A Sandglass Tiered Model for Integrating Cultural Value into Built Environment Management

Abstract

Built environment elements management involving heritage buildings requires a nuanced approach that balances cultural preservation, planning efficiency, and resource optimization. Conventional evaluation methods frequently neglect public perception, leading to misaligned priorities and ineffective heritage resource deployment. To address this gap, this study proposes a Sandglass Tiered Model that integrates public perception into the value assessment process of culturally significant buildings. By integrating multi-source perception data and cultural ontology through a structured, tiered data collection mechanism, the model translates subjective views into four quantifiable indices: symbolizability, authenticity, readability, and regionality. These indices form the basis of an AHP-GRA–driven assessment framework, facilitating value-based prioritization and spatial zoning of heritage elements within construction projects. The model was empirically validated in the Yiling Cultural Heritage Area, where it effectively facilitated differentiated building strategies, optimized resource sequencing, and improved alignment between project goals and stakeholder expectations. Importantly, the model provides a transferable framework that embeds cultural awareness into the lifecycle of heritage building projects—from pre-design evaluation to renovation and adaptive reuse. By integrating public perception into construction workflows, this approach provides a dynamic and participatory framework for managing complex heritage assets within urban development contexts. It improves the precision, responsiveness, and cultural sensitivity of construction planning, offering practical insights for policymakers, architects, and construction managers working in resource-intensive or culturally rich environments.

1. Introduction

In the context of urban regeneration, cultural heritage resources have become pivotal assets for the sustainable development of historic districts, industrial building sites, and old neighborhoods [1]. It facilitates the transformation of static heritage into dynamic cultural scene experiences through spatial narrative and experience design [2]. Such a transition infuses cultural vitality into the built environment and reinforces regional cultural identity through reinterpretation of symbolic values [3]. However, the surge of data within interactive cities has triggered structural tensions in cultural cognition: the public cognitive bias towards heritage symbols is hindering spatial sustainable development [4]. This cognitive bias reveals the disparity between maintaining cultural authenticity and adapting to public cognition in current construction regeneration projects. Current evaluation models overlook the spatial heterogeneity of public perception and oversimplify symbol diversity, leading to imbalance in spatial resource allocation [5], with high-value heritage being marginalized due to low symbol transmission or its position in peripheral locations.

Previous studies have achieved significant progress in the integration and valuation of architectural heritage resources. Regarding the development of classification systems [6,7], scholars have established classification criteria based on cultural element composition models [8], from perspectives such as temporality, spatiality, typological characteristics, and content attributes [9]. Research in heritage resource management has incorporated technologies such as GIS and 3D scanning, enabling spatial and digital management of resources [10,11,12]. In terms of value assessment [10], mixed-methods approaches are frequently employed to evaluate the value of cultural symbols and their core role in regional resource integration [13], destination image formation, and product development. For instance, Ancuta employs participatory ethnographic observation and logical framework analysis to evaluate the value of regional cultural resources [14]. Zhang established a factor analysis-based evaluation system to assess tourism development potential [15]. Alongside the growth of cultural tourism, the research scope has broadened to include semiotics, archaeology, and product design and has increasingly incorporated public perception data [16,17,18]. For instance, Sun uncovered tourists’ implicit demands for cultural symbols [19], while Cui identifies user characteristics through the analysis of online comment data [20]. Grey relational analysis (GRA) is a method designed for modeling systems characterized by uncertain information [21]. It quantifies the relationship between influencing factors and system outcomes by calculating the gray relational association. In the field of design science, the GRA is frequently applied in value assessment [22]. Peng developed a GRA model to assess the PCR of typhoon catastrophes in China, seeking to represent the ambiguity of human perception [23]. Wen introduced a GRA-MCS model framework to identify the optimal solution from a set of candidate models [24].

But the following limitations persist. Firstly, resource classification and planning overly depend on expert decision-making and lack quantitative methods for resource combination and high-value element selection that incorporate various demands [25]. Secondly, while the symbolic value assessment employs a mixed-method approach, it predominantly relies on tourism economic indicators [26,27], often overlooking the residents. Although residents are the primary users of urban renewal projects [28], their genuine and profound understanding of heritage symbols is rarely integrated into early-stage decision-making [29]. Furthermore, the GRA method quantifies relations in uncertain systems but disregards spatial carriers in the context of cultural heritage [30], neglecting the fuzzy demands of multiple target groups regarding built environment elements. Fourth, research on heritage buildings tends to focus on protective techniques or the symbolic design and application of individual structures, ignoring the contribution of regional cultural heritage in built environment elements and lacking a systematic approach to spatial intervention. These limitations frequently lead to conservation strategies being detached from community spatial experiences, intensifying the risk of deterioration for significant buildings due to symbolic misreading or locational marginalization.

To address these limitations, this study proposes a perceived demand-driven framework for assessing the value of heritage elements and facilitating tiered decision-making. The framework centers on a sandglass tiered model, dynamically collecting heritage elements by integrating the perceptual demands of multi-stakeholder groups, and innovatively combines the Analytic Hierarchy Process (AHP) with Grey Relational Analysis (GRA). Firstly, we established a hierarchical structure through perceptual demand analysis, quantifying and balancing the conflicting perceptual demands from multi-stakeholder groups. Secondly, the GRA is applied to transform these ambiguous perceptual demands into quantifiable evaluation indices, thereby constructing a reference sequence. Then, we systematically collect and screen heritage resources according to hierarchical demands, forming a comparison sequence composed of elements that align with public perceptual demands. Finally, tiered response strategies are formulated based on the value assessment outcomes. This study contributes to the existing literature by proposing some original approaches to the research issue. First of all, by dynamically integrating public perceptual demands into heritage resource collection, our approach breaks the traditional expert-centered model, balancing authenticity preservation with diverse cognitive adaptation. The second major contribution is the construction of a sandglass tiered model, which offers an actionable pathway for integrating uncertain demands and assessing the value of diverse heritage elements within complex systems. In summary, this study establishes a transformative framework structured around “demand integration-resource collection-value assessment-tiered response strategies.” This framework provides decision-making support for urban renewal that integrates heritage sustainability with community experience, offering both theoretical and practical significance.

This study is structured as follows: Section 2 details the data and describes the sandglass tiered model, with a focus on the integration of public perception demands, the transformation of indices, and a comprehensive description of the model. The performance and outcomes of the model are presented and analyzed in Section 3. Section 4 discusses the model’s applicability and replicability. Last, Section 5 contains a discussion of future work and the conclusion.

2. Sandglass Tiered Model

2.1. Model Architecture

When addressing the multi-factor correlation issue in complex systems, the AHP is widely used for breaking down intricate factors into a quantifiable system of criteria and sub-criteria [31]. However, its linear decision-making model struggles to reconcile conflicting demands among multi-objective groups. Although the Analytic Network Process (ANP) can describe the network relationships among factors, it is unsuitable for handling the interactions among multiple complex factors due to structural and operational complexity. The GRA offers a concise and effective approach for analyzing the relationships among multiple complex factors by measuring the correlation degree between individual factors and reference sequences. However, its reliability may be weakened by the subjectivity involved in the model construction process [32]. Addressing the core issue of resource collection and value assessment driven by the fuzzy perceptual demands of multi-stakeholder groups in cultural heritage, this study integrates the AHP with the GRA to construct a sandglass tiered model.

The model incorporates dual pathways for demand integration and resource screening at the data collection layer, alleviating expert reliance and the lack of systematic intervention while curbing the marginalization of non-core area elements. The intermediate layer serves as the evaluation indices layer, which mitigates the rigidity inherent in traditional linear decision models by transforming conflicting demands into quantitative indices. The value assessment and strategy layer employ a hierarchical iterative mechanism to progressively transform fuzzy demands into quantitative indices and eventually into evaluative outcomes, thereby accommodating the uncertainty inherent in cultural heritage perception. This layered decision logic allows the model to avoid the complexity caused by excessive node correlations in network models, hence offering a clearer and more operational method for modeling multiple complex relationships. The model architecture is illustrated in Figure 1.

The data collection layer refers to the collection and organization of data on the perceived subjects and objects. We identify stakeholder roles and locate multi-stakeholder perception groups, obtaining and analyzing perception demand data through a multi-source data gathering method. Then, we analyze and categorize cultural heritage resources using a categorization system for heritage elements, clarifying the perceived subjects and objects of cultural heritage resources. The cultural heritage resources at this layer have undergone initial screening according to the demands of multi-stakeholder groups. The discovered heritage elements, regarded as perceptual objects, have initially met the perceptual demands of four distinct groups.

The evaluation indices layer refers to the integration of the perceptual demands of multi-stakeholder groups. Utilizing methods such as cultural imagery word descriptions, expert assessments, and consensus meetings, fuzzy perceptual demands are converted into four-dimensional evaluation indices. These evaluation indices are general and relevant for assessing the value of different cultural resources.

The value assessment and strategy layer includes value assessment and resource optimization strategies for cultural heritage elements. The value assessment refers to the assessment of cultural heritage symbols, integrating public perceptual demands. The perceived demands from four groups are identified as evaluation indices and serve as reference sequences. The collected cultural heritage elements are utilized as evaluation objects and serve as comparison sequences. Then, the Gray Relational Analysis method is employed to quantify the index and assess symbolic value. Finally, we use semi-structured interviews to validate the consistency of the ranking results and identify the distribution areas of heritage elements. The resource optimization strategies involve differentiated construction strategies for tiered response. Following the assessment of cultural heritage elements, which serve as input data, appropriate construction strategies are formulated for heritage elements across different distribution areas.

The methodology and research protocol involved in this study were approved by the Ethics Committee of the School of Design and Art, Shaanxi University of Science and Technology. All methods were carried out in accordance with relevant guidelines and regulations. Informed consent was obtained from all participants.

2.2. Demands Integration and Indices Identification

The core of the sandglass tiered model is the demand-driven prioritization and tiered management of heritage elements. We incorporate the perceptual demands of stakeholders through three interrelated steps: multi-stakeholder demand acquisition, demand integration, and indices identification. The integrated perceptual demands are used to assess the degree of relevance with each heritage element, thereby facilitating priority ranking and tiered management.

2.2.1. Target Audience Identification and Data Collection

To achieve synergy between cultural symbols and sustainable development goals, SDG 11.4 serves as our primary focus. Data collection dimensions were derived from three sustainability pillars: social, environmental, and economic. Within the social sustainability dimension, preserving cultural identity of heritage symbols is essential while strengthening social cohesion and enabling effects. The environmental sustainability dimension requires maximizing cultural significance transmission while minimizing resource consumption. For economic sustainability, activating the industrial transformation potential of heritage symbols is critical. Building on this tripartite framework, the AHP is used to establish tiered data collection logic. We construct a sandglass tiered collection method from three dimensions: cultural authenticity, public acceptance, and policy adaptability.

At first, we identify the roles of stakeholders according to the urban population structure and adjust the population composition by combining preliminary field surveys and expert reviews. The perception subjects are divided into four groups to comprehensively cover the perception, aesthetics, and management dimensions of heritage symbols: residents, tourists, cultural workers, and government workers. Residents refers to registered individuals and new immigrants who have resided continuously for over 10 years. Cultural workers includes researchers, inheritors of intangible cultural heritage, and designers engaged for at least three years in heritage protection, cultural communication, or heritage transformation. Government workers refers to staff members in the Cultural and Tourism Department tasked with heritage resource planning.

Then, different groups have varied perception channels, so multiple data survey methods should be combined to obtain initial data on the core people. In this study, we recruited participants from diverse groups through on-site recruiting, community organizations, and government referrals while acquiring typical samples and key information through purposive sampling. Additionally, we expanded the diversity of samples using snowball sampling. Among them, the perceptual demand data from cultural workers and tourists were collected through dual-mode (online and offline) questionnaire surveys. To mitigate cultural cognitive bias stemming from varying educational backgrounds, the field survey was employed to gather data from residents. This strategy helps participants in establishing a basic understanding of the target heritage elements by suitably presenting the heritage resource attributes during the initial phase of demand data collecting. We collected perceptual demand data from government workers through semi-structured interviews, effectively capturing the policy-oriented perspectives during the dialogue process. The suitable survey methods and perceptual demand integration paths are shown in Figure 2.

After determining the target audience and varied data collection methods, this study executed questionnaire design and made adaptive adjustments. We designed a basic questionnaire focusing on public perception and design preferences as primary concerns. Subsequently, we carefully adjusted the content presentation, question complexity, question form, and data collection processes of the basic questionnaire according to various target audiences and data collecting methods. This adjustment aims to ensure that the data-collecting methods align with the features of diverse respondents, thereby obtaining more genuine and reliable information. Compared to traditional parallel group analysis, this tiered sorting logic for data gathering may accurately identify the symbol, create explanatory items, and organize target elements through tiered criteria. Additionally, it can also significantly decrease the likelihood of disconnection between the signifier and the signified.

2.2.2. Demands Integration

We used a comprehensive analysis method to analyze the gathered data. Quantitative analysis was used to process structured data from the surveys, including preference rankings and ratings. The qualitative inductive method was used to analyze open-ended texts derived from interviews and observational data. We systematically sorted the relevant options in the questionnaire data and performed preliminary categorization and organization of these options according to the established research framework and the features of the emerging data. Employees engaged in regional planning for cultural heritage development make up the fourth group. We performed semi-structured interviews with this group to thoroughly investigate their unique insights and potential needs. We strictly followed the processes of the thematic analysis method to analyze those interview data. At first, we extract all original statements related to the research topic from the interview data. Next, we follow the principle of openness to extract initial codes with significant meaning from these original statements. At last, we classify and integrate codes with similar or repetitive attributes to form thematic codes.

To address the disparities within the group and improve the objectivity of data processing and the reliability of analysis outcomes, we invited 12 people to form Expert Group A, including 5 designers, 4 cultural researchers, and 3 professionals from the cultural tourism industry. Initially, we allowed and documented the divergent opinions within each group. To identify the predominant perception trend with the highest consensus in each group, we used an expert consensus meeting to categorize the collected perception data. Representative demands were extracted based on their frequency and emphasis of occurrence. Expert Group A will address any contentious or ambiguous opinions and coding through consensus meetings to avoid oversimplifying the complexity within groups. Ultimately, following several rounds of categorization, discussion, and consensus-building, the 12 experts achieved an agreement, forming core viewpoints that represent the various groups.

2.2.3. Indices Identification

According to the expert consensus meeting in Section 2.2.2, the perceptual demands of each group were integrated sequentially:

• Cultural workers think that cultural symbols should completely reflect cultural imagery, especially the style of the region, which should be well understood and actively promoted.
• Residents hope to enhance cultural perception in many ways and develop cultural symbols that reflect more regional character.
• Tourists hope that cultural symbols are simpler and easier to understand, thus enriching their interaction experience with the culture.
• Government workers advocate for cultural heritage symbols that fit the development planning of the tourism, creating unique cultural symbols and a range of products, which may also be expanded into themed experience activities.

Once the perceptual demands of the four stakeholder groups are identified, the complexity of the demand description hinders future calculations. Therefore, the Kansei Engineering method was used to decode each perceptual demand into imagery descriptive words. Expert Group A was invited to describe the perceptual demands of four distinct groups using one or more words. Redundant or highly similar words were subsequently eliminated. Since cultural imagery words require repeated assessment for a relatively accurate evaluation index, the remaining cultural imagery words were then assessed individually using a five-point Likert Scale to ensure stylistic consistency [33]. Those words with higher mean scores were selected as evaluation indices for heritage symbol elements. As shown in Figure 3, we sorted the average values of all words from highest to lowest and selected “symbolizability (4.88), authenticity (4.52), readability (4.22), and regionality (4.12)” as the indices for the symbol value assessment of cultural heritage due to their higher average values. These four evaluation indices, derived through the collection and integration of demands, as well as the identification and quantification of indices, serve as the assessment criteria for demand-driven heritage management and form the reference sequence for subsequent relational analysis.

2.3. Cultural Heritage Resource Collection and Element Classification

The collection and sorting of cultural heritage resources form the data foundation. We systematically collect regional cultural resources through fieldwork and collaboration with LAM (libraries, archives, and museums) institutions. The original data were refined and corrected based on archaeological studies and literature. First, the survey data were cross-verified with existing literature to complement missing information, ensuring the authenticity and completeness of heritage elements. Subsequently, redundant data were consolidated to eliminate duplicates. Finally, relevant research and archeological studies were referenced to verify the aesthetic styles of the collected data. Samples that did not align with the regional aesthetic style were excluded.

The data collected from multiple sources may have issues with information fragmentation. We use tiered clustering methods to classify and describe the distribution and relationships of elements in the original dataset [34]. We construct a framework for the composition of cultural heritage elements from the views of data decomposition and correlation, as shown in

Table 1. The framework for the composition of cultural heritage elements.

Meta (Resource Type)	Chain (Cultural Category)	Cluster (Heritage Elements)
Natural Cultural Heritage (NC)	Architecture, Landform, Tourism, Resources	Individual heritage elements within each category
Historical Cultural Heritage (HC)	Regional culture, Legends, History, Handicraft

. This framework classifies the original data based on the “Meta-Chain-Cluster” system. Among them, data clusters refer to the combination of individual heritage elements and their information, which are the main research objects.

2.4. Cultural Heritage Symbol Value Assessment Based on the AHP-GRA

The presence of multiple evaluation indices introduces ambiguity and complex interrelationships among them. This study takes into full account the richness of regional cultural resources and the underlying cognitive diversity of public perception. Accordingly, we employ assessment methods based on the AHP-GRA to quantify the assessment process and identify heritage elements that more closely correspond with public perception [35]. The procedure is outlined below.

2.4.1. Construction of Reference and Comparative Sequences

Scholars generally regard the evaluation subjects as the comparison sequence and the evaluation indices as the reference sequence. Heritage elements serve as the comparison sequence, denoted as $X_i$, and the average value of the evaluation indices are used as the reference sequence, denoted as $Y$. When selecting among m heritage elements to be assessed for their alignment with public perceptual demands, the prioritization of these elements is correlated with n evaluation indices. The comparative sequence of each heritage element is expressed as follows:

$$X_i = X_i\left(k\right) \left(k = 1, 2,\cdots , n, \left.i = 1, 2,\cdots , m\right)\right.$$

Based on public perception demands, the reference sequence is set as follows:

$$Y = Y\left(k\right) \left(k = 1, 2,\cdots , n\right)$$

$k$ is the evaluation index, and $i$ is the element to be evaluated that meets public perception demands.

2.4.2. Standardization Process

Expert group A was invited to evaluate the coded heritage elements on a 10-point scale, calculating the average value for each element. Due to the incommensurability among the various assessment data, the averaging method is used to normalize the data:

$$X_{i}k = {\displaystyle \frac{x_i\left(k\right)}{\overline{x_i}}}$$

(1)

In the formula, $X_i\left(k\right)$ is the standardized average of cultural element $i$ on the $k$ index, $x_i\left(k\right)$ is the value of each sequence, and $\overline{x_i}$ is the average of the whole sequence.

2.4.3. Relevance Calculation and Ranking

The formula to calculate the gray relational degree of cultural element $X_i$ with respect to the reference sequence $Y$ at the $k$ index is as follows:

$${\xi }_i\left(k\right) = {\displaystyle \frac{\underset{i}{\mathit{min}} \underset{k}{\mathit{min}}\left|y\left(k\right) - x_{i}k\right| + \rho \underset{i}{\mathit{max}} \underset{k}{\mathit{max}}\left|y\left(k\right) - x_{i}k\right|}{\left|y\left(k\right) - x_{i}k\right| + \rho \underset{i}{\mathit{min}} \underset{k}{\mathit{min}}\left|y\left(k\right) - x_{i}k\right|}}$$

(2)

In the formula, $\left|y\left(k\right) - x_{i}k\right|$ is the absolute difference obtained by sequentially subtracting the reference object from each comparison object. A lower absolute difference shows a greater relation. $\underset{i}{\mathit{min}} \underset{k}{\mathit{min}}\left|y\left(k\right) - x_{i}k\right|$ is the minimum absolute difference of element $i$ on evaluation index $k$, while $\underset{i}{\mathit{max}} \underset{k}{\mathit{max}}\left|y\left(k\right) - x_{i}k\right|$ is the maximum absolute difference of element $i$ on evaluation index $k$. $\xi$ is the resolution coefficient. A smaller $\xi$ means a higher discriminative power. $\xi \in \left[0, 1\right]$, conventionally set $\xi \le 0.5$.

Finally, the correlation degree between each heritage element and public perception demands is calculated. The degree of correlation between the $ith$ comparison object and the reference object is formulated as follows:

$${\gamma }_i = \sum _{k=1}^n{\omega }_i{\xi }_i\left(k\right)$$

(3)

In this formula, ${\gamma }_i$ is the average relational degree of the $i\mathrm{t}\mathrm{h}$ cultural element. A greater ${\gamma }_i$ shows a higher correlation between the cultural element and public perception demands, thereby achieving the optimal ranking of heritage elements.

2.5. Differentiated Construction Strategy for Tiered Response

2.5.1. Determination of Cultural Heritage Element Zones

We invited government workers to conduct semi-structured interviews, compared the findings with rankings, and then determined the zoning of heritage elements. The interview discussed two topics: the usefulness of industrial transformation and the adaptability of cultural development planning. Government workers were asked to assess the priority heritage elements individually, taking their average values as the X and Y coordinates and using the gray relational degree to determine the size of the bubbles for a scatter plot. If the axis data and bubble area show a positive correlation, it verifies that the public perception-driven cultural heritage symbol value assessment can effectively resolve group cognitive differences.

2.5.2. Proposal of Differentiated Construction Strategy

Cultural heritage elements situated in Area I show high adaptability and usability. These elements exhibit significant value in four-dimensional indices, align closely with regional planning requirements and the potential for industrial transformation, and can be prioritized as high-value cultural resources for development. The high-value intense area should adopt a strategy of overall protection and complete interpretation, including spatial atmosphere creation, industrial chain development, and the application of intelligent interpretation technologies. The spatial atmosphere creation emphasizes the combination of cultural heritage symbols with the built environment, creating constructions or scenes that reflect cultural spirit, and an improvement of public perception through spatial art. The industrial chain development strategy involves the establishment of a systematic design project focused on cultural intellectual property (IP), including IP visual system design, brand system construction, cultural and creative products (CCPs) design, and thematic event planning. The extension of the industrial chain enables an immersive experience of the constructed environment. The application of intelligent interpretation technology uses digital techniques to create knowledge graphs of cultural heritage symbols, connecting their background, semantics, artistic features, and other attribute information. This offers structured knowledge for creative design, resulting in a comprehensive understanding of cultural symbols. Meanwhile, for the designated construction sites, intelligent interpretation technology uses digital twins, augmented reality (AR), and other technologies to achieve the digital representation of actual buildings.

Cultural heritage elements situated in Area II show high adaptability and low usability. Cultural heritage elements situated in Area IV show low adaptability and high usability. Such elements often excel in one or two specific indices, falling within the moderate-value potential area. They may possess high cultural value but face limitations in industrial transformation or contain a strong industrial foundation with relatively weak cultural significance. These heritage elements should adopt targeted development strategies so that cultural heritage elements can be reintegrated into suitable built environments.

Cultural heritage elements situated in Area III show low adaptability and usability. These elements are usually spread across various places and generally exhibit weak correlations across all four-dimensional indices, falling into the low-value reserve area. Heritage elements in this area are not prioritized in the current development plan; however, those of higher value should be selected to preserve potential for future value rediscovery.

3. Empirical Study: Cultural Heritage of Yiling

3.1. Research Area

The Yiling heritage area was selected as the validation case for this study due to its representative heritage structure and resource management context. The Yiling heritage area serves as a pivotal confluence of Ba-Chu culture and Central Plains culture. Its heritage values originate from a long-term, dynamic process of cultural integration.

This cultural integration is reflected in its folk traditions, settlement patterns, mythologies, and other forms of historical heritage. For example, the myth of “Huangniu Splits the Gorge” (HC₂₀₃) reflects ancient worship of natural forces and collective memory of reshaping the landscape, serving as a core element of regional identity and spiritual belief. The artistry and aesthetics of Yiling engraving (HC₄₁₃) blend the vigor of the Ba culture with the lyricism of the Chu culture. Moreover, this feature is also embodied in natural heritage resources, such as architectural remains, landforms, and tourism assets. The Huangling Temple (NC₁₀₃) dates from the Spring and Autumn Period and has undergone successive renovations. Its value lies in the history of stratification and the persistent spirit of flood control. Meanwhile, settlement relics such as the Bingzhai Fortress Complex (NC₁₀₆) reflect the integration of Ba-Chu culture in defense, habitation, and ecological adaptation through their distinctive architectural forms and spatial organization.

Those heritage elements reflect the cultural integration and resource-intensive features of the Yiling heritage area from multiple dimensions. However, these features have also led to challenges such as fragmented distribution, conflicting perceptions of symbol value, and a lack of systematic spatial intervention strategies. These issues make it a representative example of resource-intensive regions within the context of rapid urbanization. Therefore, given its representativeness in terms of both resource structure and current management context, this study takes Yiling District as a case to validate the model’s feasibility.

3.2. Data Collection

The cultural heritage resources of Yiling are scattered in distribution, with each element showing significant regional features. We systematically collected cultural heritage elements of Yiling using a multi-source data collection approach and integrated the perceptual demands of multi-stakeholder groups into the selection process. First, we compile data from fieldwork and demands from cultural workers to identify 71 unique elements of Yiling cultural heritage. Figure 4 and Figure 5 illustrate the inclusion of 34 elements of natural cultural heritage and 37 elements of historical cultural heritage.

Then, we selected 58 cultural heritage elements by taking the perceptual demands of residents and tourists as the identification conditions. At last, we eliminate heritage elements that are inconsistent with long-term development plans based on the demands of government workers. As shown in

Table 2. Statistical table of Yiling cultural heritage symbol elements.

Meta (Resource Type)	Cluster (Heritage Elements with Codes)
Natural Cultural Heritages (NC)
	NC101	NC102	NC103	NC104	NC106	NC109	NC201	NC202
	NC301	NC302	NC303	NC304	NC306	NC307	NC308	NC309
	NC310	NC401	NC402	NC403	NC406	NC407	NC409	NC412
Historical Cultural Heritages (HC)
	HC102	HC103	HC203	HC303	HC305	HC307	HC402	HC403
	HC404	HC405	HC406	HC409	HC413	HC414	HC415	HC416
	HC417	HC418

, 42 Yiling heritage elements meet public perception demands, which are the value assessment targets. In order to provide simpler solutions when dealing with complex relationships between data. The cleaned heritage elements are classified and encoded into natural cultural (NC) and historical cultural (HC). Each resource type contains many sub-elements, resulting in cultural element sets $NC\in \left\{{NC}_{1n}\right.,{NC}_{2n},\cdots \left.{NC}_{4n}\right\}$ and $HC\in \left\{{HC}_{1n}\right.,{HC}_{2n},\cdots \left.{HC}_{4n}\right\}$, thereby completing the organization and encoding of heritage elements. ${NC}_{11}$ represents the first cultural element in the architecture category of natural cultural heritage resources.

3.3. Value Assessment Results

The 42 heritage elements that meet public perception demands serve as evaluation objects, denoting the comparison sequence. The average value of the evaluation indices, EHL₀, serves as the reference sequence. We briefly describe all the elements of cultural heritage. Group A of experts is asked to score the 42 heritage elements in the comparable sequence and calculate the average for each index, as shown in

Table 3. Initial scores of cultural heritage elements (partial).

Index	EHL0	NC102	NC103	NC104	NC109	NC201	NC202	NC301	NC302	NC304	NC307
symbolizability	4.9	9.7	7.3	6.0	5.8	7.2	6.8	9.5	7.2	8.8	7.2
authenticity	4.5	9.3	9.2	8.2	8.5	8.2	8.7	9.8	8.7	8.8	8.3
readability	4.2	10	9.2	7.8	8.2	7.8	8.2	9.7	8.2	7.5	8.7
regionality	4.1	8.7	7.0	6.5	6.5	6.5	7.5	9.3	6.5	7.7	6.7
Index	NC310	NC402	NC403	NC406	NC412	HC103	HC203	HC303	HC305	HC413	HC415
symbolizability	8.3	8.0	7.2	6.2	7.2	8.2	8.2	6.3	9.3	9.7	5.8
authenticity	8.3	7.8	8.7	8.2	8.5	9.2	7.0	8.2	8.0	9.5	8.5
readability	7.7	8.2	8.5	7.8	8.5	9.2	6.8	7.8	8.0	9.0	8.0
regionality	8.3	6.8	7.2	6.3	7.0	7.8	7.0	6.0	7.5	9.2	6.3

. All four indices are positive indicators and exhibit a positive correlation with the comparative sequence. To ensure the recognizability and efficacy of cultural heritage symbols, heritage elements with an average value of 7 or less are eliminated.

To minimize disparities in absolute values and normalize them to a comparable range, the data in Table 3 were subjected to dimensionless processing in accordance with Equation (1). The computation was carried out using MATLAB, obtaining matrix I_i`:

$$\begin{array}{c}\hfill I\mathrm{i}`=\left[\begin{array}{ccccccccccc}1.1073& 1.0292& 0.8930& 0.8421& 0.8000& 0.9697& 0.8718& 1.1341& 0.9412& 1.0732& 1.9320\\ 1.0169& 0.9867& 1.1254& 1.1509& 1.1724& 1.1044& 1.1154& 0.9756& 1.1373& 1.0732& 1.0744\\ 0.9492& 1.0610& 1.1254& 1.0947& 1.1310& 1.0505& 1.0513& 0.9756& 1.0719& 0.9146& 1.1262\\ 0.9266& 0.9231& 0.8563& 0.9123& 0.8966& 0.8754& 0.9615& 0.9146& 0.8497& 0.9390& 0.8673\end{array}\right.\\ \hfill \left.\begin{array}{cccccccccccc}1.0184& 1.0390& 0.9114& 0.8702& 0.9231& 0.9535& 1.1310& 0.8905& 0.9922& 1.0374& 0.8112& 1.0675\\ 1.0184& 1.0130& 1.1013& 1.1509& 1.0897& 1.0698& 0.9655& 1.1590& 1.0235& 1.0160& 1.1888& 1.0184\\ 0.9448& 1.0649& 1.0759& 1.0947& 1.0897& 1.0698& 0.9379& 1.1025& 1.0131& 0.9626& 1.1189& 0.8957\\ 1.0184& 0.8831& 0.9114& 0.8842& 0.8974& 0.9070& 0.9655& 0.8481& 0.9713& 0.9840& 0.8811& 1.0184\end{array}\right]\end{array}$$

After standardizing the original data, we calculate the correlation degree between the comparison sequence and the reference sequence in each indicator according to Equation (2) and obtain the matrix ξi′:

$$\begin{array}{c}\hfill \mathsf{\xi }{\mathrm{i}}^{\prime }=\left[\begin{array}{ccccccccccc}0.6667& 0.4200& 0.3690& 0.3353& 0.5306& 0.3971& 0.8565& 0.4833& 0.8229& 0.4699& 0.6371\\ 0.8406& 0.5898& 0.5375& 0.5000& 0.6411& 0.6131& 0.7927& 0.5642& 0.7365& 0.7321& 0.9965\\ 0.5822& 0.4686& 0.5165& 0.4607& 0.6061& 0.6403& 0.8582& 0.5592& 0.8214& 0.4674& 0.9781\\ 0.9836& 0.6902& 0.9204& 0.8416& 0.7548& 0.8194& 0.9335& 0.6705& 0.9304& 0.7260& 0.6296\end{array}\right.\\ \hfill \left.\begin{array}{ccccccccccc}0.6961& 0.4421& 0.3955& 0.4574& 0.5026& 0.8715& 0.4172& 0.5750& 0.6914& 0.3436& 0.7987\\ 0.9806& 0.6495& 0.5375& 0.6826& 0.7486& 0.7537& 0.5227& 0.9648& 1.0000& 0.4748& 0.9965\\ 0.5737& 0.5511& 0.5165& 0.5253& 0.5636& 0.9374& 0.5035& 0.7105& 0.9251& 0.4780& 0.7463\\ 0.7842& 0.9156& 0.7886& 0.8457& 0.8922& 0.8024& 0.6658& 0.7791& 0.7323& 0.7764& 0.6296\end{array}\right]\end{array}$$

Formula (3) is used to evaluate the correlation between each heritage element and public perception demands as follows:

$$\begin{array}{c}\hfill ri=\left[\begin{array}{ccccccccccc}0.7683& 0.5422& 0.5859& 0.5344& 0.6332& 0.6085& 0.8602& 0.5693& 0.8278& 0.5989& 0.8103\end{array}\right.\\ \hfill \left.\begin{array}{ccccccccccc}0.7587& 0.6396& 0.5595& 0.6278& 0.6768& 0.8413& 0.5273& 0.7574& 0.8372& 0.5182& 0.7928\end{array}\right]\end{array}$$

The relational degree reflects the alignment between heritage elements and public perception demands, thus enabling optimal prioritization of heritage elements (Figure 6):

NC₃₀₁ > HC₂₀₃ > HC₄₁₃ > NC₃₀₄ > NC₃₁₀ > HC₄₁₈ > NC₁₀₂ > NC₄₀₂ > HC₃₀₅ > HC₁₀₃ > NC₄₀₃ > NC₂₀₁ > NC₄₁₂ > NC₂₀₂ > NC₃₀₇ > NC₁₀₄ > NC₃₀₂ > NC₄₀₆ > NC₁₀₃ > NC₁₀₉ > HC₃₀₃ > HC₄₁₅.

Finally, a scatter plot is generated with the semi-structured interview data from government workers to perform a consistency test on the correlation ranking. As shown in Figure 7, the bubble area grows with the increase in quadrant coordinate values, and the two data sets show a positive correlation. This confirms that the cultural cognition among government workers, the public, and expert groups is consistent. This finding verifies that the public perception-driven cultural heritage symbol value assessment can effectively resolve group cognitive differences.

3.4. Sustainable Design Solutions

We select the high-value heritage resources of Area I as elements of the built environment and illustrate an example of the application of spatial construction strategies. Elements in this area exhibit strong correlations across all four indices and align closely with the heritage development plan. It shows that the area is vital to cultural identity and spatial experience, necessitating a design strategy that prioritizes the preservation of authenticity and the exploration of symbolic value. In accordance with the differentiated construction strategy for tiered response outlined in Section 2.5.2, we have formulated a series of design schemes, including spatial atmosphere creation, industrial chain construction, and digital knowledge retrieval. The primary focus of this study remains on the assessment of heritage elements and distinct spatial construction strategies. Therefore, this section provides examples of the application of spatial construction strategies without elaborating on the specific symbol design process.

We use design techniques like symbol extraction and illustration design to transform heritage elements into visual forms. As shown in Figure 8, simplified heritage symbols are applied to building exteriors, contributing to an urban cultural atmosphere through the use of distinctive architectural materials and light-shadow techniques. Illustration design, which carries richer cultural connotations and attribute information, can improve public engagement and recognition when applied to interior spaces.

In addition, we are identifying shared attributes of cultural heritage in Area I to develop thematic cultural experience spaces. Figure 9 shows how we use spatial layout and design schemes to make the experience spaces more distinctive. We also design a series of creative and cultural products (CCPs) to enhance the immersive interactive experience. User interactions with these products, through photo-taking, social media sharing, and online commentary, amplified the cultural influence of cultural heritage. These participatory activities further expanded dissemination channels.

A primary reason for public cognitive bias in the digital era is deficient contextual cultural data. In the thematic cultural experience field, we employ the cultural gene map for the digital management and presentation of cultural heritage knowledge. As shown in Figure 10, we choose typical pictures from the collected dataset as image panels and combine relevant information to decode artistic features, background stories, and contemporary implications. Then, we extract cultural symbols from cultural imagery and image panels and deconstruct the symbols into individual symbol elements that contain cultural implications. Finally, pixelation methods are applied to analyze color images, and the Adobe Illustrator CC 2019 (version 23.1.0, Adobe Inc., San Jose, CA, USA) color picker is used to extract color data, thereby creating a cultural gene map. This map provides a comprehensive description of cultural heritage symbols by multi-dimensional feature linkages, allowing users a digital learning pathway and supplying abundant design elements and structured knowledge for urban renewal projects.

4. Discussion

4.1. Analysis of the Results

This study compares the quadrant distribution of heritage elements derived from semi-structured interviews with the priority ranking generated by grey relational degree calculation. The results demonstrate that the bubble size increases with higher quadrant coordinate values, and the two datasets show a significant correlation. This confirms that the cultural perceptions of government departments, the public, and expert groups are consistent. This finding indicates that heritage value assessment methods based on public perception can effectively mitigate cognitive biases among groups. This empirically supports the information processing theory [36], highlighting the importance of public participation in bridging the disjunction between the signifier and the signified in cultural symbols.

Some research has explored the decision-making framework of the built environment by integrating users’ ambiguous requirements. For instance, Ding constructed a group decision-making model using the IF-TODIM method [37], Hsueh created a fuzzy evaluation method for multi-attribute decision-making [38], and Li analyzed the aspects of festival attractiveness from the perspectives of organizers and tourists [39]. The study can simultaneously preserve cultural authenticity and improve design efficiency from the perspective of public perception integration. The essence is reducing individual cognitive biases by aligning users’ perceptual demands with built spatial elements.

During the data collection process, we observed that some residents possess a deep understanding of heritage elements. But owing to the intrinsic cultural value of the heritage, protective measures, and developmental plans, some heritage sites are incongruent with current urban development plans. These heritages elements are often assigned a lower value ranking and located in Area III or Area IV. Some heritage elements may be identified from them, and their common features can be explored. By establishing pertinent themed neighborhood sceneries and employing a series of promotional and marketing schemes, their developmental potential can be reshaped.

Compared to cultural workers, the general public tends to possess less knowledge of cultural heritage, which contributes significantly to public cognitive bias. When cultural symbols are presented in isolation, public interest remains relatively low. However, once heritage elements are accompanied by their historical narratives, the time of interaction significantly increases. Participants often begin to compare symbols, artifacts, and related cultural information, suggesting that hierarchical knowledge presentation may enhance user cognition. Li extracted symbols according to the cultural levels theory and verified the efficacy of hierarchical presentation in addressing cognitive biases [38]. Thus, a comprehensive and structured presentation of heritage is necessary to provide a systematic learning pathway.

4.2. Significance and Contribution of the Sandglass Tired Model

This study constructs a sandglass tiered model that effectively addresses the issue of cognitive spatial gradient dissolution. It mitigates value perception bias of heritage symbols between central transmission zones and periphery regions, as well as among tourists. Compared to traditional parallel group data analysis, this model employs a dynamic value assessment mechanism to account for variations in public perception in recognizing heritage value. This dynamic stems from the integration of real-time perceived demand data as assessment criteria. This mechanism enables the model to adapt to the spatial development of regional heritage. Case studies have confirmed its significant benefits in data logic, regional cognitive disparities, cultural authenticity, and policy execution.

The research identified four-dimensional evaluation indices: symbolizability, authenticity, readability, and regionality, providing quantitative standards for assessing the value of regional cultural heritage. These standards may be applied singly or in combination, depending on the specific situations. For example, regionality can serve as a scientific basis for delineating heritage corridors, providing data support for particular protection strategies. At the same time, the differentiated construction strategies for tiered response provide a direct decision-making tool for regional cultural heritage management. This form of spatial governance, based on the gradient of symbol value, effectively avoids the resource waste related to traditional homogenized protection and improves the efficiency of cultural heritage resource development.

4.3. Model Application and Transferability

The sandglass tiered model is transferable, and its framework provides a practical tool for assessing heritage elements’ value and managing resources in resource-intensive regions. The general model application procedure can be organized into four core processes:

• Contextual Adaptation and Subject-Object Definition. This involves clarifying the research objectives by defining application scenarios and core issues, thereafter identifying and recruiting pertinent stakeholders as perceptual subjects, and specifying the scope of heritage resources and spatial data for evaluation.
• Demand Integration and Data Collection. On the demand integration path, multi-source data collection methods are used to gather original perceptual data from multi-stakeholder groups, and the sandglass tiered framework is used for structured decoding to provide quantifiable demand indices. On the resource selection path, the spatial, morphological, and cultural attributes of heritage resources are systematically collected based on the perceptual demands, ensuring all elements possess evaluable properties.
• Value Assessment of Heritage Elements. This process translates the perceptual demands of diverse stakeholder groups into quantitative indices to establish reference sequences while treating various heritage elements as comparative sequences. The relationship between each cultural element and group perceptual demand is assessed to determine the valuation and ranking of elements.
• Differentiated Strategy for Tiered Response. Heritage elements are categorized into distinct priority areas for protection and utilization according to their correlation score and development planning. Then, differentiated strategies are formulated for each area based on the specific project contexts.

The transferability of the sandglass tiered model derives from its structured problem decomposition logic and flexible data input interface. Adhering to the above core process, the model can be adapted to the valuation and management of heritage resources across different regions and themes, offering a systematic solution for sustainable development projects.

5. Conclusions

This study constructs a sandglass tiered model that integrates public perception, quantifying the perceptual demands of multi-stakeholder groups into four evaluative dimensions: symbolization, authenticity, readability, and regionality. Heritage elements within the research region are assessed using these four indices, which results in the formulation of tiered response strategies that are based on overall correlation ranking and alignment with the development plan. This model provides a transferability pathway from value cognition to different decision-making in heritage management within resource-intensive regions.

Compared to traditional research on the evaluation of symbol value, this study incorporates the public’s ambiguous perceptual demands and transforms them into quantitative indices for evaluation. This method expands the research perspective while simultaneously improving the representativeness and consensus of value assessment using a fourfold division of roles. Additionally, we identify the value attributes of different heritage elements from the outcomes of value assessment and heritage development planning, thus enabling the formulation of tiered response heritage management strategies. The assessment results show that heritage elements in the high-value intensive area exhibit strong correlations across four value dimensions and development planning. Therefore, holistic protection and comprehensive interpretative strategies should be adopted. Heritage elements in the moderate-value potential area generally show one or two prominent indices among the four dimensions. These elements should focus on targeted development to unlock potential value. Heritage elements in low-value reserve areas generally show weak correlations across four-dimensional indices, but their presence is crucial to the overall landscape. It is advisable to use controlled development strategies to reserve space for future value development. Those differentiated strategies directly respond to the four-dimensional indices of the heritage elements, ensuring that management measures are closely aligned with perceived heritage values.

The sandglass tiered model is essentially a tool for element screening and decision recommendation suited to resource-intensive regions. It encompasses three aspects: spatial screening, value screening, and decision screening. Spatial screening identifies heritage elements that align with public perception through multi-source data collection, thereby mitigating cognitive biases resulting from regional perception gradients. Value screening selects high-value heritage elements using a four-dimensional evaluation index system, preventing resource waste caused by homogenized protection. Decision screening selects suitable strategies based on the element distribution quadrant, thereby improving the efficiency of resource allocation. Therefore, this model can serve as a routine decision-making tool, integrated into regular management practices such as periodic heritage surveys, project approval, and fund allocation. By regularly updating the value assessment of heritage elements, it enables real-time spatial and value-based screening of heritage resources, thereby providing data support for annual conservation and resource distribution plans. Furthermore, users can embed the differentiated strategies into the heritage management system to implement zoning-based precision management. For example, the system can automatically assign targeted development strategies based on the value area of heritage elements. For high-value areas (Area I), it triggers priority investment and activation procedures. For moderate-value areas (Areas II and IV), it initiates potential evaluation and targeted investment modules. For low-value areas (Area III), it executes controlled development focused on monitoring and maintenance. This data-driven automated decision-making significantly enhances the efficiency and scientific rigor of heritage management. Overall, through a standardized process, this study systematically integrates public perceptual demands into heritage management programs, providing a replicable technical pathway for precision management in resource-intensive regions. However, further improvements are needed to address current limitations through broader data collection, automated data acquisition, and more diverse data linkages to improve both the academic and practical value of this research. Furthermore, future research could explore the integration of digital twin technology with architectural symbol extraction, using digital methods to offer the public an accessible and detailed platform for visualizing and interpreting architectural heritage, thereby facilitating more effective knowledge acquisition and reuse of cultural building sites.