nine university lecturers and scholars specialising in the conservation and regeneration of urban public spaces and waterfront areas. The third group (Group C) comprised five designers from design institutes engaged in the design and practice of historic and cultural district renewal. Existing research indicates that expert consensus using the Delphi method is typically achieved within two rounds of consultation [91]. Consequently, this study conducted two rounds of expert consultation from 1 to 5 September 2025.

4.3.2. Expert Feedback and Indicator Revision

This study calculated the arithmetic mean (hereafter referred to as ‘M’), standard deviation (here after referred to as “SD”), and coefficient of variation (hereafter referred to as ‘CV’) to evaluate expert scoring. These statistical indicators are widely employed to assess and quantify the degree of consensus in expert judgements [92,93]. M is used to evaluate central tendency; an M > 3.75 indicates that a particular category or indicator possesses sufficient significance. SD is used to evaluate the degree of convergence; where SD < 1 indicates high reliability of the indicator. CV measures dispersion, with CV < 25% signifying strong expert consensus on a specific item [93]. This study defined M > 3.75, SD < 1, and CV < 0.25 as conditions for consensus attainment. The overall coherence of expert opinions was assessed using Kendall’s coefficient of concordance, ranging from 0 to 1, with higher values indicating stronger agreement.

Table 5 presents the outcomes of the first and second rounds of expert consultation. Based on the analysis and synthesis of feedback data, three principal revisions were made to the indicator system following the initial consultation.

Table 5. Feedback Results of the Delphi Album Consultation Method.

Number of Participating Experts: 20						Number of Participating Experts: 20
Number	Element Name	First Round				Number	Element Name	Second Round
		M	SD	CV	Result			M	SD	CV	Result
A1	Spatial Structure Dimension	4.80	0.41	0.086	Retained	A1	Spatial Structure Dimension	3.95	0.22	0.057	Retained
A2	Landscape Elements Dimension	4.95	0.22	0.045	Retained	A2	Landscape Elements Dimension	4.00	0.32	0.081	Retained
A3	Environmental Perception Dimension	4.85	0.37	0.076	Retained	A3	Environmental Perception Dimension	4.05	0.22	0.055	Retained
A4	Sociocultural dimension	4.60	0.60	0.130	Retained	A4	Socio-Cultural Dimension	4.95	0.22	0.045	Retained
A5	Facility System Dimension	4.55	0.60	0.133	Retained	A5	Facility System Dimension	4.20	0.41	0.098	Renamed
B1	Microclimate adaptability	4.65	0.49	0.105	Retained	B1	Spatial Connectivity	4.95	0.22	0.045	Retained
B2	Spatial privacy	4.35	0.88	0.201	Retained	B2	Spatial Hierarchy and Sequence	3.95	0.22	0.057	Retained
B3 *	Spatial recognisability	3.40	0.94	0.277	Integrated	B3	Spatial diversity	4.05	0.22	0.055	Retained
B4	Spatial diversity	4.85	0.49	0.101	Retained	B4	Spatial accessibility	4.05	0.22	0.055	Retained
B5 *	Spatial safety	4.90	0.31	0.063	Integrated	B5	Spatial scale appropriateness	4.05	0.22	0.055	Retained
B6	Spatial accessibility	4.15	0.37	0.088	Retained	B6	Water Landscape Quality	4.10	0.31	0.075	Retained
B7 *	Spatial Integrity and Continuity of Historical Layout	3.25	0.91	0.280	Integrated	B7	Vegetation landscape diversity	4.05	0.22	0.055	Retained
B8	Spatial scale appropriateness	4.80	0.52	0.109	Retained	B8	Hard surfacing and surface materialsy	4.95	0.22	0.045	Renamed
B9 *	River-crossing accessibility	3.35	1.18	0.353	Remove	B9	Landscape Node Design	4.95	0.22	0.045	Retained
B10 *	Spatial accessibility	4.80	0.41	0.086	Integrated	B10 *	Landscape Viewing Areas and Corridors	4.05	0.22	0.055	Remove
B11	Water Body Landscape Quality	4.90	0.31	0.063	Retained	B11	Preservation of historical landscape elements	4.95	0.22	0.045	Retained
B12	Vegetation landscape diversity	4.00	0.32	0.081	Retained	B12	Green Infrastructure Integration	4.95	0.22	0.045	Retained
B13	Hard surfacing and surface materials	4.10	0.31	0.075	Retained	B13	Landscape Colour and Character Coordination	4.95	0.22	0.045	Retained
B14	Landscape Node Design	4.10	0.31	0.075	Retained	B14	Visual comfort	4.05	0.22	0.055	Retained
B15	Landscape Viewing Areas and Corridors	4.20	0.41	0.098	Retained	B15	Thermal environmental comfort	4.90	0.31	0.063	Retained
B16	Preservation of historical landscape elements	4.90	0.31	0.063	Retained	B16	Acoustic Environment Quality	4.95	0.22	0.045	Retained
B17	Green Infrastructure Integration	4.15	0.37	0.088	Retained	B17	Air Quality and Olfactory Perception	4.05	0.22	0.055	Retained
B18	Landscape Colour and Character Coordination	4.90	0.31	0.063	Retained	B18	Ecological Friendliness	4.90	0.45	0.091	Retained
B19 *	Landscape Aesthetic Experience	4.90	0.31	0.063	Adjustment	B19	Space Cleaning and Maintenance Status	4.05	0.22	0.055	Retained
B20	Thermal environmental comfort	4.90	0.31	0.063	Retained	B20	Historical and Cultural Expression	4.95	0.22	0.045	Retained
B21 *	Lighting environment quality	4.10	0.31	0.075	Adjustment	B21	Place Memory and Sense of Place	4.95	0.22	0.045	Retained
B22	Acoustic Environment Quality	4.90	0.31	0.063	Retained	B22	Social inclusion	3.95	0.22	0.057	Retained
B23	Air Quality and Olfactory Perception	4.10	0.45	0.109	Retained	B23	Publicness and Spatial Equity	4.00	0.32	0.081	Retained
B24 *	Climate resilience	4.90	0.31	0.063	Integrated	B24	Cultural Activities and Festivals	4.00	0.32	0.081	Retained
B25	Ecological Friendliness	4.90	0.31	0.063	Retained	B25	Sanitation facilities completeness	3.75	0.55	0.147	Retained
B26	Space Cleaning and Maintenance Status	4.85	0.37	0.076	Retained	B26	Comfort of Resting Facilities	4.95	0.22	0.045	Retained
B27	Historical and Cultural Expression	4.90	0.31	0.063	Retained	B27	Accessibility facilities completeness	3.85	0.67	0.174	Retained
B28	Place Memory and Sense of Place	4.15	0.37	0.088	Retained	B28	Wayfinding and Information System	4.95	0.22	0.045	Retained
B29 *	Local cultural identity	3.15	0.93	0.296	Remove	B29	Diversity of recreational facilities	4.95	0.22	0.045	Retained
B30 *	Immersion in local culture	3.05	0.83	0.271	Integrated	B30	Safety and Emergency Facilities	4.25	0.79	0.185	Retained
B31 *	Community Interaction	4.00	0.32	0.081	Integrated	B31	Completeness of anti-slip facilities	4.30	0.86	0.201	Retained
B32	Social inclusion	4.00	0.32	0.081	Retained	B32	The completeness of lighting facilities	4.35	0.93	0.215	Retained
B33	Publicness and Spatial Equity	4.90	0.31	0.063	Retained	B33	Smart and Digital Management Facilities	4.50	0.83	0.184	Retained
B34	Cultural Activities and Festivals	3.90	0.31	0.079	Retained	B34	Management, Maintenance and Operational Mechanisms	4.95	0.22	0.045	Retained
B35	Sanitation facilities completeness	4.85	0.37	0.076	Retained
B36	Comfort of Resting Facilities	3.95	0.39	0.100	Retained
B37	Accessibility facilities completeness	4.10	0.55	0.135	Retained
B38	Wayfinding and Information System	4.60	0.82	0.178	Retained
B39	Diversity of recreational facilities	4.45	0.89	0.199	Retained
B40	Safety and Emergency Facilities	4.30	0.86	0.201	Retained
B41 *	Shared Sports Facilities and Spaces	4.20	0.41	0.098	Integrated
B42	Completeness of anti-slip facilities	4.10	0.85	0.208	Retained
B43	Smart and Digital Management Facilities	4.20	0.77	0.183	Retained
B44	Management, Maintenance and Operational Mechanisms	4.00	0.73	0.181	Retained

Note: Indicators marked with an asterisk (*) in Table 5 have not passed expert review.

(1)

Five indicators failing to meet consensus standards (M < 3.75, SD > 1, CV > 0.25) were reassessed and revised based on expert recommendations. All five non-consensus indicators were secondary indicators: C3, C7, C9, C29, and C30. Eleven experts deemed that secondary indicator “C9 River-crossing accessibility” within dimension B1 involved the destruction of historic waterfront areas and unpredictable negative impacts from new bridge construction. They proposed improvements through path planning along both riverbanks and utilisation of existing bridges. Consequently, this study removed C9. Seven experts deemed “C29 Local cultural identity” within Dimension B4 to have limited relevance to this study, with its content failing to encompass diverse groups. Consequently, C29 was removed. Furthermore, integrating the opinions of 14 experts, “C3 Spatial Legibility” from Dimension B1 was logically consolidated into “C27 Cultural Representation” and “C38 Wayfinding and Information System”; “C7 Integrity of Historical Layout” was integrated into “C16 Historic Landscape Conservation” and “C18 Colour and Aesthetic Harmony”; “C30 Immersion in Local Culture” from Dimension B4 was integrated into “C28 Sense of Place and Memory”.

(2)

Re-evaluate and revise six indicators meeting consensus criteria (M < 3.75, SD > 1, CV > 0.25) yet receiving multiple expert feedback comments. All six are secondary indicators: C5, C10, C24, C30, C31, and C41. Based on expert input:—“C5 Spatial Safety” from Dimension B1 was logically integrated into “C40 Safety and Emergency System”, while “C10 Spatial Accessibility” was merged into “C6 Accessibility and Connectivity”; “C24 Climate Resilience” from Dimension B3 was integrated into “C20 Thermal Comfort”; Integrating “C30 Tourist Cultural Experience” from Dimension B4 into “C28 Sense of Place and Memory”, and integrating “C31 Community Interaction” into “C4 Functional Variety” and “C39 Functional Facilities Diversity”; “C41 Shared Sports Facilities and Spaces” from Dimension B5 has been integrated into “C39 Functional Facilities Diversity”.

(3)

Individual indicators were revised. Following expert advice, to avoid conceptual ambiguity and differing interpretations, the name of B3 dimension “C19 Aesthetic Experience” was changed to ‘Visual Comfort’; concurrently, “C21 Lighting and Night Ambience” was renamed ‘The Completeness of Lighting Facilities’ and reclassified under B5 dimension.

The revised indicator system underwent a second round of expert consultation. All data points in the first-round modified system met consensus standards (M < 3.75, SD > 1, CV > 0.25), whilst two additional expert comments were obtained. Both experts noted that the concept of “C10 View Corridor and Visual Axis” within Dimension B2 was rather ambiguous. Furthermore, following the first round of adjustments, its meaning was already encompassed within other indicators. They recommended its removal or integration. Acting upon this advice, we undertook a second textual revision, eliminating indicator C10.

Finally, Kendall’s coefficient of concordance was calculated using SPSS 31 software to assess expert consensus on each indicator’s importance. Results showed the W value increased from 0.498 in the first round to 0.657 in the second round, with both rounds yielding statistically significant results (p < 0.001). This indicates a marked improvement in expert consensus [94], rendering further consultation rounds unnecessary.

4.3.3. Final Evaluation System Confirmation

Following two rounds of Delphi consultations, this study refined the evaluation frame work for public spaces within historic urban waterfront areas based on expert assessment scores and recommendations. While retaining the existing five core dimensions, the number of secondary indicators was streamlined from 44 to 33 (Table 6).

Table 6. Final Evaluation System.

System Name	First-Level Dimension	Secondary Indicators
A Perceptual Evaluation System for Public Spaces in Historic Urban Waterfront Areas	B1 Spatial Structure Dimension	C1 Spatial Connectivit
		C2 Spatial Hierarchy and Sequence
		C3 Spatial diversity
		C4 Spatial accessibility
		C5 Spatial scale appropriateness
	B2 Landscape Elements Dimension	C6 Water Landscape Quality
		C7 Vegetation landscape diversity
		C8 Hard surfacing and surface materialsy
		C9 Landscape Node Design
		C10 Preservation of historical landscape elements
		C11 Green Infrastructure Integration
		C12 Landscape Colour and Character Coordination
	B3 Environmental Perception Dimension	C13 Visual comfort
		C14 Thermal environmental comfort
		C15 Acoustic Environment Quality
		C16 Air Quality and Olfactory Perception
		C17 Ecological Friendliness
		C18 Space Cleaning and Maintenance Status
	B4 Socio-Cultural Dimension	C19 Historical and Cultural Expression
		C20 Place Memory and Sense of Place
		C21 Social inclusion
		C22 Publicness and Spatial Equity
		C23 Cultural Activities and Festivals
	B5 Facility System Dimension	C24 Sanitation facilities completeness
		C25 Comfort of Resting Facilities
		C26 Accessibility facilities completeness
		C27 Wayfinding and Information System
		C28 Diversity of recreational facilities
		C29 Safety and Emergency System
		C30 Completeness of anti-slip facilities
		C31 The completeness of lighting facilities
		C32 Smart and Digital Management Facilities
		C33 Management, Maintenance and Operational Mechanisms

4.4. Analytic Hierarchy Process for Weight Determination

The Analytic Hierarchy Process (AHP) is a multi-criteria decision-making method [95], involving the construction of a hierarchical model comprising three levels: the objective level, the criterion level, and the factor level. It assists decision-makers in reducing bias and minimising consensus differences among expert teams [96], and has been extensively applied to evaluate influencing factors of urban public spaces, including urban waterfront areas [97,98,99,100]. This study employs the Analytic Hierarchy Process to construct a perceived evaluation framework for Yangzhou’s Xiaoqinhuai River urban historical waterfront area. Experts were invited to score indicators within the evaluation system, with pairwise comparisons used to assess relative importance. This yielded quantitative indicator weights reflecting the role of different influencing factors in shaping perceptions of the urban historical waterfront’s public space. By integrating subjective judgements with objective analysis through mathematical reasoning, this method significantly enhances the transparency and systematic nature of the decision-making process.

4.4.1. Constructing Hierarchical Models

First, complex decision-making problems are decomposed into multiple hierarchical levels, with interrelated and subordinate relationships established between each tier. The hierarchical model constructed herein, based on the finalised evaluation indicator system for urban historic waterfront public spaces outlined above, comprises three principal levels: the objective level, the criterion level, and the factor level.

4.4.2. Construct a Decision Matrix

This study employs an expert questionnaire method to conduct pairwise comparative assessments of hierarchical elements, utilising a 1–9 scale system to quantify their relative importance. During the assessment process, for any two elements i and j within the same level, intensity values are assigned based on a predefined 9-level proportional scale (Table 7). This enables the quantitative conversion of qualitative judgements through the construction of a pairwise comparison matrix. Here, 1 denotes equivalence between the two elements, 9 indicates absolute dominance, while intermediate values correspond to differing degrees of importance.

Table 7. 9-grade graduated scale.

Paired Comparison Standard	Definition	Content
1	Equally important	Two factors are of equal importance
3	Slightly more important	One factor is slightly more important than the other
5	Considerably more important	Strongly inclined towards one factor
7	Significantly more important	Very strongly inclined towards one factor
9	Absolutely essential	When comparing two factors, one is significantly more important, representing the maximum controllable possibility that one factor can surpass the other
2, 4, 6, 8		A compromise value applied to the above criteria
The reciprocal of the aforementioned value		When comparing factor A to factor B, if either is assigned a value on the above scale, then factor B’s importance relative to A is the reciprocal of that scale value

4.4.3. Merge Expert Matrix

In multi-expert group decision-making processes, divergent expert opinions complicate the construction of comprehensive judgement matrices. This paper introduces a matrix merging approach: experts first independently construct judgement matrix A. Provided individual matrices pass consistency tests, multiple matrices undergo geometric mean integration processing. This preserves expert independence while achieving group decision convergence. The computational process is as follows:

Let ${\mathrm{A}}^{(\mathrm{k})}$ denote the judgement matrix of the kth expert among z experts, where k = 1, 2, …, z:

$${\mathrm{A}}^{(\mathrm{k})}={({\mathrm{a}}_{\mathrm{i}\mathrm{j}})}_{\mathrm{n}\times \mathrm{n}}=\left[\begin{array}{cccc}1& {\mathrm{a}}_{12}^{(\mathrm{k})}& \dots & {\mathrm{a}}_{1\mathrm{n}}^{(\mathrm{k})}\\ {\mathrm{a}}_{21}^{(\mathrm{k})}& 1& \dots & {\mathrm{a}}_{2\mathrm{n}}^{(\mathrm{k})}\\ \dots & \dots & \dots & \dots \\ {\mathrm{a}}_{\mathrm{n}1}^{(\mathrm{k})}& {\mathrm{a}}_{\mathrm{n}2}^{(\mathrm{k})}& \dots & 1\end{array}\right]$$

(1)

By taking the geometric mean of the corresponding elements in the judgement matrices of z experts, the integrated judgement matrix I is obtained, as per the following formula:

$$\mathrm{I}={({\mathrm{i}}_{\mathrm{i}\mathrm{j}})}_{\mathrm{n}\times \mathrm{n}}$$

(2)

In the formula: i_ij denotes the value in row i, column j of the integrated judgement matrix, calculated as follows:

$${\mathrm{i}}_{\mathrm{i}\mathrm{j}}={({\prod }_{\mathrm{k}=1}^{\mathrm{z}}{\mathrm{a}}_{\mathrm{i}\mathrm{j}}^{\mathrm{k}})}^{{\displaystyle \frac{1}{\mathrm{z}}}}$$

(3)

4.4.4. Weighting and Consistency Check

Step 1: Multiply the row vectors of the integration judgement matrix I to obtain a new vector B = bi (i = 1, 2, …, n), calculated as follows: where i denotes the row index and j denotes the column index, with i = 1, 2, …, n and j = 1, 2, …, n:

$$\mathrm{B}=\left[\begin{array}{c}{\mathrm{b}}_1\\ {\mathrm{b}}_2\\ \dots \\ {\mathrm{b}}_{\mathrm{n}}\end{array}\right]={\prod }_{\mathrm{j}=1}^{\mathrm{n}}{\mathrm{i}}_{\mathrm{i}\mathrm{j}}$$

(4)

Step 2: Raise each component b_i of vector B to the nth power to obtain the eigenvectors: $\mathrm{M}={\mathrm{m}}_{\mathrm{i}}(\mathrm{i}=1, 2, \dots , \mathrm{n})$:

$$\mathrm{M}=\left[\begin{array}{c}{\mathrm{m}}_1\\ {\mathrm{m}}_2\\ \dots \\ {\mathrm{m}}_{\mathrm{n}}\end{array}\right]=\sqrt[\mathrm{n}]{{\mathrm{b}}_{\mathrm{i}}}$$

(5)

Step 3: Normalise the feature vector M to obtain the weight vector: $\mathrm{W}={\mathrm{w}}_{\mathrm{i}}(\mathrm{i}=1, 2, \dots , \mathrm{n})$

$$\mathrm{W}={\displaystyle \frac{{\mathrm{m}}_{\mathrm{i}}}{{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{m}}_{\mathrm{i}}}}=\left[\begin{array}{c}{\mathrm{w}}_1\\ {\mathrm{w}}_2\\ \dots \\ {\mathrm{w}}_{\mathrm{n}}\end{array}\right]$$

(6)

Step 4: Conduct a consistency check on the integration judgement matrix I, calculating its consistency index CI and random consistency ratio CR. If CR is less than 0.1, the judgement matrix passes the test and the data is deemed valid (Table 8).

$$\mathrm{CR}={\displaystyle \frac{\mathrm{CI}}{\mathrm{RI}}}$$

(7)

where the consistency index CI and the maximum eigenvalue ${\mathrm{\lambda }}_{\max }$ of the judgement matrix are calculated as follows:

$$\mathrm{CI}={\displaystyle \frac{{\mathrm{\lambda }}_{\max }-\mathrm{n}}{(\mathrm{n}-1)}}$$

(8)

$${\mathrm{\lambda }}_{\max }={\displaystyle \frac{1}{\mathrm{n}}}{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\displaystyle \frac{{(\mathrm{I}\mathrm{w})}_{\mathrm{i}}}{{\mathrm{w}}_{\mathrm{i}}}}$$

(9)

Table 8. Determining the Randomised Consistency Index (RI) Value for Matrix Averages.

Matrix Order	1	2	3	4	5	6	7	8	9	10	11	12
RI	0	0	0.52	0.89	1.12	1.26	1.36	1.41	1.46	1.49	1.52	1.54

4.4.5. Expert Scoring and Data Computation

To scientifically determine indicator weights, this study distributed an evaluation system questionnaire to 35 experts, scholars, practitioners, and operational management personnel involved in urban spatial planning and waterfront enhancement between 10 and 15 September 2025. The invitees comprised 10 academics and scholars from relevant university disciplines, 12 specialists in urban spatial planning, 8 practitioners in urban spatial design, and 5 personnel responsible for the day-to-day management and operation of the Xiao Qinhuai River waterfront area. Drawing upon their professional expertise and practical experience, these invitees rated the importance of the perception evaluation indicators for the public spaces within the Xiao Qinhuai River’s historic urban waterfront area. Subsequently, the questionnaire data was analysed and weighted using the Analytic Hierarchy Process (AHP).

Criteria layers B1, B2, B3, B4, and B5 within the target layer A of the assessment system for the perceived public spaces of the Xiaoqinhuai River’s historic urban waterfront. After verifying each matrix met consistency requirements, the 35 expert judgement matrices were merged using Formulas (1)–(3) to form an integrated matrix. The integrated matrix then underwent weight and consistency verification using Formulas (1)–(9). Similarly, the above procedure was applied to indicators C1 to C33 within each criterion layer.

The calculated results were aggregated to derive relative weights for each indicator. These relative weights were then multiplied sequentially to obtain comprehensive weights, representing the hierarchical ranking of the lowest-level indicators relative to the overall objective (Appendix A Table A3).

4.5. Comprehensive Evaluation of Cloud Modelling

The cloud model is an innovative mathematical framework designed to address issues of uncertainty, proposed by Academician Li Deyi of the Chinese Academy of Engineering in 1995 [101,102]. This theory uniquely integrates the essence of probability theory and fuzzy mathematics, revealing the inherent core uncertainty characteristics of both fuzziness and randomness. It achieves a seamless transition and interconnection between qualitative description and quantitative measurement [103], thereby unifying these two traditionally separate modes of expressing uncertainty. This study employs a cloud model to translate residents’ and visitors’ fuzzy evaluations of the public spaces within the Xiaoqinhuai River’s historic urban waterfront area into quantifiable parameters.

4.5.1. Fundamental Concepts and Numerical Characteristics

Define a numerical range U as a specific quantitative domain. Within this domain, establish a qualitative concept C to describe some abstract property or category of elements within U. For any quantitative value x within domain U, if regarded as a random realisation of qualitative concept C, then x’s membership degree µ(x) is set as a random number between 0 and 1. This random number exhibits a stable tendency, reflecting the extent to which x conforms to the description of concept C. Each such quantitative value x and its corresponding membership degree µ(x) collectively constitute a ‘cloud droplet’.

The three key numerical characteristics of clouds—Expectation (Ex), Entropy (En), and Hyperentropy (He)—possess a unique capacity to quantify qualitative concepts. Figure 4 illustrates a cloud model schematic with an expected value of 10, entropy of 2, and hyperentropy of 0.1. The expected value (Ex) represents the core numerical value of a qualitative concept, serving as its most representative point for quantification. Entropy (En) measures the degree of ambiguity within a qualitative concept, reflecting its vagueness and randomness; a higher entropy value indicates greater conceptual obscurity. Hyperentropy (He) measures the degree of variation in entropy, reflecting the dispersion of cloud droplets. Higher hyperentropy values indicate greater dispersion in droplet distribution, increased randomness in membership degrees, and thicker cloud layers.

Figure 4. Cloud Modelling Diagram (self-drawn by the author).

4.5.2. Algorithm Flow

The cloud model comprises two core algorithms: the forward cloud generator and the inverse cloud generator. The forward cloud generator produces cloud droplets based on known expected values (Ex), entropy (En), and hyperentropy (He), thereby converting qualitative descriptions into quantitative values. Conversely, the inverse cloud generator analyses existing cloud droplet data to infer these three characteristic values, facilitating the extraction of qualitative concepts from quantitative samples.

(1)

Algorithm Flowchart for the Positive Cloud Generator

First, three key parameters must be established: the expected value (Ex), entropy (En), and hyperentropy (He), along with the desired number of cloud droplets to be generated (N). Subsequently, to simulate the random fluctuations of entropy, a normal random number En’ is generated for each cloud droplet, with En as its expected value and He as its standard deviation. This En’ represents the current cloud droplet’s fluctuation in terms of ambiguity and randomness. Subsequently, using Ex as the mean and the En’ generated in the previous step as the standard deviation, another normal random number x is generated. This x represents the specific quantitative value of the current cloud droplet within the quantitative domain, i.e., the ‘cloud droplet’.

Next, to measure the degree to which x belongs to a qualitative concept, a degree of certainty y is calculated. The method for calculating certainty may vary depending on the specific implementation of the cloud model, but it typically relies on the relative position of x with respect to Ex and the magnitude of En’. Generally, the closer x is to Ex and the smaller En’ is, the closer y will be to 1, indicating a higher degree of certainty.

Finally, each pair (x, y) is output as a cloud droplet, where x represents the quantitative value and y denotes the degree of certainty of that value’s affiliation with the qualitative concept. Through this sequence of steps, the forward cloud generator produces a series of cloud droplets. Their distribution across the quantitative domain, coupled with their respective degrees of certainty, collectively form a complete description of the qualitative concept, thereby achieving the transformation from qualitative to quantitative representation.

$$\mathrm{y}={\mathrm{e}}^{{-(\mathrm{x}-\mathrm{E}\mathrm{x})}^2/{2((\mathrm{E}\mathrm{n}))}^2}$$

(10)

(2)

Reverse Cloud Generator Algorithm Flow

First, obtain N cloud droplets xi. Solve for the expected value Ex, entropy En, and hyperentropy He corresponding to the qualitative concepts of these N droplets.

Calculate the sample mean:

$$\overline{\mathrm{x}}={\displaystyle \frac{1}{\mathrm{n}}}{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{x}}_{\mathrm{i}}$$

(11)

Calculate the sample variance:

$${\mathrm{S}}^2={\displaystyle \frac{1}{\mathrm{n}-1}}{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{({\mathrm{x}}_{\mathrm{i}}-\overline{\mathrm{x}})}^2$$

(12)

Calculate the expected value:

$$\mathrm{E}\mathrm{x}=\overline{\mathrm{x}}$$

(13)

Calculating entropy:

$$\mathrm{E}\mathrm{n}=\sqrt{{\displaystyle \frac{\mathrm{\pi }}{2}}}\times {\displaystyle \frac{1}{\mathrm{n}}}{\sum }_{\mathrm{i}=1}^{\mathrm{n}}|{\mathrm{x}}_{\mathrm{i}}-\overline{\mathrm{x}}|$$

(14)

Calculating superentropy:

$$\mathrm{H}\mathrm{e}=\sqrt{|{\mathrm{E}\mathrm{n}}^2-{\mathrm{S}}^2|}$$

(15)

(3)

Cloud Similarity Calculation

Having established the comprehensive evaluation cloud model and standard cloud model for each indicator, the cloud similarity γ is employed as a quantitative standard to measure the degree of proximity between a project’s risk status and different evaluation grades. Specifically, a higher cloud similarity value indicates a greater alignment between the current project’s risk status and a particular evaluation grade. The assessment grade is determined by the level corresponding to the evaluation interval with the highest cloud similarity. To achieve this objective, the following steps must be followed:

In evaluating Cloud C, generate a random number ${\mathrm{E}}_{\mathrm{j}}$ that conforms to the normal distribution specified by En and He.

$${\mathrm{E}}_{\mathrm{j}}=\mathrm{N}({\mathrm{E}}_{\mathrm{n}},{\mathrm{H}}_{\mathrm{e}}^2)$$

(16)

Generate a normal random number ${\mathrm{X}}_{\mathrm{j}}$ in the evaluation cloud C that satisfies the conditions Ex and ${{\mathrm{E}}_{\mathrm{j}}}^2$.

$${\mathrm{X}}_{\mathrm{j}}=\mathrm{N}({\mathrm{E}}_{\mathrm{x}},{{\mathrm{E}}_{\mathrm{j}}}^2)$$

(17)

Generate a random number ${{\mathrm{E}}_{\mathrm{j}}}^{\prime }$ from a normal distribution with mean Ev and variance ${\mathrm{H}}_{\mathrm{e}\mathrm{v}}^2$ in Standard Cloud C.

$${{\mathrm{E}}_{\mathrm{j}}}^{\prime }=\mathrm{N}({\mathrm{E}}_{\mathrm{v}},{\mathrm{H}}_{\mathrm{e}\mathrm{v}}^2)$$

(18)

Calculate the membership degree ${\mathrm{\mu }}_{\mathrm{j}\mathrm{v}}$ in the standard cloud C for ${\mathrm{X}}_{\mathrm{j}}$.

$${\mathrm{\mu }}_{\mathrm{j}\mathrm{v}}={\mathrm{e}}^{{-({\mathrm{x}}_{\mathrm{j}}-\mathrm{E}\mathrm{v})}^2/{2({{\mathrm{E}}_{\mathrm{j}}}^{\prime })}^2}$$

(19)

Repeat steps 10–14 until n ${\mathrm{\mu }}_{\mathrm{j}\mathrm{v}}$ values have been generated, then compute the similarity.

$${\mathrm{\gamma }}_{\mathrm{v}}={\displaystyle \frac{1}{\mathrm{n}}}{\sum }_{\mathrm{j}=1}^{\mathrm{n}}{\mathrm{\mu }}_{\mathrm{j}\mathrm{v}}$$

(20)

Normalise γ to obtain the membership degree ${\mathrm{\theta }}_{\mathrm{v}}$ indicating that the current evaluation cloud belongs to the vth evaluation level.

$${\mathrm{\theta }}_{\mathrm{v}}={\displaystyle \frac{{\mathrm{\gamma }}_{\mathrm{v}}}{{\sum }_{\mathrm{v}=1}^{\mathrm{n}}{\mathrm{\gamma }}_{\mathrm{v}}}}$$

(21)

(4)

Generate a comprehensive evaluation cloud map

Employ a forward cloud generator to create a comprehensive evaluation cloud map within the domain space.

4.5.3. Empirical Analysis and Computation

(1)

Establishment of Evaluation Criteria and Standard Cloud Grading

Drawing upon the specific context of the historic waterfront public spaces along the Xiaoqinhuai River and expert consultation, linguistic terminology—‘extremely dissatisfied’, ‘dissatisfied’, ‘neutral’, ‘satisfied’, and ‘extremely satisfied’—is employed to describe public space perception levels. Corresponding score ranges are set between [0, 100], with higher scores indicating better perceived quality. [0.0, 25.0] denotes ‘Very Dissatisfied’, [25.0, 50.0] denotes ‘Dissatisfied’, [50.0, 75.0] corresponds to ‘Neutral’, [75.0, 90.0] denotes ‘Satisfied’, and [90.0, 100] denotes ‘Very Satisfied’ (Table 9). Based on these rating intervals, a set of reasonable evaluation terms was established. Using the evaluation items, cloud model parameters were derived Via a forward cloud generator to generate the standard evaluation cloud model. The corresponding cloud model parameters are shown in Table 10.

Table 9. Correspondence Table of Comments and Scores.

Rating Level	Grade Range
Very dissatisfied	[0.0, 25.0]
Dissatisfied	(25.0, 50.0]
Neutral	(50.0, 75.0]
Satisfied	(75.0, 90.0]
Very satisfied	(90.0, 100.0]

Table 10. Score Range and Corresponding Values.

Rating Level	Grade Range	Ex	En	He
Very dissatisfied	[0.0, 25.0]	12.5	4.167	0.5
Dissatisfied	(25.0, 50.0]	37.5	4.167	0.5
Neutral	(50.0, 75.0]	62.5	4.167	0.5
Satisfied	(75.0, 90.0]	82.5	2.5	0.5
Very satisfied	(90.0, 100.0]	95	1.667	0.5

Based on the intervals in the table above, cloud model calculations were performed on the five comments to derive the standard cloud. The calculation formula is as follows:

$$\left\{\begin{array}{c}{\mathrm{E}\mathrm{x}}_{\mathrm{v}}=\left({\mathrm{x}}_{\mathrm{v}}^{\mathrm{m}\mathrm{a}\mathrm{x}}+{\mathrm{x}}_{\mathrm{v}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\right)/2\\ {\mathrm{E}\mathrm{n}}_{\mathrm{v}}=\left({\mathrm{x}}_{\mathrm{v}}^{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{x}}_{\mathrm{v}}^{\mathrm{m}\mathrm{i}\mathrm{n}}\right)/6\\ {\mathrm{H}\mathrm{e}}_{\mathrm{v}}=\mathrm{k}\end{array}\right.$$

(22)

where the vth sub-interval is [${\mathrm{x}}_{\mathrm{v}}^{\mathrm{m}\mathrm{i}\mathrm{n}}$, ${\mathrm{x}}_{\mathrm{v}}^{\mathrm{m}\mathrm{a}\mathrm{x}}$], and the numerical characteristics of the standard cloud Couldv corresponding to this sub-interval are (Exv, Env, Hev); k is a constant, taken as 0.5 in this paper.

The calculation results are shown in Table 10:

The cloud map of the standard cloud, calculated from the computed values of Ex, En, and He, is shown in Figure 5:

Figure 5. Standard Cloud Diagram (self-drawn by the author).

(2)

Computational Analysis of the Integrated Cloud

This research team conducted a study from 21 to 27 September 2025, deploying two two-person teams (Team A and Team B) along both banks of the Xiaoqinhuai Riverside Area. They distributed questionnaires assessing perceptions of public spaces to explore residents’ and visitors’ perceptions of public spaces within the city’s historic waterfront district. The questionnaire employed an on-site scoring method with a maximum score of 100 points, where higher scores indicated stronger perceptions of the evaluation indicators. To ensure the accuracy of feedback, the questionnaire was distributed and collected on-site: Team A targeted tourists, distributing and collecting questionnaires in public spaces at key tourist attractions along the Xiaoqinhuai Riverside; Group B conducted door-to-door surveys and collections among indigenous residents living along both banks of the Xiaoqinhuai River, assisted by the local residents’ committee. This approach mitigated potential resistance or wariness among residents, ensuring the accuracy and validity of responses. Throughout the distribution and collection process, Teams A and B maintained constant communication to ensure the number of valid responses and demographic proportions remained broadly consistent across their respective respondent groups. All surveys were conducted with informed consent from participants. A total of 240 questionnaires were distributed, yielding 223 valid responses, representing a 93% valid response rate. Among valid samples, 105 (47%) were residents and 118 (53%) were tourists. Gender distribution showed 109 males (48.87%) and 114 females (51.12%). Age distribution comprised: 25 respondents aged 18–25 (11.21%); 46 aged 26–30 (20.62%); 51 aged 31–40 (22.86%); 41 aged 41–50 (18.38%); 37 aged 51–60 (16.59%); and 23 aged over 60 (10.31%) (Appendix A Table A4). The valid sample data exhibits high representativeness, comprehensiveness, and reliability, fully reflecting the perceptions of public space in the Xiaoqinhuai Riverside Area among two core user groups. Based on questionnaire scores, an evaluation matrix was constructed. A backward cloud generator was employed to compute numerical characteristics for target-layer, standard-layer, and factor-layer cloud models. Finally, a forward cloud generator produced a normalised cloud map, providing scientific basis for subsequent research. The detailed process is as follows:

Employing the expert scoring method to evaluate each factor layer yields the indicator evaluation matrix Z. With m experts and n evaluation indicators, zij denotes the evaluation result of the jth indicator by the ith expert. i = 1, 2, …, m; j = 1, 2, …, n. Processing the indicator evaluation matrix Z yields the indicator evaluation cloud Via the following formula: The evaluation cloud for the jth indicator is denoted as Cj(Exj, Enj, Hej), where j = 1, 2, …, n (Appendix A Table A3).

$$\left\{\begin{array}{c}{\mathrm{E}\mathrm{x}}_{\mathrm{j}}={\displaystyle \frac{1}{\mathrm{m}}}{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{Z}}_{\mathrm{i}\mathrm{j}}\\ {\mathrm{E}\mathrm{n}}_{\mathrm{j}}=\sqrt{{\displaystyle \frac{\mathrm{\pi }}{2}}}\times {\displaystyle \frac{1}{\mathrm{m}}}{\sum }_{\mathrm{i}=1}^{\mathrm{m}}\left|{\mathrm{Z}}_{\mathrm{i}\mathrm{j}-}{\mathrm{E}\mathrm{x}}_{\mathrm{j}}\right|\\ {\mathrm{H}\mathrm{e}}_{\mathrm{j}}=\sqrt{\left|{\mathrm{E}\mathrm{n}}_{\mathrm{j}}^2-{\mathrm{S}}_{\mathrm{j}}^2\right|}\end{array}\right.$$

(23)

After calculating the numerical eigenvectors of the secondary indicators, the comprehensive weighting values of the indicators are combined with the corresponding cloud feature parameters using the formula to derive the comprehensive evaluation cloud for the entire subject. This is then compared with the standard evaluation cloud to yield the final assessment outcome (Table 11).

$$\begin{array}{c}{\mathrm{E}}_{\mathrm{x}}={\displaystyle \frac{{\sum }_{\mathrm{j}=1}^{\mathrm{n}}{\mathrm{E}\mathrm{x}}_{\mathrm{j}}\times {\mathrm{E}\mathrm{n}}_{\mathrm{j}}\times {\mathrm{W}}_{\mathrm{j}}}{{\sum }_{\mathrm{j}=1}^{\mathrm{n}}{\mathrm{E}\mathrm{n}}_{\mathrm{j}}\times {\mathrm{W}}_{\mathrm{j}}}}\hfill \\ {\mathrm{E}}_{\mathrm{n}}={\sum }_{\mathrm{j}=1}^{\mathrm{n}}{\mathrm{E}\mathrm{n}}_{\mathrm{j}}\times {\mathrm{W}}_{\mathrm{j}}\hfill \\ {\mathrm{H}}_{\mathrm{e}}={\displaystyle \frac{{\sum }_{\mathrm{j}=1}^{\mathrm{n}}{\mathrm{H}\mathrm{e}}_{\mathrm{j}}\times {\mathrm{E}\mathrm{n}}_{\mathrm{j}}\times {\mathrm{W}}_{\mathrm{j}}}{{\sum }_{\mathrm{j}=1}^{\mathrm{n}}{\mathrm{E}\mathrm{n}}_{\mathrm{j}}\times {\mathrm{W}}_{\mathrm{j}}}}\hfill \end{array}$$

(24)

Table 11. Eigenvalue calculation results.

Integrated Cloud Value	Ex	En	He
Eigenvalue	59.537	6.593	2.656

Substituting the digital characteristics of the composite cloud into Formula (20) yields similarity results for the five evaluation criteria. These similarity results are then processed according to Formula (21) to calculate the composite cloud’s membership degrees for each criterion. Applying the principle of maximum membership degree, the composite cloud’s final evaluation is determined to be ‘average’ (Table 12).

Table 12. Membership degree results.

Integrated Cloud	Very Dissatisfied	Dissatisfied	Neutral	Satisfied	Very Satisfied
Similarity	0.0000	0.0159	0.5159	0.0041	0.0000
Membership degree	0.0000	0.0297	0.9626	0.0077	0.0000

Based on the integrated cloud computing results, Ex, En, and He were input into the MATLAB 2025 programme to generate cloud maps. The standard cloud and integrated cloud were merged to produce the cloud map results as shown in Figure 6.

Figure 6. Comparison of Integrated Cloud and Standard Cloud (self-drawn by the author).

Evaluation characteristics may be quantitatively represented using the cloud’s numerical attributes (Ex, En, He). Here, the expected value Ex serves as the quantitative centre point for qualitative concepts, corresponding to the centre of mass of cloud droplets within the domain. Higher values indicate a more favourable overall assessment of public space perception in the Xiaoqinhuai Riverside Area. Entropy En measures the fuzziness and probability of qualitative concepts, reflecting the ambiguity and uncertainty in cloud droplet distribution. A higher value indicates a wider distribution span of cloud droplets on the cloud map, signifying greater instability in the overall assessment. Hyperentropy He serves as a measure of uncertainty, reflecting the cohesion among cloud droplets. A higher value indicates reduced cohesion among droplets, signifying greater uncertainty in evaluators’ perceptions of the riverside public space and lower levels of consensus. The generated cloud map and membership values reveal a clearly defined distribution area for cloud droplets, with an expected value of 59.537 within the 50–75 range. This indicates that the overall perceived evaluation of public spaces in the Xiaoqinhuai Riverside Area falls between ‘dissatisfied’ and ‘average’, closely aligned with ‘average’, with a membership degree of 0.9626.

(3)

Calculation of Individual Indicators

Normalisation was performed based on the composite weights of the secondary indicators. Repeating the comprehensive cloud calculation process described above, the composite clouds for the primary dimensions and secondary indicators were computed (Appendix A Table A5). Similarly, the membership degree results for each indicator were calculated (Appendix A Table A6), followed by integration and summarisation. Finally, a normalised cloud plot was generated using a forward cloud generator for subsequent analysis. Due to space constraints, the computational workflow for individual indicators is not reproduced here; the generated cloud plot will be presented in the findings section below.

5. Results

5.1. Overall Assessment Results

Based on the cloud model calculations and generated cloud map (Figure 7), the comprehensive perception assessment of the public spaces within the Xiaoqinhuai River’s historic urban waterfront area is rated as ‘average’, with an expected value of 59.937—nearly overlapping with the ‘average’ benchmark of 60. The super-entropy value stands at 2.656, indicating a degree of instability. This suggests a discernible cognitive disparity in how visitors and residents perceive the public spaces within this waterfront area. Within the assessment framework, the perceived evaluation sequence for primary dimensions is: ‘B4 Sociocultural Dimension’, ‘B1 Spatial Structure Dimension’, ‘B5 Facility System Dimension’, ‘B3 Environmental Perception Dimension’, and ‘B2 Landscape Element Dimension’, with respective expected values of 78.165, 76.541, 55.289, 49.495, and 49.006. The expected values for both the ‘Sociocultural Dimension’ and the ‘Spatial Structure Dimension’ exceed the comprehensive assessment expectation. The generated cloud maps substantially overlap with the ‘Satisfactory’ cloud, exhibiting hyperentropy values of 2.074 and 1.515, respectively—both lower than the hyperentropy value for the comprehensive perception assessment. The cognitive disparity between residents and visitors was smaller than the existing gap in the comprehensive perception assessment. Consequently, both the ‘B4 Sociocultural Dimension’ and ‘B1 Spatial Structure Dimension’ were classified as ‘satisfactory’ attributes. The expected value for the “B5 Facility System Dimension” was 55.289, closely matching the overall perceived assessment expectation. However, its hyperentropy value of 3.023 was the highest among the five primary dimensions, revealing significant cognitive divergence between visitors and residents when evaluating the public facility system of the waterfront area. This dimension was thus classified as “Average”. The expected values for the ‘B3 Environmental Perception Dimension’ and ‘B2 Landscape Elements Dimension’ are close, yet both fall short of the comprehensive perception assessment score at 49.495 and 49.006, respectively, with hyperentropy values of 2.261 and 1.690, respectively, both lower than the hyperentropy threshold for the comprehensive perception assessment. This indicates that both visitors and residents exhibit dissatisfaction with the environmental perception and landscape elements of the waterfront public space, with relatively minor cognitive differences. Urgent improvement and optimisation are required. Therefore, based on membership functions, both the ‘B3 Environmental Perception Dimension’ and ‘B2 Landscape Elements Dimension’ are classified as ‘unsatisfactory’.

Figure 7. Overall Assessment Cloud Map (self-drawn by the author).

5.2. Perceptual Evaluation Results for ‘B1 Spatial Structure’

In the comprehensive perception assessment of public spaces within the historical riverside district of the Xiaoqinhuai River, the primary dimension ‘B1 Spatial Structure’ exhibits the second-highest expected value of 76.541 alongside the lowest super-entropy value of 1.515. This indicates high overall satisfaction and strong stability. Although the generated cloud map displays some degree of dispersion, the overall perception assessment yields favourable results, categorised as ‘satisfactory’ (Figure 8). Within this dimension, the expected values for ‘C1 Spatial Connectivity’, ‘C2 Spatial Hierarchy and Sequence’, and ‘C5 Spatial Scale Appropriateness’ were closely aligned at 82.848, 83.009, and 82.883, respectively. The cloud maps for these three secondary indicators substantially overlap with the ‘satisfactory’ cloud in the standard cloud map, and their hyperentropy values are all relatively low, demonstrating a high degree of consistent satisfaction. Combined with their respective membership degrees, they are all judged to be of ‘satisfactory’ quality. Overall, The Xiao Qinhuai River’s well-developed continuous riverside walkways, pedestrian accessibility, balanced ratio of open and semi-enclosed spaces, richly layered pedestrian environments, and appropriate spatial scale for human activities constitute its existing strengths (Figure 9). These advantages stem from the historically accumulated grid of roads and dense water network characteristic of urban historic waterfronts, representing crucial spatial structural resources for the area. Within this historically developed urban waterfront area, where vehicular and non-motorised traffic is restricted, both visitors and residents express high satisfaction with the walkability, accessibility, and the rich yet balanced spatial composition and stratification. The favourable assessment results for the three secondary indicators C1, C2, and C5 largely stem from current governmental and societal efforts to preserve the historical and cultural heritage district. Consequently, the urban historical waterfront area retains its original spatial structure to the greatest extent possible, preserving its core competitive advantage in the perceived quality of its public spaces. The expected values for ‘C3 Spatial diversity’ and ‘C4 Spatial accessibility’ are lower than the preceding three indicators, at 63.233 and 63.054, respectively. These overlap to some extent with the “average” category, exhibiting slightly elevated super-entropy values and relatively dispersed data points. Nevertheless, they remain above the overall comprehensive assessment threshold. Combined with their respective membership degrees, both are classified as exhibiting ‘average’ attributes. This indicates that while possessing three advantageous high-satisfaction indicators (C1, C2, and C5), there remains scope for improvement in diversity, the planning of multi-scale landscape spaces, the density of waterfront and external transport routes, and connectivity with public transport systems. This would provide users of the waterfront public space with a greater variety of landscape space types and more convenient and efficient transport accessibility.

Figure 8. B1 Dimension Perception Assessment Results (self-drawn by the author).

Figure 9. Xiao Qinhuai Waterfront Promenade (photograph by the author).

5.3. Perceptual Assessment Results for ‘B2 Landscape Elements’

In the comprehensive perception assessment of public spaces within the Xiaoqinhuai River Urban Historical Waterfront Area, the primary dimension ‘B2 Landscape Elements’ exhibited the lowest expected value among the five primary dimensions at 49.006, with a relatively low super-entropy value of 1.690. The cloud drop distribution for secondary indicators C6–C12 within this dimension showed considerable dispersion except for C10, indicating overall low satisfaction among both visitors and residents. This dimension was categorised as ‘dissatisfactory’ (Figure 10). Within the B2 dimension, only one secondary indicator—‘C10 Preservation of historical landscape elements’—met the ‘satisfactory’ standard, with an expected value of 82.924; ‘C6 Water Landscape Quality’, ‘C11 Green Infrastructure Integration’, and ‘C12 Landscape Colour and Character Coordination’ had expected values of 62.767, 63.475, and 62.951, respectively. Combined with the generated cloud map and their respective membership degrees, all were classified as ‘average’. The expected values for ‘C7 Vegetation landscape diversity’, ‘C8 Hard surfacing and surface materials’, and ‘C9 Landscape Node Design’ were comparatively low at 37.915, 37.834, and 38.314, respectively. Based on the generated cloud maps and their respective membership degrees, these were classified as ‘Unsatisfactory’. Overall, the historic waterfront area along the Xiaoqinhuai River has preserved its heritage landscape elements—such as old wharves, revetments, and bridges—relatively well (Figure 11). However, it has yet to achieve the revitalisation and utilisation of existing features to enhance the overall perceived landscape quality. Firstly, key landscape perception elements such as water body cleanliness and overall accessibility scored only at an average level. Secondly, high-quality waterfront design features widely employed in other urban waterfronts—such as rain gardens, ecological revetments, and vegetated buffer zones—have yet to be integrated into the Xiaoqinhuai River historical waterfront. Consequently, the overall landscape elements remain relatively monotonous, lacking systematic and targeted planning and design. Finally, deficiencies exist in the selection and design of landmark paving materials, the selection and planting of complementary vegetation, and the aesthetic and functional design of waterfront platforms, viewing decks, and plaza spaces within the waterfront area, with assessment scores for these aspects deemed unsatisfactory. The current ground surfaces within the Xiaoqinhuai River waterfront public spaces exhibit poor quality, inconsistent material selection, and non-uniform paving methods. During the survey, multiple instances of cracked and damaged surfaces and steps were observed (Figure 12), alongside irregular undulations in ground surfaces at numerous traffic junctions (Figure 13). Significant improvement potential exists in both landscape aesthetics and alignment with the historical context. Concurrently, the selection of plants and landscaping methods along both banks of the waterfront area is monotonous. Extensive greenery exhibits fragmentation, damage, and contamination (Figure 14), necessitating urgent re-selection, replanting, and maintenance of landscape vegetation. Currently, apart from the redeveloped cultural tourism zone at the Xiaodongmen section, the Xiaoqinhuai Riverside area suffers from a scarcity of waterfront platforms, viewing decks, and plaza spaces. These existing facilities are poorly maintained, failing to provide visitors and residents with convenient access or an enhanced visual experience. During our research, we also observed that some riverside buildings have undergone renovation and repurposing (including residential use). However, the colours of these renovated building facades do not align with the overall aesthetic and colour palette of the Xiaoqinhuai Riverside Area. This mismatch is one reason why visitors and residents rate the coordination of colours between the public space architecture and landscape in the area as rather mediocre.

Figure 10. B2 Dimension Perception Assessment Results (self-drawn by the author).

Figure 11. The ancient bridge in good condition (photograph by the author).

Figure 12. Damaged stair paving (photograph by the author).

Figure 13. Stairs with an excessively steep angle (photograph by the author).

Figure 14. Damaged coastal greenery (photograph by the author).

5.4. Perception Assessment Results for ‘B3 Environmental Perception Dimension’

The expected value for the primary dimension ‘B3 Environmental Perception Dimension’ in the comprehensive perception assessment of public spaces within the Xiaoqinhuai River Urban Historical Waterfront Area is relatively low at 49.006. Concurrently, the hyperentropy value is the highest at 2.261, with a significant dispersion of cloud droplets. This indicates that visitors and residents exhibit low satisfaction with the environmental perception of public spaces along the Xiaoqinhuai River waterfront, alongside discernible cognitive differences (Figure 15). Within the B3 dimension, only one secondary indicator—‘C15 Acoustic Environment Quality’—met the ‘Satisfactory’ standard, exhibiting a relatively high expected value of 83.368. with a compact cloud-drop distribution overlapping substantially with the ‘Satisfactory’ cloud. Its membership degree confirms this as a ‘Satisfactory’ attribute. Situated within a historical and cultural conservation zone, the Xiaoqinhuai Riverside Area benefits from distance from high-traffic routes and concentrated commercial activity, enabling effective noise control. Natural flowing water bodies and riparian trees further cultivate a favourable natural soundscape and tranquil atmosphere (Figure 16). The expected values for ‘C13 Visual comfort’, ‘C17 Ecological Friendliness’ and ‘C18 Space Cleaning and Maintenance Status’ are 62.924, 63.749, and 63.502, respectively, with excess entropy values of 2.713, 2.959, and 2.377. The expected values and excess entropy values for these three secondary indicators are relatively close, yielding largely consistent assessment results. Combined with their respective membership degrees, all three are classified as ‘average’ attributes. This outcome indicates that the public spaces along the Xiaoqinhuai Riverside area currently fail to provide users with a high-quality visual and ecological sensory experience in terms of spatial aesthetics, spatial imagery creation, ecological enhancement of pedestrian corridors, and biodiversity presence. Key contributing factors include the uniformity of the riverside public space’s morphological layers, the poor maintenance of greenery (Figure 17), and the resulting scarcity of biological populations in the riverside zone. Concurrently, spatial tidiness warrants attention in the Xiaoqinhuai Riverside public spaces. Field investigations revealed extensive disorderly accumulation of debris and haphazard parking of residents’ non-motorised vehicles, constituting an ‘encroachment’ upon the public space (Figure 18). Overall spatial tidiness was low (Figure 19), with no evidence of dedicated personnel conducting regular maintenance during the survey period. ‘C14 Thermal environmental comfort’ and ‘C16 Air Quality and Olfactory Perception’ were the two indicators with the lowest expected values in this study’s assessment framework, both around 16, with relatively low super-entropy values. Combined with their respective membership degrees, both were judged to be ‘unsatisfactory’ attributes. This phenomenon indicates that both visitors and residents consistently express significant dissatisfaction with the climate suitability, walking comfort during extreme temperatures, air cleanliness, and odour control of the Xiaoqinhuai River public space. Based on survey findings and expert feedback, the Xiaoqinhuai River lacks adequate shade and heat-relief facilities beyond natural vegetation cover during extreme temperatures, particularly summer heatwaves. This significantly compromises walking comfort and daily living. Similarly, cold temperatures from stagnant water during winter’s rainy, overcast weather prove uncomfortable, especially in the river’s historic urban waterfront areas where elderly residents congregate. Furthermore, despite dredging works undertaken during the river’s regeneration, persistent water odour remains a significant issue. During the study period, when temperatures reached approximately 28 °C, the water along the Xiaoqinhuai River’s waterfront emitted a noticeable odour. Numerous interviewees and survey respondents, including both visitors and residents, expressed strong negative reactions to this. Sustained and targeted measures to control river water odour warrant significant attention in future regeneration efforts.

Figure 15. B3 Dimension Perception Assessment Results (self-drawn by the author).

Figure 16. A more favourable soundscape (photograph by the author).

Figure 17. Encroaching foliage cluttering the pavement (photograph by the author).

Figure 18. The encroachment of clutter upon communal spaces (photograph by the author).

Figure 19. The current state of pedestrian footways (photograph by the author).

5.5. Perceived Assessment Results for the ‘B4 Socio-Cultural Dimension’

The expected value for the primary dimension ‘B4 Socio-Cultural Dimension’ in the comprehensive perception assessment of public spaces within the Xiaoqinhuai River Urban Historical Waterfront Area reached its highest value of 78.165. The cloud drop distribution was relatively concentrated with a narrow span, accompanied by a low super-entropy value. This indicates a high level of satisfaction coupled with excellent stability (Figure 20). Within the B4 dimension, all secondary indicators’ expected values exceeded 60, ranked from highest to lowest as follows: ‘C19 Historical and Cultural Expression (95.516)’, ‘C23 Cultural Activities and Festivals (83.767)’, ‘C20 Place Memory and Sense of Place (83.139)’, ‘C22 Publicness and Spatial Equity (82.794)’, and ‘C21 Social Inclusion (62.148)’. Except for indicator C21, all indicators exhibit superentropy values below 2, demonstrating favourable cloud droplet cohesion and minimal lateral spread. Combined with their respective membership degrees, all are classified as “Satisfactory” or higher attributes, with C19 achieving ‘Highly Satisfied’ status. This phenomenon indicates that the preservation and authenticity maintenance of the Xiaoqinhuai Historical Riverside Area substantially enhance residents’ and visitors’ satisfaction with socio-cultural perceptions (Figure 21). Under the emphasis and protection of government departments, social organisations, and planning/design units towards various historical and intangible heritage—such as cultural symbol displays, historical narrative storytelling, and the continuation of local traditions—a distinctive sense of uniqueness and emotional belonging has been forged for this historical site. Concurrently, during the implementation of substantial renewal projects at key nodes within the historic waterfront area, local culture and distinctiveness have been integrated into the festive events, exhibitions, and cultural activities of the renewed zones, with commercial development being pursued in a relatively restrained manner (Figure 22). This has effectively enhanced the cultural richness and diversity of the public spaces along the Xiaoqinhuai waterfront. These measures collectively constitute significant contributing factors to the positive evaluation outcomes for Dimension B4 and its secondary indicators. C21 exhibits lower expected values and higher super-entropy compared to the other four indicators. Its cloud drop cohesion and lateral span in the generated cloud map are suboptimal, resulting in a ‘fair’ rating. Based on research findings, the current renewal process of the Xiaoqinhuai Riverside public space has not prioritised accessibility for vulnerable groups, including children, the elderly, and persons with disabilities. Future public space renewal initiatives should incorporate child-friendly, age-appropriate, and disability-accessible design elements and supporting facilities to enhance social inclusivity.

Figure 20. B4 Dimension Perception Assessment Results (self-drawn by the author).

Figure 21. Bridge Protection Signage (photograph by the author).

Figure 22. Cultural and Creative Cluster (photograph by the author).

5.6. Perceptual Evaluation Results for the ‘B5 Facility System Dimension’

The assessment results for the primary dimension ‘B5 Facility System Dimension’ in the comprehensive perception evaluation of public spaces within the Xiaoqinhuai River Urban Historical Waterfront Area were moderate, with an expected value of 55.289, slightly below the overall perception assessment score. Although the cloud-drop distribution was relatively concentrated, the extent of the cloud map spanned considerably. The super-entropy value reached 3.023, the highest among the five primary dimensions, with all secondary indicators within this dimension exhibiting substantial cloud droplet spans and super-entropy values exceeding 2.5. This indicates a degree of individual cognitive variation in visitors’ and residents’ perceptions of the facility system within the Xiaoqinhuai River waterfront public space (Figure 23). Analysis of cloud droplet continuity, concentration, and respective membership results reveals three indicators within Dimension B5 receiving ‘dissatisfied’ evaluations: ‘C25 Comfort of Resting Facilities’, ‘C28 Diversity of Recreational Facilities’, and ‘C30 Completeness of Anti-Slip Facilities’, each with an expected value around 38. The remaining seven secondary indicators—‘C24 Sanitation facilities completeness,’ ‘C26 Accessibility facilities completeness,’ ‘C27 Wayfinding and Information System,’ ‘C29 Safety and Emergency System,’ ‘C31 The completeness of lighting facilities,’ ‘C32 Smart and Digital Management Facilities,’ and ‘C33 Management. Maintenance and Operational Mechanisms’ are all classified as ‘average’. Among these, C24, C26, C29, C32, and C33 exhibit significant cognitive divergence, with entropy values exceeding 3. Overall, facility systems constitute a crucial component of public spaces. Many supporting facilities in historic waterfront public areas have deteriorated or been damaged over time due to historical circumstances. The perceived quality of these facilities by visitors and residents directly contributes to the generally average assessment scores. Visitor groups prioritise the experiential aspects of navigating waterfront public spaces, thus focusing on amenities that directly enhance tourism experiences: recreational facilities, accessibility features, wayfinding and information systems, alongside smart and digital management infrastructure. Residents who regard waterfront public spaces as living environments place greater emphasis on sanitation facilities, activity provisions, safety and emergency measures, anti-slip surfaces, lighting installations, and management/maintenance/operational mechanisms—all closely tied to the quality of their living environment. Overall, differing purposes and needs shape distinct perspectives on facility system evaluations, influencing cognitive variations within this dimension and its sub-indicators. Research findings indicate that the Xiaoqinhuai Riverside currently features only one small pocket park with associated activity facilities, which are poorly maintained and largely in a state of semi-dereliction (Figure 24). The density of public seating and backrest facilities is low, with rudimentary conditions; some have suffered damage or contamination (Figure 25). The primary pedestrian walkways and staircases along the riverbank present significant safety risks due to inadequate slip resistance during wet weather or high humidity. These conditions directly account for the unsatisfactory perceived assessment results for sub-indicators C25, C28, and C30. Secondly, according to survey interviews, although the completeness of sanitation facilities, safety and emergency facilities, and lighting facilities was rated as ‘average’ in the perceived assessment scores, visitors and residents expressed stronger negative reactions towards these three categories. There was unanimous agreement that the public spaces within the Xiaoqinhuai River waterfront area should feature a greater number and higher density of litter bins (Figure 26), lifebuoys (Figure 27), and street lamps. Finally, whilst ensuring the optimisation and enhancement of the aforementioned facilities, a more efficient and professional maintenance mechanism should be implemented. This will achieve the dual objectives of improving the convenience of facility usage and elevating the overall perceived assessment level.

Figure 23. B5 Dimension Perception Assessment Results (self-drawn by the author).

Figure 24. The sole, semi-derelict pocket park (photograph by the author).

Figure 25. Rest facilities in poor condition (photograph by the author).

Figure 26. Refuse collection area (photograph by the author).

Figure 27. Emergency life-saving equipment (photograph by the author).

6. Discussion

6.1. Comparison with Previous Research Findings

The findings of this study bear a certain degree of relevance to domestic and international research on public spaces within waterfront areas and historic districts, while simultaneously achieving substantive expansion in terms of research perspectives and methodological integration. Existing research consistently indicates that the environmental quality [39,49,57,60] and cultural characteristics [1,27,33,67] of waterfront spaces are pivotal factors influencing public perception, with their spatial appeal stemming from physical attributes such as openness [50], accessibility [22,81,104], and landscape continuity [23]. Findings from the Xiaoqinhuai Riverside area corroborate this perspective—the highest scores were recorded in spatial structure and socio-cultural dimensions, indicating that the continuity of historical patterns [15,76] and the accumulation of cultural memory [59] remain decisive factors in residents’ and visitors’ perceptions.

However, this study also reveals significant differences between historic waterfronts undergoing renewal and traditional static heritage zones: both their spatial forms and user groups are undergoing dynamic change [7,20,26], while perceptions of environment and facilities exhibit phased imbalances [35,65,105]. This characteristic distinguishes it from previous research primarily focused on large-scale port-type [18] or bay-type waterfronts [13,14], aligning more closely with recent discussions on ‘micro-scale renewal’ and ‘residential waterfront spaces’ [19,30,31]. The Xiaoqinhuai River case demonstrates that within historically protected urban waterfronts guided by conservation-oriented renewal, spatial structure and cultural continuity maintain stable advantages. However, environmental perception and facility experience are more susceptible to dynamic renewal disturbances, exhibiting weakened aspects. This presents a characteristic of ‘static cultural stability—dynamic experiential fluctuation’.

Methodologically, existing public space perception studies predominantly employ questionnaire surveys [15] and fuzzy comprehensive evaluation [2,24], focusing on subjective preference analysis. This approach struggles to accurately reflect the systematic differences in perceptions across diverse groups. This study establishes a comprehensive analytical framework capable of addressing ‘multi-stakeholder, multi-indicator, and fuzziness’ issues through a structured process: literature review and synthesis → user interviews → Delphi expert consultation → AHP hierarchical analysis for weighting → cloud model perception analysis. This approach addresses the shortcomings of traditional evaluation models in integrating subjective and objective elements and expressing uncertainty, enabling the quantification and visualisation of perceptual differences between residents and visitors. Consequently, it deepens the analytical scope of studies on public space perception in historic waterfront areas.

Moreover, unlike existing studies that predominantly focus on either ‘visitor experience’ [67] or ‘resident satisfaction’ [50] as singular evaluation targets, this research employs a dual-subject comparative analysis. It reveals residents’ steady-state demands for ‘safety, cleanliness, and everyday functionality’ during renewal processes, alongside visitors’ dynamic focus on ‘cultural experiences and spatial orientation’. This structural divergence between groups corroborates Han’s (2025) [52] assertion that ‘differing usage purposes lead to stratified perception dimensions,’ underscoring that optimising public spaces in historic waterfront areas must simultaneously address both everyday functionality and experiential value.

Overall, this study maintains the foundational logic of the triadic framework—‘structure-environment-culture’—in research on public spaces within waterfront areas. Simultaneously, it achieves expansion in terms of research subjects, methodologies, and analytical dimensions:

(1)

Subjectively, shifting focus from large-scale waterfront regeneration to small-scale, residential, and historically significant waterfront areas undergoing dynamic renewal.

(2)

Methodologically, extending beyond singular subjective evaluation to incorporate subject-object coupling and fuzzy perception models.

(3)

Perspective: Shifting from single-group cognition to comparative analysis involving both residents and visitors as dual subjects.

These expansions enable the study not only to validate core theoretical propositions from existing research but also to provide a replicable research paradigm in terms of evaluation frameworks and analytical logic. This offers new theoretical support for understanding the perceptual evolution mechanisms of public spaces in historic waterfront areas during renewal processes.

6.2. Strategies for Enhancing and Optimising Public Spaces in the Historical Riverside Area of the Xiaoqinhuai River Based on Perceptual Evaluation

Based on the combined perception assessments of residents and visitors, the overall quality of public spaces in the Xiaoqinhuai Riverside Area is rated as moderate. The spatial structure and socio-cultural dimensions demonstrate strong performance, whereas landscape elements, environmental perception and facility systems are relatively weaker. Drawing upon cloud model analysis and weighting results, this study proposes corresponding optimisation and enhancement strategies across five primary dimensions.

6.2.1. Preservation and Micro-Renewal of Traditional Historical Spatial Patterns

The spatial structure dimension demonstrated the strongest overall performance among the five primary dimensions, with both residents and visitors assigning “satisfactory” ratings. Within this dimension, the highest scores were achieved for “spatial connectivity”, “sense of spatial hierarchy”, and “spatial scale appropriateness”, indicating the rational spatial organisation and continuity of historical fabric within the Xiaoqinhuai Riverside Area. According to the cloud model analysis results, spatial connectivity scored 81.923, spatial hierarchy scored 83.022, and spatial scale appropriateness scored 79.845. This indicates that both residents and visitors generally gave high ratings to the spatial experience of the area. The AHP analysis revealed these three indicators carry significant weight, particularly within the spatial structure dimension. This further validates the positive impact of the area’s historical spatial layout on user perception, confirming respondents’ widespread recognition of the rational spatial organisation and continuity of historical fabric in the Xiaoqinhuai Riverside Area. This high spatial perception quality is not coincidental but rather the dual outcome of historical accumulation and policy direction. On the one hand, the Xiaoqinhuai Riverside Area emerged from Yangzhou’s ancient city layout, where ‘waterways form streets, and streets run alongside water’. Rivers, bridges, and alleyways collectively constitute a unique linear spatial system. This interwoven structure of ‘water-street-neighbourhood-lane’ evolved over time into a highly pedestrian-scale network, offering residents and visitors a spatial experience characterised by ‘clear directionality, distinct layering, and a comfortable scale’. On the other hand, in recent years, government, society, and design departments have regarded the historical spatial pattern as a vital element of cultural heritage. During the renewal process, they adopted a strategy of ‘preserving the original layout, restoring key nodes, and optimising through micro-interventions’. For instance, the Xiaoqinhuai River conservation project explicitly stipulated principles such as ‘preserving street and lane alignments,’ ‘regulating building heights and maintaining skyline continuity,’ and ‘restoring traditional spatial order through micro-renewal techniques.’ This combination of top-down and bottom-up conservation measures has effectively prevented excessive spatial alterations and landscape alienation. Finally, we must not overlook the relatively lower scores for ‘spatial diversity’ (63.233 points) and ‘spatial accessibility’ (63.054 points) compared to the other three high-satisfaction indicators. Enhancing diversity, planning multi-scale landscape spaces, increasing the density of waterfront areas and external transport routes, and improving connectivity with public transport systems could further elevate the overall level of spatial structure.

To this end, we propose the following improvement measures:

(1)

Preserving historical layout coherence: Maintain the traditional pattern of ‘waterways forming streets, streets following waterways’, while strengthening the protection of historical elements during renewal to ensure spatial structures are not excessively altered. For instance, when planning footpaths and public facilities, avoid changing the traditional waterway layout and retain original street and alleyway alignments to ensure spatial accessibility and continuity of historical character.

(2)

Optimisation through micro-interventions: Given the high scores for ‘spatial connectivity’, further enhance the coherence of the pedestrian network. Building upon existing foundations, optimise connections between ‘waterways-streets-neighbourhoods-lanes’ by adding accessible nodes and transitional spaces, thereby improving the walking experience for residents and visitors.

(3)

Optimisation and integration of internal/external transport: Further refine traffic flow within the Xiaoqinhuai River historic waterfront area while improving external access convenience and public transport connectivity to the waterfront.

Consequently, the high spatial satisfaction of the Xiaoqinhuai River waterfront stems both from its historically evolved layout advantages and the protective regeneration mechanisms enabled by multi-stakeholder participation. Future spatial enhancements should continue adhering to the three principles of ‘preserving authenticity–sustaining walkability–enhancing recognisability’. While maintaining the traditional fabric, micro-interventions should refine node connections and waterfront accessibility, forming a continuous, open, and locally distinctive system of public waterfront spaces.

6.2.2. Systematic Reconstruction and Planning of Landscape Elements

The landscape elements dimension scored lowest among the five primary dimensions, indicating that the Xiaoqinhuai Riverside Area still exhibits significant shortcomings in landscape integrity, ecological continuity, and visual harmony. Analysis of the cloud model revealed markedly low satisfaction ratings for three indicators: “Vegetation Landscape Diversity” (score 37.915), “Hard Surfacing and Surface Materials” (score 37.834), and “Landscape Node Design” (score 36.732). Conversely, “Preservation of Historical Landscape Elements” (score 58.297) and “Integration of Green Infrastructure” (score 52.234) received comparatively higher scores. and ‘Landscape Node Design’ (score 36.732) exhibited notably low satisfaction levels. Meanwhile, ‘Preservation of Historical Landscape Elements’ (score 58.297), ‘Integration of Green Infrastructure’ (score 52.234), and ‘Colour and Character Coordination’ (score 50.321) demonstrated moderate performance. This indicates that while the existing landscape system possesses a degree of historical continuity, it remains fragmented in terms of systemic composition and functional experience.

This perceived landscape disparity stems from three primary causes: Firstly, inadequate holistic planning of the landscape system. During initial renovations, certain sections prioritised waterfront façade improvements and building restoration over continuous landscape design, resulting in fragmented greenery and isolated nodes, with no comprehensive ecological corridor along the waterfront. Secondly, inappropriate vegetation structure and material control. Certain areas suffer from insufficient green coverage, with monotonous tree-shrub-grass layering lacking seasonal variation. Concurrently, the use of mixed paving materials and overly stark colour contrasts disrupts visual continuity and compromises pedestrian comfort. Finally, node spaces exhibit limited functionality. Most nodes serve primarily as viewing points or thoroughfares, lacking the spatial experience of ‘pause-interact-rest’ and failing to meet the diverse needs of different user groups.

To address these issues, landscape optimisation should adopt a core approach centred on ‘enhancing ecological resilience—reconstructing visual order—integrating node functions’:

(1)

Enhancing Ecological Resilience: Given the low vegetation landscape diversity score (37.915), it is recommended to increase native plant species and establish multi-layered vegetation communities. The design approach of ‘multi-layered communities + flower border edges + ecological swales’ will enhance ecological diversity. This will be integrated with stormwater retention functions to form a green infrastructure system. This approach will both enrich ecological landscape diversity and effectively mitigate waterlogging issues during the rainy season.

(2)

Visual Order Reconstruction: Given the low score for hard paving and surface materials (37.834), it is recommended to employ historically resonant and visually harmonious materials such as blue bricks and traditional slate—low-saturation materials that avoid excessive colour contrast. This ensures harmony between the waterfront’s hard paving materials and the surrounding environment, enhancing the landscape’s visual coherence.

(3)

Functional Integration at Nodes: Landscape node design (score 36.732) represents the weakest aspect of the current assessment, reflecting the limited functionality of existing nodes. It is recommended to incorporate additional facilities such as waterside platforms and rest pavilions at key nodes (e.g., bridge approaches, street corners), providing more opportunities for lingering and interactive experiences. These nodes should not only serve as viewing points but also offer interactive and recreational functions, transforming them into multifunctional, composite spaces.

(4)

Revitalisation of Historical Landscapes: Building upon the high score for integrating historical landscape elements (58.297), while preserving the authenticity of historical features such as old quays, bridges, and ancient trees, incorporate display and participatory designs. This dual approach enhances both aesthetic appeal and educational value, strengthening the ‘cultural legibility’ of the site.

In summary, landscape enhancement should transcend mere visual restoration. It must address holistic coordination across ecosystems, spatial experiences, and historical character to achieve the integrated objectives of ‘systematised landscape structure, diversified landscape experiences, and unified landscape character’. Through systematic reconstruction, the Xiaoqinhuai River holds promise to evolve from ‘localised embellishment’ to ‘comprehensive renewal’, reshaping its waterside landscape into one imbued with historical depth and ecological vitality.

6.2.3. Holistic Environmental Enhancement Guided by User Perception

The relatively low scores in environmental perception dimensions indicate significant shortcomings in the Xiaoqinhuai Riverside Area concerning microclimate regulation, ecological friendliness, and environmental cleanliness. Cloud modelling results indicate that ‘thermal environmental comfort’ and ‘air quality and olfactory perception’ scored lowest, with both residents and visitors expressing dissatisfaction in these areas. ‘Acoustic environmental quality’, ‘visual comfort’, ‘ecological friendliness’, and ‘spatial cleanliness and maintenance’ were rated at moderately low levels, indicating that the overall environmental perception assessment quality remains inadequate.

The root causes of low scores in environmental perception dimensions can be summarised in three aspects: Firstly, inadequate microclimate and thermal comfort. Some pedestrian sections lack shading facilities, exhibit strong embankment reflections, and suffer from poor ventilation, resulting in poor thermal comfort during summer. Narrow walkways with extensive exposed areas lack continuous shading and green canopy systems. Secondly, suboptimal air and olfactory experiences. Limited water flow in the Xiaoqinhuai River, coupled with ageing drainage systems, has resulted in odour issues along certain stretches. Inconsistent frequency of household waste and kitchen refuse collection further compromises air cleanliness. Thirdly, visual disharmony persists. Inconsistent façade and paving colours disrupt spatial order and visual rhythm.

To address these issues, future environmental enhancements should follow a pathway of ‘microclimate regulation—ecological enhancement—multisensory coordination—maintenance system improvement’:

(1)

Optimising thermal environment and microclimate comfort: Given the low thermal comfort score (16.987), it is recommended to enhance continuous shading facilities (such as tree arrays, pergolas, or canopy structures) along the walkway areas. These should be integrated with permeable paving, rain gardens, and green belts to establish a microclimate regulation system combining ‘shading + ventilation + cooling’. Additionally, consider incorporating water features (e.g., fountains, artificial lakes) at key nodes (such as plazas and leisure zones) to enhance thermal comfort within these spaces.

(2)

Enhancing Air Quality and Olfactory Experience: Addressing the low air quality and olfactory perception score (16.444), it is recommended to improve water quality and minimise odour sources. Installing water purification facilities such as ecological floating islands and micro-aeration devices can effectively enhance water quality and improve air quality. Concurrently, optimising the surrounding waste management and collection systems will reduce the generation of odour sources.

(3)

Refining Visual and Acoustic Environmental Harmony: Visually, unify façade and paving colour schemes, minimise materials with overly intense hues, and avoid disrupting the spatial visual rhythm. Cloud model analysis indicates that façade colour coordination is pivotal to enhancing the visual environment. Consequently, grey-white stone and traditional blue bricks should be employed to reinforce visual harmony with historic architecture. For the acoustic environment, introduce soundscape planning by installing green noise barriers and water feature sound nodes at key junctions to optimise the overall acoustic experience.

(4)

Enhance ecological friendliness and environmental resilience: Increase biodiversity through multi-layered vegetation communities and ecological corridors; employ renewable materials and Low Impact Development (LID) techniques to achieve sustainable design, establishing ecological circulation systems.

(5)

Establishing a Dynamic Maintenance Mechanism: For spatial cleanliness and upkeep, implement a zoned responsibility system and regular inspection protocols to clarify operational accountability. Integrate a smart management system for real-time monitoring and feedback on waste collection, water quality, and facility maintenance, ensuring long-term environmental stability.

In summary, optimising the environmental perception dimension must transcend isolated technical fixes, adopting a user-experience orientation to achieve synergistic enhancement across six elements: thermal environment, air quality, visual perception, soundscape, ecology, and maintenance. By transitioning from ‘passive improvement’ to ‘active regulation’, the Xiaoqinhuai Riverside Area can develop high-quality public spaces combining environmental resilience with perceptual comfort, substantially enhancing the multi-sensory experience for residents and visitors alike.

6.2.4. Continuity of Historical Context and Optimisation of Spatial Inclusivity

The socio-cultural dimension demonstrated particularly prominent performance among the five primary dimensions, indicating significant achievements in cultural heritage preservation and social identity within the Xiaoqinhuai Riverside Area. Findings reveal that both residents and visitors highly value the area’s rich historical atmosphere and cultural continuity. This outcome is closely linked to Yangzhou’s ancient city’s profound historical and cultural heritage, alongside the prioritisation of cultural heritage preservation during its renewal process. Nevertheless, certain shortcomings persist in ‘social inclusivity’ and ‘publicness and spatial equity’. Commercial development at some waterfront nodes has led to unequal access to public space usage rights. While ‘cultural activities and festivities’ maintain a moderately high level, event frequency and resident participation remain limited, indicating a need to broaden the depth of cultural experiences.

This mild structural disparity reflects the tension between ‘cultural continuity’ and ‘spatial equity’ in historic waterfront areas. On one hand, government and design agencies prioritised preserving historical character and cultural narratives during renewal—restoring bridges, quays, and ancient structures to reinforce historical continuity and visual memory, making cultural expression central to spatial identity. Conversely, as cultural tourism functions intensify, certain waterfront nodes have gradually been overtaken by commercial facilities, compressing residents’ daily activity spaces and diminishing the openness and shared nature of cultural heritage transmission.

Future optimisation should follow an overarching strategy of ‘Cultural Continuity—Social Inclusion—Equitable Sharing—Activity Activation’:

(1)

Enhance historical and cultural expression: Building upon the preservation of traditional architecture, bridges, and landscape elements, further explore the historical narratives and local memories along the Xiaoqinhuai River. Employ narrative spatial design, informational guidance, and immersive display techniques to render history and culture ‘visible, tangible, and experiential.’

(2)

Deepen place memory and sense of locality: Maintain the spatial logic of waterways-lanes-courtyards, preserving the traditional imagery of ‘water reflecting street scenes’; Through atmospheric design and material continuity, strengthen residents’ emotional attachment and visitors’ sense of place, forming a chain of ‘historical continuity—emotional resonance—cultural transmission’ in spatial experience.

(3)

Enhance social inclusivity: Address diverse group needs by improving barrier-free access systems, creating age-friendly pedestrian spaces, and establishing children’s activity nodes to ensure accessibility and safety for vulnerable groups. Through ‘inclusive facilities + equitable spaces’ design, transform waterfront public spaces into genuine social platforms shared by diverse groups.

(4)

Strengthening Publicness and Spatial Equity: Limit commercial development along waterfront interfaces and prevent private encroachment on public shorelines; establish oversight mechanisms for public space usage rights to guarantee equality between residents and visitors. Institutionalise practices like ‘Public Open Days’ and ‘space-sharing initiatives’ to sustain the extension of public attributes.

(5)

Enriching Cultural Activities and Festivities: Promote a multi-tiered cultural event system integrating Yangzhou’s traditional festivals (e.g., the Mid-Autumn Moon Festival, Canal Culture Festival) with community-participatory activities (e.g., craft markets, intangible cultural heritage experiences). Increase the frequency of cultural events and broaden societal engagement, transforming cultural heritage from ‘static display’ to ‘dynamic co-creation’.

Overall, enhancing the socio-cultural dimension requires shifting from ‘static preservation of historical heritage’ to ‘dynamic perpetuation of cultural identity.’ While safeguarding cultural authenticity, mechanisms for social inclusion and equity should establish symbiotic spatial relationships between residents, visitors, and the city’s cultural fabric. Thus, the Xiaoqinhuai Riverside District will transcend its role as a repository of local memory to become a modern urban exemplar for cultural exchange and social cohesion.

6.2.5. Renewal, Layout and Maintenance of Public Facilities

The facilities system dimension exhibited the greatest fluctuation in residents’ and visitors’ overall perceptions, indicating pronounced differences in facility requirements between the two user groups. According to cloud model analysis, recreational facility comfort (score 38.013) and accessibility facility completeness (score 37.516) ranked as the lowest-scoring indicators, revealing deficiencies in the facilities system’s comfort, safety, and accessibility. Overall, the average expectation for this dimension remains at a moderate level, suggesting that the facility framework within the Xiao Qinhuai Riverside Area has taken initial shape, though shortcomings persist in comfort, accessibility, and maintenance. Cloud model results indicate relatively high scores for ‘sanitary facility completeness,’ ‘lighting facility adequacy,’ and ‘safety and emergency facilities,’ reflecting a reasonably robust basic safety and public health support system. However, scores for ‘comfort of recreational facilities,’ ‘completeness of accessibility facilities,’ and ‘anti-slip facilities’ were low, revealing shortcomings in user convenience and pedestrian safety. Additionally, scores for ‘smart and digital management facilities’ and ‘management, maintenance, and operational mechanisms’ were moderate, indicating that the digital management system is still in its infancy, with facility maintenance primarily relying on traditional manual methods.

Combined findings from interviews and field research reveal that residents prioritise the enhancement of facilities addressing daily necessities such as sanitation, lighting, slip resistance, and safety, whereas visitors place greater emphasis on experience-oriented facilities including wayfinding systems, information access, recreational amenities, and digital services. This divergence constitutes a key direction for improving the facility system.

Future optimisation should follow a systematic approach of ‘tiered provision—detailed refinement—smart management—sustained maintenance’:

(1)

Enhance fundamental public service facilities

Improve the balanced distribution of sanitation facilities, ensuring rational placement of public toilets, litter bins, and cleaning stations. Increase cleaning frequency and extend night-time lighting coverage. Addressing the high humidity and frequent night-time use of waterfront spaces, prioritise continuous anti-slip paving and safety lighting to guarantee pedestrian and recreational safety.

(2)

Enhance Rest and Accessibility Experiences

Addressing lengthy walking distances and insufficient resting points along the waterfront, increase shaded seating, waterside leisure platforms, and covered pergolas to improve spatial comfort. For accessibility, ensure seamless integration of tactile paving, ramps, and public transport systems to form a continuous barrier-free walking network. Strengthen anti-slip treatments in ground surfacing at key nodes to safeguard elderly and child users.

(3)

Optimising Wayfinding and Information Systems

Wayfinding systems should adhere to the principles of simplicity, intuitiveness, and multilingualism. They should integrate tourist guidance, cultural interpretation, and safety advisories to enhance spatial recognition efficiency and information accessibility for visitors. By installing guide maps and digital QR code systems at key nodes, an integrated online-offline guidance experience can be achieved.

(4)

Enhancing Recreational and Activity Facility Diversity

Introducing appropriate activity facilities within limited spaces, such as children’s play areas, fitness trails, and waterside social nodes, catering to diverse age groups. Concurrently, integrating festival and community events to transform waterside open spaces into temporary public venues, thereby increasing facility reuse and flexibility.

(5)

Promoting Smart and Digital Management

Employ digital solutions including smart lampposts, environmental sensors, and navigation apps to monitor facility status and manage energy consumption. Utilise IoT platforms to enable coordinated responses for lighting, drainage, and security surveillance, elevating operational automation and precision.

(6)

Establishing Long-Term Operation and Maintenance Mechanisms

The sustainability of the facility system relies on dynamic maintenance and accountability management. A collaborative mechanism should be established based on the principle of ‘government leadership—corporate participation—resident oversight’. Through Service Level Agreements (SLAs) and digital inspection systems, real-time feedback and rapid response to facility malfunctions, cleaning management, and spatial governance are enabled, forming a quantifiable and traceable maintenance framework.

Overall, optimising the facility system concerns not only physical provision but also reflects comprehensive public space governance capabilities. Through a continuous strategy of ‘infrastructure enhancement—user experience improvement—intelligent management—long-term operational maintenance’, the Xiaoqinhuai Riverside Area can achieve modernisation and human-centred design of its historic waterfront facilities while ensuring convenience for residents and comfort for visitors.

6.2.6. Mid-Term and Long-Term Planning and Enhancement Strategy

To ensure the continuous enhancement of public spaces along the Xiao Qinhuai Riverside area over the coming years, while balancing historical preservation with modern requirements, future planning and improvement strategies will be divided into medium-term and long-term phases.

Medium-Term Plan (1–3 years)

(1)

Optimisation and Renewal of Spatial Functions: The medium-term plan should focus on enhancing the functionality and accessibility of public spaces. Firstly, micro-upgrades should be implemented at key nodes along the waterfront, introducing waterfront platforms, pedestrian pergolas, and leisure seating to enhance existing recreational facilities, thereby improving comfort and convenience. Secondly, connectivity of footpaths should be strengthened, particularly along traffic routes linking historic districts with water bodies, ensuring smooth pedestrian and bicycle access to enhance the waterfront’s accessibility.

(2)

Preliminary Landscape System Optimisation: Based on cloud model analysis indicating low scores for vegetation diversity and hardscaping, the medium-term plan should prioritise landscape node design and green corridor enhancement. By introducing native flora and seasonal green belts, ecologically resilient green infrastructure will be established, elevating ecological functionality and aesthetic appeal.

(3)

Promotion of Cultural Activities: To strengthen cultural identity within the historic waterfront area, the mid-term plan should establish cultural exhibition zones and organise regular cultural festivals (such as Yangzhou traditional celebrations and folk art performances). This will foster residents’ and visitors’ connection to and participation in local culture. Concurrently, develop distinctive leisure and cultural programmes incorporating traditional cultural elements to enhance the area’s cultural appeal.

(4)

Establishing a Long-Term Effect Tracking and Evaluation Mechanism: Considering the limitations in assessing long-term effects within this study, it is recommended to establish a long-term tracking and evaluation mechanism during the mid-term phase. This involves regularly collecting data and assessing the sustained impact of renewal measures. Through periodic questionnaires, field interviews, and environmental monitoring, the effects of renewal measures on spatial perception should be continuously tracked. Strategies should be promptly adjusted and optimised to ensure the sustained and stable effectiveness of public space transformations.

Long-Term Planning (3–10 Years)

(1)

Holistically enhance ecological resilience and sustainability: The long-term plan should further strengthen ecosystem construction and conservation. For instance, developing more comprehensive ecological greenways and wetland parks will forge closer connections between the waterfront and surrounding natural environments, safeguarding regional biodiversity and enhancing ecological friendliness. Concurrently, introducing facilities such as rain gardens and ecological floating islands can improve the area’s stormwater retention capacity and water purification functions.

(2)

Deepening Historical and Cultural Revitalisation: The long-term vision must continue prioritising the preservation and display of historical heritage. Designs incorporating heritage trails and cultural landmark nodes should transform the Xiao Qinhuai Riverside into a living showcase of history and culture. For instance, installing digital displays and interactive cultural interpretation platforms at key points would enable visitors and residents to experience history and understand culture within the space, thereby enhancing its cultural appeal and educational function.

(3)

Integration of Smart Technology and Digital Management: With technological advancement, future public spaces should progressively incorporate intelligent management systems. By introducing Internet of Things (IoT) technology, real-time monitoring and feedback on environmental quality, facility operations, and visitor flow can be achieved, optimising management and service standards. Furthermore, the long-term plan should pursue the application of digital twin technology for spatial simulation and optimisation, enabling real-time monitoring of public space usage to enhance operational efficiency and responsiveness.

(4)

Sustainable Operations and Multi-Stakeholder Management Models: To ensure the long-term viability of public spaces, long-term planning should establish multi-stakeholder management mechanisms. By combining the strengths of government, social organisations, commercial entities, and residents, explore a management model characterised by ‘government leadership, market-oriented operations, and resident participation.’ This ensures waterfront public spaces can be sustainably operated and optimised through the collective efforts of diverse stakeholders.

(5)

Ongoing Evaluation and Adjustment of Long-Term Outcomes: To guarantee the sustained effectiveness of renewal measures, the long-term plan should establish a mechanism for evaluating and feeding back on long-term outcomes. By integrating long-term data collection, environmental monitoring, and user feedback, the effectiveness of public space usage and shifts in user perceptions should be regularly assessed, enabling timely adjustments to planning strategies. Tracking and evaluating long-term outcomes will form a crucial foundation for ensuring the sustainable renewal of historic waterfront areas.

Through systematic mid-term and long-term planning and implementation, the future Xiao Qinhuai Riverside District will not only achieve new progress in preserving historical and cultural heritage but also realise comprehensive enhancements in ecological, social functions, and intelligent management. It will emerge as a vibrant, inclusive, and sustainable model for modern historical waterfront districts.

7. Conclusions

7.1. Key Findings of the Study

This research employs the historical waterfront area of Yangzhou’s Xiaoqinhuai River during its renewal process as an empirical case study. By establishing a systematic research framework comprising ‘indicator system construction–weight determination—perceptual evaluation–optimisation strategy’, it conducts a comprehensive assessment of the perceived quality of public spaces within urban historical waterfront areas undergoing renewal. Targeted enhancement and optimisation strategies are subsequently proposed. Firstly, the study successfully established a perceptual evaluation system for urban historic waterfront public spaces, comprising five primary dimensions and 33 secondary indicators. This framework underwent multi-stage validation through literature review, field research, and the Delphi expert consultation method, ensuring the scientific rigour and comprehensiveness of its metrics. It provides a reference evaluation framework for similar studies. Secondly, findings indicate that the comprehensive perceived quality of public spaces along the Xiaoqinhuai River waterfront remains at an average level. This reflects initial progress in renewal efforts while highlighting significant room for improvement. Specifically, socio-cultural aspects and spatial structure demonstrate excellent performance, achieving satisfactory levels; the facilities system dimension remains average; whereas landscape elements and environmental perception receive relatively negative feedback. This outcome reveals an imbalance between ‘static cultural preservation and dynamic usage experience’ within the renewal process. Cloud model data and visualisations further indicate that both visitors and residents express notably low satisfaction and a distinctly negative attitude towards factors including the diversity of vegetation landscapes; hard paving and surface materials; landscape node design; thermal environmental comfort; air quality and olfactory perception; the comfort of rest facilities; the diversity of activity facilities; and the adequacy of anti-slip measures. This highlights areas that require urgent attention and improvement. Secondly, the findings reveal that residents’ and visitors’ perceptions exhibit characteristics of ‘coincidence and divergence coexisting’. Both groups show high consistency in evaluating the socio-cultural and spatial structure dimensions, yet display significant divergence in the facilities system dimension. This discrepancy stems from the tension between residents’ emphasis on daily functionality and visitors’ focus on experiential quality, highlighting the dual-attribute contradiction of historic waterfront areas as both ‘living spaces’ and ‘tourist spaces’. Finally, based on the evaluation outcomes, this study proposes differentiated enhancement and optimisation strategies for public spaces in urban historic waterfront areas across five dimensions, grounded in perceptual assessments.

7.2. Theoretical Methodological Innovation and Practical Significance

This study introduces a novel framework for assessing perceptions of public spaces within historic urban waterfront areas, addressing limitations of randomness and ambiguity in previous models. The findings provide a more precise and scientifically robust evaluation, holding significance for both theoretical advancement and practical conservation efforts. Through the integration of the AHP–Cloud Model, the framework achieves quantitative treatment of the fuzziness and randomness inherent in public space perceptions. The three parameters of the Cloud Model (Ex, En, He) not only reflect the overall level of perception but also precisely delineate the distribution patterns of group differences. Compared to traditional evaluation methods, this framework simultaneously captures the central tendency (Ex) and dispersion characteristics (He) of subjective evaluations, offering novel insights for analysing perceptual conflicts among diverse user groups. Theoretically, this study extends the research perspective from large-scale port waterfronts to small-scale, residential historical waterfronts, revealing the particularities of spatial quality evolution within the context of ‘micro-renewal’. The proposed four-dimensional evaluation model—encompassing structure, environment, culture, and facilities—enriches the theoretical framework for urban public space research, offering fresh perspectives for balancing the tensions between conservation and development. The derived differentiated enhancement and optimisation strategies are not only applicable to the Xiaoqinhuai Riverside but also provide an operational reference template for renewal practices in similar historic waterfront areas elsewhere. Particularly for traditional neighbourhoods undergoing ‘conservation-oriented renewal,’ the ‘diagnosis-evaluation-optimisation’ pathway proposed in this study holds significant demonstrative value.

By integrating the Analytic Hierarchy Process (AHP) with cloud computing models, traditional questionnaire-based evaluation methods have been effectively optimised. Traditional questionnaire-based assessments typically rely on subjective feedback. While reflecting user needs, they often lack quantitative analysis and are susceptible to individual differences and response biases. The AHP–Cloud Model transforms qualitative aspects of perception evaluation into quantitative analysis, providing a more precise and scientific assessment framework. The AHP method clarifies the importance of each evaluation dimension through expert weight allocation, thereby more accurately reflecting the impact of different factors on user perception. The cloud model further addresses ambiguity and uncertainty in perception, ensuring that assessment outcomes possess quantitative foundations while illustrating interrelationships and priority hierarchies across dimensions. Through these methodologies, this study achieves clearer identification of specific requirements among distinct user groups (such as residents and visitors), circumventing potential biases arising from subjective judgements inherent in conventional approaches.

Compared to conventional questionnaire-based assessments, the AHP–Cloud Model offers a more objective and precise evaluation approach. This enables research findings to not only reflect the issues themselves but also provide actionable, evidence-based improvement strategies for subsequent spatial optimisation. Through this innovative methodology, this study delivers theoretical support for optimising public spaces in historic waterfront areas while establishing a new methodological framework for future comparable research. Consequently, it significantly elevates the academic contribution and practical significance of the investigation.

7.3. Research Limitations and Future Prospects

The limitations of this study are primarily reflected in four aspects:

(1)

The case selection exhibits strong regional characteristics: This research focuses on the historical waterfront area of the Xiaoqinhuai River in Yangzhou, situated within China’s quintessential Jiangnan water town region, which possesses a profound water culture heritage. The form and function of public spaces within this area are profoundly shaped by regional culture and history. Consequently, the study’s conclusions may exhibit significant regional specificity, potentially limiting their applicability to other parts of China or different cultural contexts. For instance, historical waterfront areas in northern cities or inland regions may exhibit notable differences in spatial perception, historical-cultural continuity, and social functions, thereby reducing the generalisability of findings to such areas. Future research could conduct comparative multi-case studies to further explore commonalities and differences in public space perceptions within historical waterfront areas across diverse regional cultural contexts, thereby validating the generalisability of this study’s conclusions. Moreover, the study’s sample size is relatively small, and regional cultural particularities may limit the generalisability of its findings. Therefore, expanding the research sample to include additional waterfront case studies from diverse cultural contexts and geographical locations would enhance the extrapolation of these conclusions.

(2)

Cross-sectional data struggles to capture the dynamic evolution of perceptions during renewal processes: This study employed cross-sectional data, primarily reflecting public space perceptions at a single point in time. The absence of longitudinal tracking data prevents the capture of dynamic shifts in perceptions during renewal processes. Consequently, future research should employ longitudinal tracking designs to analyse perceptual changes across different time points and examine the long-term impact of renewal measures on spatial perceptions.

(3)

Cloud model calculations are relatively complex, necessitating tool simplification for practical implementation: While the cloud model provides a scientific quantitative approach for perception assessment, its computational process is intricate and requires substantial technical support. A key challenge for future research lies in simplifying this model and developing toolkits for broader application in assessing public spaces within other historic waterfront areas. Streamlined tools would enhance the method’s dissemination and implementation in practical projects.

(4)

Seasonal factors’ influence on perception assessment remains under-examined: While this study acknowledges climate’s impact on thermal comfort, it does not thoroughly explore seasonal factors’ potential effects on data collection and perception. Weather conditions (such as extreme heat or cold) may significantly skew thermal comfort ratings, thereby influencing overall perceptions. Consequently, future studies should collect data across different seasons to comprehensively evaluate climate factors’ impact on spatial perception. Furthermore, incorporating correlation analyses between meteorological data and thermal comfort ratings would provide a more scientific understanding of how climatic conditions shape users’ spatial experiences.

(5)

Strengthening the monitoring and analysis of objective indicators: Although this study employed quantitative methods such as cloud modelling to analyse perceived outcomes, reliance on subjective evaluations may still introduce biases stemming from individual differences and environmental factors. To enhance the scientific rigour and accuracy of future research, greater emphasis should be placed on monitoring and analysing objective indicators.

Future research may address the limitations of this study and provide more comprehensive theoretical support for optimising public spaces in historic waterfront areas through the following avenues:

(1)

Multi-case comparative studies: As this research focused on the historic waterfront area of Xiaoqinhuai River in Yangzhou, future work could expand to other regions, conducting cross-regional comparative studies to explore commonalities and differences in the perception of public spaces within historic waterfront areas across distinct cultural contexts, thereby validating the generalisability of findings.

(2)

Longitudinal tracking design and long-term impact analysis: Future research should employ longitudinal tracking designs to continuously monitor data at different time points within historic waterfront areas. This would analyse the long-term effects of renewal measures on perceptual changes, thereby revealing evolving trends in perception throughout the renewal process.

(3)

Integration and Analysis of Climatic Factors: Considering seasonal variations’ impact on public space perception, future studies may combine meteorological data with thermal comfort ratings to examine climate change’s specific effects on user perceptions, thereby providing climate-adaptive design solutions.

(4)

Digital Twins and Real-Time Monitoring Systems: Future research may integrate evaluation frameworks with digital twin technology to enable real-time monitoring and optimisation of spatial quality. Through virtual simulation and big data analysis, this approach enhances the efficiency and precision of public space management.

(5)

Utilising environmental sensors to gather objective data: To enhance the scientific rigour of the research, future studies may employ environmental sensors (such as air quality monitors and noise level detectors) to collect actual environmental data. This approach combines objective indicators including air quality parameters (e.g., PM2.5 and CO₂ concentrations) and noise levels. Such data provides stable, reliable environmental conditions that serve to validate and supplement subjective evaluation outcomes.

By deepening these research directions, future guidance for the design and management of public spaces within historic waterfront areas can be rendered more scientific and sustainable.