New Quality Productive Forces Enabling the Sustainable Development of Culture–Tourism Integration in China

Abstract

With the rapid advancement of information and communication technologies, new quality productive forces (NQPF) have emerged as a critical driver of the sustainable development of culture–tourism integration (SDCTI). Using panel data from 31 Chinese provinces spanning 2011–2023, this study constructs composite indices of NQPF and SDCTI based on the entropy method and the coupling coordination model, and empirically examines the impact and underlying mechanisms of NQPF on SDCTI. The results indicate that NQPF significantly promote SDCTI. Mechanism analysis shows that NQPF enhance SDCTI primarily by improving innovation capacity and stimulating entrepreneurial activity, with the mediating effect of entrepreneurship being more pronounced. In addition, NQPF contribute to SDCTI by alleviating labor misallocation, whereas the mediating effect of capital misallocation is not statistically significant. Heterogeneity analysis further reveals that the positive impact of NQPF is stronger in regions with higher levels of artificial intelligence development, lower tourism resource endowments, and weaker transportation infrastructure, highlighting both technological synergy and resource substitution effects. These findings remain robust after a series of endogeneity and robustness tests. Based on these findings, this study highlights the importance of promoting NQPF, strengthening innovation-oriented actors and technological applications, and optimizing labor allocation to fully unleash their enabling role in advancing SDCTI.

1. Introduction and Literature Review

Culture and tourism are inherently interconnected. Cultural landscapes, heritage sites, and events serve as key drivers of tourism, while tourism, in turn, facilitates the preservation, transmission, and revitalization of culture (Greg Richards, 2018) [1]. The expansion of cultural tourism, in both production and consumption, generates substantial economic and social benefits for destinations. However, unchecked tourism development may erode cultural authenticity, place pressure on local environments, and ultimately constrain the sustainable development of the tourism sector (Lu et al., 2023) [2]. New-quality productive forces (NQPF) represent an advanced form of productive capacity characterized by innovation-driven growth and a focus on efficiency, equity, and sustainability (Chen et al., 2025) [3]. By improving resource allocation efficiency, promoting industrial upgrading, and supporting environmentally friendly development [4], NQPF create new opportunities for the sustainable integration of culture and tourism. In particular, emerging digital technologies enable new forms of cultural dissemination, while digital platforms have fostered the development of smart tourism systems, significantly enhancing the value of cultural and tourism resources.

Nevertheless, the development of digital infrastructure also introduces a range of challenges, including environmental pressures, the over-commercialization of culture, and widening digital divides. Against this backdrop, a systematic examination of the impact of NQPF on the sustainable development of culture–tourism integration (SDCTI) is of considerable theoretical and practical importance.

NQPF represent an advanced form of productivity characterized by innovation-driven development and continuous industrial transformation and upgrading. They emphasize the accumulation of human capital while seeking to balance economic growth with resource and environmental sustainability [5,6,7]. In this regard, NQPF provide valuable insights into improving energy efficiency, mitigating climate pressures, and alleviating poverty on a global scale [8,9]. With respect to measurement, existing studies have adopted either single-indicator approaches [10] or composite index methods. The latter typically include entropy-based measures, entropy-weighted techniques such as TOPSIS, as well as productivity-oriented approaches like the Malmquist index and the SBM model [11,12,13]. These approaches, typically constructed around labor, means of production, and objects of labor, have improved the characterization of the spatiotemporal evolution of NQPF. However, they tend to emphasize static efficiency and are less capable of capturing the structural and systemic features inherent in NQPF. In terms of empirical effects, existing research has primarily focused on the agricultural, manufacturing, employment, and energy sectors, while relatively limited attention has been paid to the service sector particularly the culture and tourism domain. In agriculture, NQPF significantly promotes high-quality development and modernization, with stronger effects observed in less-developed and resource-constrained regions (Lin et al., 2024; Huang et al., 2024) [14,15]. In manufacturing, NQPF facilitate intelligent transformation, enhance industrial resilience, and act as a key mediating force in synergistic agglomeration (Liu and He, 2024; Cui, 2025) [11,16]. In the labor market, NQPF improve employment quality by increasing labor productivity (Shi, 2026) [17]. In the energy and environmental domain, NQPF drive green and low-carbon transitions, promote green development, and foster inclusive green growth (Zhang et al., 2024; Xu et al., 2024; Wang and Chen, 2024) [18,19,20]. Regarding mechanisms, prior studies have mainly examined channels such as technological innovation, industrial upgrading, economic development, and government intervention. However, relatively little attention has been given to the roles of entrepreneurial activity and factor allocation, which are crucial for translating technological advances into real economic outcomes. In terms of heterogeneity, existing analyses primarily focus on geographical or economic differences, while insufficient attention has been paid to variations in artificial intelligence infrastructure, tourism resource endowments, and transportation conditions.

From the perspective of culture–tourism integration, culture and tourism are fundamentally rooted in the human spiritual domain (Elavarasan et al., 2022) [21]. Cultural resources are preserved and revitalized through tourism, while tourism gains depth and identity through cultural embedding, forming a dynamic process of mutual reinforcement. SDCTI refers to a state in which cultural and tourism systems achieve coordinated development while continuously generating economic, social, and ecological benefits (Zhao et al., 2023) [22]. This contributes to community sustainability and more resilient tourism systems (Canavan, 2016; Mzembe et al., 2023) [23,24]. Existing studies have applied coupling–coordination models to evaluate eco-cultural tourism systems and identify key drivers of coordinated development (Lu et al., 2023) [2]. Meanwhile, the rapid advancement of information and communication technologies (ICT) has broadened access to tourism services and enhanced firms’ marketing capabilities (Ukpabi and Karjaluoto, 2016; Alkan et al., 2025) [25,26]. Emerging technologies are also found to reshape tourism systems through stages of opportunity, disruption, integration, and appropriation (Gössling, 2021) [27]. However, current research often treats NQPF, culture–tourism integration, and sustainability as a simple additive relationship, lacking a coherent theoretical framework to explain their complex interdependencies. In particular, empirical evidence on the direct impact of NQPF on SDCTI remains limited, and the underlying transmission mechanisms have not been sufficiently explored.

In summary, although existing studies have examined the concept, measurement, and impacts of NQPF, relatively limited attention has been paid to the service sector, particularly in the context of cultural tourism. Against this backdrop, this study makes several contributions. First, by integrating NQPF and the sustainable development of culture–tourism integration (SDCTI) into a unified analytical framework, this study extends the theoretical scope of NQPF research. Second, by applying Jenkins’ natural breakpoint decomposition method to examine the dynamic evolution of SDCTI, it moves beyond the predominantly static analyses in the existing literature. In addition, by constructing an evaluation index system that captures the emerging characteristics of NQPF from the perspectives of foundational support and technological penetration, this study enriches the measurement of labor, means of production, and objects of labor. Third, while prior research has paid limited attention to the mediating roles of factor allocation and entrepreneurial activity, this study investigates the mechanisms through which NQPF influences SDCTI across multiple dimensions, including labor allocation, capital allocation, innovation, and entrepreneurship. Finally, whereas most existing studies focus on geographic or economic heterogeneity, this study further explores heterogeneity from the perspective of NQPF and SDCTI characteristics, specifically examining differences in AI development levels, tourism resource endowments, and transportation infrastructure.

2. Theoretical Framework and Research Hypothesis

2.1. Direct Effect

Leveraging advanced digital and intelligent technologies, NQPF plays a crucial role across the production, dissemination, experience, and governance of cultural tourism. In the production stage, digital platforms and intelligent technologies facilitate the integration of culture and technology, enabling both the preservation and innovative revitalization of cultural heritage. In the dissemination stage, algorithm-driven platforms and social media have reshaped content creation and distribution, lowering information search costs and significantly expanding the reach and influence of cultural tourism products (Balakrishnan et al., 2023) [28]. On the experiential side, emerging technologies such as virtual reality (VR), augmented reality (AR), and blockchain foster new business models—such as immersive tourism and metaverse-based experiences—thereby enhancing visitor engagement and experiential quality (Cranmer et al., 2023) [29]. On the governance side, big data and Internet of Things technologies enable real-time monitoring and intelligent regulation of tourism markets, improving governance efficiency and sustainability. Moreover, the integration of online and offline consumption scenarios broadens market boundaries and enhances resource utilization efficiency (Zhao et al., 2023) [22]. Through these multidimensional transformations, NQPF improve production efficiency, optimize resource allocation, and enhance environmental performance, thereby strengthening the sustainable development capacity of culture–tourism integration.

Hypothesis 1.

NQPF positively promote SDCTI.

2.2. Mediating Mechanism: Innovation and Entrepreneurship

NQPF promote SDCTI not only directly but also indirectly through the dual channels of innovation capacity and entrepreneurial activity. First, NQPF drive advancements in frontier technologies such as artificial intelligence, big data, and blockchain, improving R&D efficiency and knowledge production, thereby enhancing urban innovation capacity (Liu and He, 2024) [11]. Enhanced innovation capability facilitates the efficient integration of culture, tourism, transportation, and information resources on digital platforms, enabling precise factor matching and improving resource allocation efficiency. It also helps identify structural bottlenecks within the culture–tourism value chain and promotes cross-regional industrial linkages (Ur Rahman et al., 2020) [30], thereby improving environmental performance and sustainability. Second, NQPF stimulate entrepreneurial activity by lowering market entry barriers through digital platforms, improving access to financing, and fostering innovation ecosystems (Xu and Yang, 2024) [31]. This reduces startup costs and risks, enabling small and medium-sized enterprises and innovative entities to participate more actively in the cultural and tourism market. Increased entrepreneurial activity diversifies product supply, enhances market vitality, and accelerates the formation of new business models, such as “cultural IP + tourism experience” and “digital platform + tourism services.” These dynamics promote the deep integration of culture and tourism while improving service quality and responsiveness to consumer demand.

Hypothesis 2.

NQPF promote SDCTI through enhancing innovation capacity and entrepreneurial activity.

2.3. Mediating Mechanism: Factor Allocation

NQPF further influence SDCTI through optimizing factor allocation and alleviating factor misallocation. The development of NQPF accelerates the digital and intelligent transformation of the culture and tourism industries, reshaping labor demand structures (Shi et al., 2026) [17]. It generates high-skilled, high-value-added employment opportunities in fields such as artificial intelligence and data analytics (Wei et al., 2026) [32], while also facilitating labor reallocation from low-productivity sectors to high-productivity sectors, thereby optimizing labor structure. From the perspective of production factors, new-quality labor—characterized by interdisciplinary skills—enhances decision-making efficiency based on data and market signals, reducing information asymmetry and improving service systems. New-quality means of production, including digital technologies and intelligent equipment, enhance the precision and flexibility of resource allocation (Rezaee et al., 2021) [33]. Meanwhile, new-quality objects of labor, such as data and digital content, transform culture and tourism resources from static consumption to dynamic value creation. In addition, platform-based and networked systems enable cross-regional allocation and sharing of resources, reducing spatial constraints and market segmentation (Gan et al., 2021) [34]. These processes alleviate capital and labor misallocation, improve resource utilization efficiency, and ultimately enhance the sustainability of culture–tourism integration.

Hypothesis 3.

NQPF promote SDCTI by alleviating factor misallocation.

2.4. Heterogeneity Mechanism

The impact of NQPF on SDCTI is likely to exhibit significant heterogeneity depending on technological, resource, and infrastructural conditions. First, as a general-purpose technology, artificial intelligence (AI) serves as a core carrier of NQPF. Higher levels of AI development enhance data processing capacity, algorithmic efficiency, and intelligent decision-making, thereby strengthening the enabling effect of NQPF on SDCTI (Tuo et al., 2025) [35]. Second, in regions with limited tourism resource endowments, NQPF can compensate for resource constraints through digital reconstruction and technological embedding. Technologies such as virtual reality and digital content creation enable the reconfiguration of cultural and tourism value, demonstrating a clear resource substitution effect (Balakrishnan et al., 2023) [28]. Third, NQPF reduce spatial constraints through digital platforms, online marketing, and smart tourism systems, enabling cross-regional dissemination and consumption of tourism products (Chon and Hao, 2025) [36]. This effect is particularly pronounced in regions with underdeveloped transportation infrastructure, where digital technologies can substitute for physical accessibility.

Hypothesis 4.

The effect of NQPF on SDCTI varies with the level of AI development.

Hypothesis 5.

The effect of NQPF on SDCTI varies with tourism resource endowments.

Hypothesis 6.

The effect of NQPF on SDCTI varies with transportation infrastructure.

Based on the above theoretical analysis, a conceptual diagram is presented to illustrate the relationship between the two (see Figure 1).

3. Methodology

3.1. Methods

3.1.1. Entropy Method

This study applies the entropy weight method (Han et al., 2023) [37] to evaluate the sustainable development levels of the culture and tourism sectors across 31 Chinese provinces, municipalities, and autonomous regions. The cultural and tourism sustainability indices are then computed using the following expressions:

$$A (x)={\displaystyle \sum _{i=1}^m{w}_j{M}_{ij}}$$

(1)

$$B (x)={\displaystyle \sum _{i=1}^m{w}_j{N}_{ij}}$$

(2)

In these equations, A(x) and B(x) represent the cultural and tourism sustainability indices, respectively; M_ij and N_ij are the standardized values of indicator j for year i; and w_j denotes the objectively derived entropy weight for indicator j.

3.1.2. Coupling–Coordination Degree Model (CCDM)

While the entropy method quantifies subsystem performance, it does not capture the extent to which culture and tourism develop interactively. Cultural and tourism subsystems can evolve asynchronously, potentially creating “pseudo-coordination” even when integration appears superficially strong. To evaluate both interaction intensity and coordination quality, we adopt the widely used coupling–coordination degree model (Liu et al., 2024) [38].

The coupling degree, measuring interaction strength, is expressed as:

$$C=\sqrt{A(x)\times B(y)/{[A(x)\times B(y)]}^2}$$

(3)

The comprehensive development level combining cultural and tourism systems is defined as:

$$T=\alpha A(x)+\beta B(y)$$

(4)

where α and β represent the relative importance of cultural and tourism systems. Following the principle of equal emphasis, we set α = β = 0.5.

Finally, the coupling–coordination degree is calculated as:

$$D=\sqrt{C\times T}$$

(5)

Values of D range from 0 to 1. Based on established thresholds, we classify D into ten levels to assess regional integration quality, identify spatial disparities, and track temporal progression (Sun and Cui, 2018) [39]: [0–0.1) indicates extreme imbalance; [0.1–0.2) signifies severe imbalance; [0.2–0.3) denotes moderate imbalance; [0.3–0.4) represents mild imbalance; [0.4–0.5) borderline imbalance, [0.5–0.6) barely balanced, [0.6–0.7) primary balance, [0.7–0.8) intermediate balance, [0.8–0.9) good balance, and [0.9–1] excellent balance.

3.1.3. Model Setting

To conduct the empirical analysis, this study first specifies Model (1) to examine the impact of NQPF on SDCTI.

$$Y_{i,t}={\alpha }_0+{\alpha }_{1}NQP{F}_{i,t}+{\alpha }_{2}Control{s}_{i,t}+{\mu }_{\mathrm{i}}+{\epsilon }_{i,t}$$

(6)

In this model, i denotes the individual unit and t represents time. Y is the dependent variable, referring to the sustainable development of culture–tourism integration (SDCTI). NQPF is the key explanatory variable, and Control represents a set of control variables. u_i captures individual fixed effects, and ε_i,t denotes the random error term. Standard errors are clustered at the individual level; the same approach is applied in subsequent models.

Second, Model (2) is constructed to examine the effect of NQPF on the mediating variables:

$$Medi{a}_{i,t}={\alpha }_0+{\alpha }_{1}NQP{F}_{i,t}+{\alpha }_{2}Control{s}_{i,t}+{\mu }_{\mathrm{i}}+{\epsilon }_{i,t}$$

(7)

where Media represents the mediating variables, including innovation capacity, entrepreneurial activity, capital allocation, and labor allocation.

Finally, Model (3) is specified to test the mediating effects in the relationship between NQPF and SDCTI. Both NQPF and the mediating variables are included in the regression to assess whether the mediators transmit the effect of NQPF on SDCTI.

$$Y_{i,t}={\alpha }_0+{\alpha }_{1}Medi{a}_{i,t}+{\alpha }_{2}NQP{F}_{i,t}+{\alpha }_{3}Control{s}_{i,t}+{\mu }_{\mathrm{i}}+{\epsilon }_{i,t}$$

(8)

3.2. Variables and the Data

Dependent Variable: SDCTI is measured using a comprehensive evaluation system incorporating five dimensions—innovation, coordination, green development, openness, and sharing—covering both cultural and tourism subsystems. The indicator system comprises 32 variables. After standardization and entropy weighting, subsystem indices for cultural sustainability (A) and tourism sustainability (B) are generated. These feed into the CCDM to compute the final coupling–coordination index representing SDCTI.

A detailed indicator system is listed in

Table 1. Indicators of SDCTI.

Type	Target Layer	Indicator	Indicator Explanation
Sustainable Cultural Development	Innovation	A1 Cultural Industry R&D Expenditure	Total social R&D expenditure × total output of above-designated-size cultural manufacturing enterprises/GDP
		A2 Cultural Industry R&D Personnel	Total social R&D personnel × cultural employees/total regional employment
		A3 Culture-related Patent Grants	Total number of granted patents in cultural and related industries
	Coordination	A4 Cultural and Entertainment Consumption	Per capita cultural and entertainment expenditure/per capita consumption expenditure
		A5 Advanced Structure of Cultural Industries	Operating revenue of cultural service enterprises/total operating revenue
		A6 Fixed-asset Investment in Cultural Industries	Fixed-asset investment in cultural and related industries/total fixed-asset investment
	Green	A7 SO2 Emissions per Unit of Cultural Development	SO2 emissions × total output of above-designated-size cultural manufacturing enterprises/industrial value added
		A8 Solid Waste Generation per Unit of Cultural Development	Solid waste generation × total output of above-designated-size cultural manufacturing enterprises/industrial value added
		A9 Total Wastewater Discharge	Total volume of wastewater discharged
		A10 Municipal Solid Waste Treatment Rate	Treatment rate of municipal solid waste
	Opening	A11 Participants in International Cultural Exchange Activities	Number of participants in international cultural exchange activities
		A12 International Cultural Exchange Programs	Number of international cultural exchange programs
	Sharing	A13 Per Capita Public Expenditure on Culture, Sports, and Media	Local government expenditure on culture, sports, and media/year-end permanent population
		A14 Number of Public Cultural Facilities	Total number of libraries, art galleries, museums, and cultural centers
		A15 Share of Cultural Employment	Number of cultural employees/total regional employment
		A16 Cultural and Tourism Public Service Expenditure	Cultural and tourism public service expenditure/year-end permanent population
Sustainable Tourism Development	Innovation	B1 Tourism R&D Expenditure	Total social R&D expenditure × total tourism revenue/GDP
		B2 Tourism R&D Personnel	Total social R&D personnel × tourism employees/total regional employment
		B3 Tourism-related Patent Applications	Number of patent applications related to tourism
	Coordination	B4 Industrial Coordination	Total tourism revenue/value added of the tertiary industry
		B5 Tourism Economic Coordination	Total tourism revenue/number of tourism employees
		B6 Urban–Rural Coordination	Urban–rural income gap measured using the Theil index
	Green	B7 SO2 Emissions per Unit of Tourism Development	SO2 emissions × total tourism revenue/industrial value added
		B8 Solid Waste Emissions per Unit of Tourism Development	Solid waste emissions × total tourism revenue/industrial value added
		B9 Forest Coverage	Regional forest coverage rate
		B10 Per Capita Park Green Area	Per capita public park green space area
	Opening	B11 Number of Inbound Tourists	Number of inbound tourist arrivals
		B12 International Tourism Foreign Exchange Earnings	Foreign exchange earnings from international tourism
	Sharing	B13 Tourism Fixed-asset Investment	Fixed-asset investment × total tourism revenue/GDP
		B14 Number of Travel Agencies	Total number of travel agencies
		B15 Number of Star-rated Hotels	Total number of star-rated hotels
		B16 Number of A-level and Above Scenic Spots	Number of scenic spots rated A-level or above

.

Independent Variable: The novelty of NQPF is reflected not only in the innovation and optimized allocation of new production factors—such as new labor, new means of production, and new objects of labor—but also in the emergence of new technologies, new industries, and new business models (Wang et al., 2025) [40]. Building on the framework proposed by Han Wenlong et al. (2024) [41], and drawing on the approaches of Li (2024) [7] and Liu and He (2024) [11], this study constructs a comprehensive NQPF index system using the entropy method, as shown in

Table 2. Indicators of NQPF.

Dimension	Component	Sub-Indicator	Basic Indicator	Description and Data Source	Note
Real Foundation	New Labor	Quantity of New Labor	Number of employees in emerging industries	Number of employees in listed companies of strategic emerging industries. Source: annual reports of listed companies	+
		Structure of New Labor	Educational structure of employees in emerging industries	Proportion of employees with a bachelor’s degree or above in listed companies of strategic emerging industries. Source: corporate annual reports	+
			Skill structure of employees in emerging industries	Proportion of employees in technical departments in listed companies of strategic emerging industries. Source: corporate annual reports	+
	New Means of Production	New Production Tools	Industrial robot penetration	Based on Acemoglu and Restrepo (2020) [42]. Source: IFR	+
			Integrated circuit output	Source: Ministry of Industry and Information Technology (MIIT)	+
		New Infrastructure	Length of optical fiber cable lines	Source: China Statistical Yearbook	+
			Broadband Internet access ports per capita	Source: MIIT	+
	New Objects of Labor	New Energy	Share of new energy generation	Ratio of new energy generation to total electricity generation (including hydro, nuclear, wind, solar, and other renewables). Source: National Bureau of Statistics (NBS)	+
			Number of ultra-high voltage transmission lines	Measures the capacity for new energy consumption. Compiled from official documents	+
			Efficiency of new energy utilization	GDP/new energy generation. Source: NBS	+
		New Materials	Output value of new materials industry	Operating revenue of listed companies related to new materials. Source: corporate annual reports	+
			Number of listed companies in new materials industry	Number of listed companies related to new materials. Source: corporate annual reports	+
Penetration Performance	New Technology	Technology R&D	R&D expenditure in high-tech industries	R&D expenditure of high-tech enterprises. Source: China Statistical Yearbook on Science and Technology	+
			Number of R&D institutions in high-tech industries	Number of R&D institutions in high-tech enterprises. Source: China Statistical Yearbook on Science and Technology	+
		Innovation Output	Patents in strategic emerging industries	Number of patents obtained by strategic emerging industries. Source: annual reports of listed companies	+
			Operating revenue of strategic emerging industries	Source: annual reports of listed companies	+
	New Industries	Intelligence	Number of Taobao villages	Source: China Taobao Village Research Report	+
			Number of AI enterprises	Source: Tianyancha database	+
		Green Development	Environmental intensity	Environmental protection expenditure/government public fiscal expenditure. Source: China Statistical Yearbook	+
			Green patent applications	Green patent applications/total patent applications of listed companies. Source: National Intellectual Property Administration	+
	New Model	Big Data Generation	Mobile Internet data traffic	Measures the scale of big data generation. Source: MIIT	+
		Big Data Trading	Number of data exchanges	Measures the scale of big data trading. Compiled from official documents	+
		Digitalization Level	Enterprise digitalization level	Measured using keyword frequency in annual reports of listed companies, aggregated at the provincial level	+

Note: “+” indicates a positive indicator.

. The index is developed along the practical foundation and the degree of penetration.

Control Variables: Building on prior studies (Pu et al., 2022; Lu et al., 2023) [2,42], this study includes a set of control variables. Economic development (lngdp) is measured as the logarithm of GDP per capita. Financial development (finance) is proxied by the ratio of total deposits and loans of financial institutions to regional GDP. The level of urbanization (urban) is measured by the proportion of the urban population to the total population. The degree of openness (open) is measured as the ratio of the total value of imports and exports (converted using the USD/CNY exchange rate) to regional GDP.

Mediating Variables: Innovation and entrepreneurship are selected as key mediating variables. Innovation capacity (innovation1) is measured using the Innovation Capacity Index, with data sourced from the Report on the Evaluation of Innovation Capacity in Chinese Regions. Entrepreneurial activity (innovation2) is proxied by the number of newly established enterprises per 100 people, with data obtained from the “Tianyancha” database.

In addition, resource misallocation is incorporated as a mediating variable. Following the framework of Hsieh and Klenow (2009) [43], and, drawing on the empirical approach of Bai Junhong and Liu Yuying (2018) [44], this study measures capital misallocation (capmis) and labor misallocation (labmis). These indicators are constructed as inverse measures, with higher values indicating a greater degree of misallocation.

3.3. Data and Sources

The study covers 31 provinces in China from 2011 to 2023. Data are drawn from: China Tourism Yearbook; China Culture and Tourism Yearbook; China Culture Yearbook; EPS statistical database; Provincial statistical yearbooks and bulletins. Missing data is handled using forward or backward imputation or linear interpolation. The results of the descriptive statistical analysis for each variable are presented in

Table 3. Descriptive Statistics of Variables.

Variable	Sample Size	Mean	Std. Dev.	Min	Max
SDCTI	403	0.278	0.074	0.148	0.626
culture	403	0.085	0.071	0.037	0.734
tourism	403	0.086	0.056	0.012	0.696
quality	403	0.276	0.132	0.072	0.747
lngdp	403	9.811	1.011	6.416	11.818
finance	403	3.538	1.141	1.688	8.164
urban	403	0.597	0.129	0.227	0.896
open	403	0.264	0.275	0.008	1.464
innovation1	388	1.509	0.964	−0.005	6.174
innovation2	372	28.804	10.695	15.780	65.490
capmis	390	0.279	0.177	0.001	0.733
labmis	390	0.395	0.388	0.002	2.716
IV1	403	12.877	0.923	10.001	15.899
IV2	403	16.955	1.433	13.001	20.577
IV3	341	34.653	33.448	7.360	218.733
A	403	347.548	235.315	32.000	1292.000
aifirm	403	21,692.740	37,784.800	50.000	341,254.000
traffic	403	0.005	0.005	0.001	0.034

.

4. Empirical Results

4.1. Spatiotemporal Evolution of SDCTI

Based on the results of the coupling coordination model, this study employs ArcGIS 10.3 (Esri, Redlands, CA, USA) to generate dynamic evolution maps of SDCTI for the years 2011, 2017, and 2023, using Jenkins’ natural breakpoint classification method (Figure 2).

4.1.1. Conditions in 2011: Low Coordination and Strong Spatial Clustering

In 2011, most provinces exhibited varying degrees of imbalance. Guangdong Province was close to a balanced state, while Zhejiang, Jiangsu, Beijing, Shanghai, Shandong, and Fujian showed only mild imbalance. In contrast, less developed regions, including Xinjiang, Tibet, Gansu, Ningxia, and western Qinghai, experienced severe imbalance. This pattern can largely be attributed to the relatively low levels of sustainability in both the cultural and tourism sectors in these regions, which resulted in weak agglomeration effects, limited synergy between the two sectors, and consequently low levels of SDCTI.

4.1.2. Conditions in 2017: Significant Improvement and Shrinking Imbalance

In 2017, SDCTI scores improved markedly across all provinces, with only Qinghai and Ningxia remaining in a state of severe imbalance. This can be largely attributed to their relatively limited cultural and tourism resource endowments, as well as the fragile ecological conditions in their arid regions, which constrain the coordinated development of the two sectors. In terms of regional transitions, Guangdong Province moved from being on the verge of imbalance to a barely coordinated state, while Jiangsu Province shifted from mild imbalance to the verge of imbalance. Zhejiang, Shandong, Beijing, Sichuan, Shanghai, Hunan, and Fujian all transitioned to a state of moderate imbalance. These changes are likely associated with the implementation of national strategies such as the “Beautiful China” initiative, the “Cultural Powerhouse” strategy, and the “All-for-One Tourism” policy, which have promoted the role of cultural development in enhancing tourism efficiency and, in turn, strengthened the coupling and coordination between the two sectors.

4.1.3. Conditions in 2023: Broad Advancement and Regional Breakthroughs

By 2023, all provinces had moved out of a state of severe imbalance. Guangdong Province advanced to an initial stage of coordination, while Zhejiang and Jiangsu remained close to the threshold of imbalance. These patterns suggest that SDCTI is closely associated with the level of regional economic development. Meanwhile, provinces such as Beijing, Shandong, Hubei, Yunnan, Sichuan, Hunan, Shanghai, Jiangxi, Fujian, Hebei, Anhui, and Henan transitioned to a state of mild imbalance, indicating a broadly shared trend across regions. From a spatial perspective, SDCTI across provinces shows a gradual movement toward equilibrium, although the extent of improvement remains relatively modest.

4.2. Baseline Regression Results

This study employs OLS to examine the impact of NQPF on SDCTI and its sub-dimensions. As reported in

Table 4. Benchmark Regression Results.

Variable	Coordination	Culture	Tourism
NQPF	0.320 ***	0.418 ***	0.202 ***
	(8.833)	(4.754)	(3.965)
lngdp	0.036 ***	0.004	0.026 ***
	(13.273)	(1.304)	(8.571)
finance	0.014 ***	0.013 ***	0.009 ***
	(6.879)	(3.462)	(7.657)
urban	−0.115 ***	−0.088 ***	−0.067 ***
	(−5.701)	(−3.940)	(−4.826)
open	0.005	−0.029	−0.007
	(0.413)	(−1.111)	(−0.666)
cons	−0.151 ***	−0.061 *	−0.220 ***
	(−6.337)	(−1.870)	(−10.405)
N	403	403	403
R2	0.810	0.487	0.610

Note: t-statistics for estimated coefficients are in parentheses; * p < 0.10, *** p < 0.01.

, NQPF significantly promote SDCTI, cultural sustainability, and tourism sustainability at the 1% significance level, with the strongest effect observed for cultural sustainability. These findings provide strong empirical support for Hypothesis 1. Regarding the control variables, the level of economic development (lngdp) has a significant positive impact on SDCTI and tourism sustainability, but its effect on cultural sustainability is not statistically significant, suggesting that cultural sustainability relies more on long-term historical accumulation rather than short-term economic growth. Financial development (finance) exerts a significant positive influence across all dimensions, highlighting the critical role of financial support in fostering sustainable development. Urbanization (urban) has a significantly negative effect on all dimensions, which may be attributed to the crowding-out of cultural and tourism resources during rapid urban expansion. In contrast, openness (open) does not exhibit a significant effect, indicating that the sustainable development of SDCTI is driven more by endogenous factors than by external openness.

4.3. Analysis of Transmission Mechanisms

4.3.1. Analysis of Innovation and Entrepreneurship

The baseline regression captures the overall relationship between NQPF and SDCTI. To further investigate the underlying mechanisms, this study constructs a mediation model incorporating two key dimensions—innovation and entrepreneurship, as well as resource allocation—to identify the transmission pathways. As reported in

Table 5. Mediation results: innovation capacity channels.

Variable	Innovation1	Coordination	Innovation2	Coordination
NQPF	2.792 ***	0.299 ***	49.654 ***	0.239 ***
	(3.552)	(8.366)	(11.629)	(6.007)
lngdp	−0.213 **	0.039 ***	1.019 ***	0.029 ***
	(−2.105)	(14.241)	(2.856)	(11.081)
finance	0.129 *	0.008 ***	0.934 ***	0.010 ***
	(1.906)	(4.044)	(3.190)	(5.002)
urban	6.263 ***	−0.092 ***	1.376	−0.107 ***
	(11.429)	(−4.295)	(0.644)	(−5.532)
open	−2.822 ***	0.016	10.435 ***	−0.018
	(−8.829)	(1.170)	(6.064)	(−1.314)
innovation1		0.008 ***
		(4.072)
innovation2				0.002 ***
				(3.853)
cons	−0.658	−0.182 ***	−1.852	−0.106 ***
	(−0.798)	(−7.685)	(−0.595)	(−4.821)
N	388	388	372	372
R2	0.317	0.827	0.867	0.833

Note: t-statistics for estimated coefficients are in parentheses; * p < 0.10, ** p < 0.05, *** p < 0.01.

, with respect to innovation capacity, NQPF significantly enhance the level of scientific and technological innovation. After introducing innovation capacity as a mediating variable, the estimated coefficient of NQPF remains significantly positive, although it declines from 0.320 in the baseline regression to 0.299, indicating a partial mediating effect. Similarly, in terms of entrepreneurial activity, NQPF exert a significantly positive impact on the number of innovative enterprises, suggesting that the development of NQPF effectively stimulates entrepreneurship. After including this variable in the model, the coefficient of NQPF remains significantly positive, but decreases from 0.320 to 0.239, further supporting the existence of a mediation effect. These findings provide strong empirical support for Hypothesis 2.

The bootstrap results provide additional evidence for these mechanisms. The indirect effect of innovation capacity is significantly positive, with a coefficient of 0.021, while the indirect effect of entrepreneurial activity reaches 0.105, which is substantially larger. This finding indicates that, compared with macro-level innovation capacity, micro-level innovation entities play a more prominent role in promoting SDCTI. Overall, NQPF facilitate the high-quality development of SDCTI by enhancing innovation capacity, fostering entrepreneurial activity, and improving the allocation and utilization of cultural and tourism resources, thereby enriching product supply and strengthening integrated development.

4.3.2. Analysis of Resource Allocation

As reported in

Table 6. Mediation results: resource allocation channels.

Variable	Capmis	Coordination	Labmis	Coordination
NQPF	0.629 ***	0.310 ***	−1.841 ***	0.286 ***
	(5.567)	(8.388)	(−9.677)	(7.409)
lngdp	−0.110 ***	0.039 ***	0.031	0.039 ***
	(−7.799)	(12.835)	(1.469)	(14.454)
finance	−0.005	0.010 ***	0.013	0.010 ***
	(−0.450)	(4.565)	(0.613)	(4.694)
urban	0.289 *	−0.049 **	0.146	−0.044 **
	(1.944)	(−2.499)	(0.674)	(−2.287)
open	−0.339 ***	−0.002	1.399 ***	0.018
	(−5.711)	(−0.131)	(8.490)	(1.011)
capmis		0.010
		(1.028)
labmis				−0.017 **
				(−2.542)
cons	1.118 ***	−0.201 ***	0.099	−0.188 ***
	(9.441)	(−7.244)	(0.458)	(−8.158)
N	390	390	390	390
R2	0.218	0.819	0.596	0.822

Note: t-statistics for estimated coefficients are in parentheses; * p < 0.10, ** p < 0.05, *** p < 0.01.

, NQPF exert a significantly negative effect on resource misallocation, indicating that they contribute to improving factor allocation efficiency. After incorporating resource misallocation as a mediating variable, the coefficient of NQPF remains significantly positive, although its magnitude is smaller than that in the baseline regression, suggesting a partial mediating effect. Specifically, NQPF have a significant impact on labor misallocation. After including labor misallocation in the model, the coefficient of NQPF remains significantly positive, but declines from 0.320 in the baseline regression to 0.286, further confirming the mediating role of labor allocation. These findings provide strong empirical support for Hypothesis 3.

The bootstrap results in

Table 7. Bootstrap results of mediation effects.

Mediator	Effect	Coef	t	95%CI	Proportion
innovation1	Indirect	0.021 ***	2.898	[0.007, 0.035]	6.6%
	Direct	0.299 ***	8.184	[0.227, 0.370]
	Total	0.320 ***	8.662	[0.247, 0.392]
innovation2	Indirect	0.105 ***	3.804	[0.051, 0.158]	30.6%
	Direct	0.239 ***	6.027	[0.161, 0.316]
	Total	0.343 ***	9.922	[0.275, 0.411]
capmis	Indirect	0.006	1.091	[−0.005, 0.018]	1.9%
	Direct	0.310 ***	8.209	[0.236, 0.384]
	Total	0.316 ***	8.594	[0.244, 0.388]
labmis	Indirect	0.030 **	2.530	[0.007, 0.054]	9.5%
	Direct	0.286 ***	7.414	[0.210, 0.361]
	Total	0.316 ***	8.594	[0.244, 0.388]

Note: ** p < 0.05, *** p < 0.01.

provide additional support for these findings. The indirect effect of capital misallocation is not statistically significant, whereas, the indirect effect of labor misallocation is 0.030 and significant at the 5% level. This indicates that NQPF primarily enhance SDCTI by improving the allocation of labor resources and increasing the efficiency of factor matching.

4.4. Heterogeneity Analysis

Given the heterogeneity across regions, this study employs a two-way fixed-effects model to examine whether the impact of NQPF on SDCTI varies under different regional conditions. The heterogeneity analysis is conducted along three dimensions. First, considering the close linkage between artificial intelligence (AI) development and NQPF, the sample is divided into high- and low-AI groups based on the overall median of the stock of AI enterprises across provinces and years. Second, tourism resource endowments provide the foundational conditions and application scenarios for value creation in SDCTI. Accordingly, the sample is split into high- and low-resource groups using the overall median of the average number of Grade A scenic spots in each province. Third, transportation infrastructure determines the accessibility and market reach of cultural and tourism products. It is measured by transportation network density—defined as the ratio of total highway and railway mileage to provincial administrative area. The sample is then divided into high- and low-infrastructure groups based on the overall median of this indicator.

As reported in

Table 8. Results of the heterogeneity analysis.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
	Low AI	High AI	Low Resources	High Resources	Low Transport	High Transport
NQPF	0.222 ***	0.287 ***	0.247 ***	0.205 ***	0.335 ***	0.074 *
	(6.316)	(8.651)	(7.401)	(5.048)	(9.243)	(1.836)
lngdp	0.028 ***	0.054 ***	0.025 ***	0.053 ***	0.041 ***	0.041 ***
	(8.416)	(9.387)	(7.779)	(8.860)	(7.209)	(15.493)
finance	0.014 ***	0.012 ***	0.011 ***	0.018 ***	0.005	0.015 ***
	(5.681)	(3.413)	(6.192)	(4.108)	(1.408)	(5.871)
urban	−0.069 ***	−0.233 ***	−0.101 ***	−0.098 **	−0.028	−0.092 ***
	(−3.491)	(−4.738)	(−5.084)	(−2.121)	(−0.556)	(−4.658)
Open	0.020	0.038 **	0.033 ***	0.088 ***	0.001	0.021
	(1.621)	(2.067)	(2.845)	(2.952)	(0.026)	(0.853)
cons	−0.079 ***	−0.239 ***	−0.036	−0.323 ***	−0.222 ***	−0.166 ***
	(−2.694)	(−4.067)	(−1.443)	(−5.820)	(−4.842)	(−6.564)
N	215	188	202	201	204	199
R2	0.698	0.817	0.813	0.819	0.794	0.701

Note: t-statistics for estimated coefficients are in parentheses; * p < 0.10, ** p < 0.05, *** p < 0.01.

, the effect of NQPF on SDCTI exhibits clear heterogeneity across regions.

In regions with stronger AI capabilities, the promoting effect of NQPF is more pronounced, with a coefficient of 0.287. This can be attributed to more advanced digital infrastructure and a higher degree of integration between big data, AI, and the culture–tourism sector. These conditions enable NQPF to be more effectively embedded in resource development, product innovation, and service optimization, thereby amplifying their enabling effects. In contrast, in regions with weaker AI foundations, limited digital infrastructure and fewer application scenarios constrain the translation of NQPF into tangible development momentum for SDCTI. In regions with lower tourism resource endowments, the effect of NQPF is also more significant, with a coefficient of 0.247, indicating a clear resource substitution effect. Resource-scarce regions have greater scope for technological embedding. Through digital technologies, NQPF facilitate the recreation and value restructuring of existing cultural and tourism resources, enabling these regions to overcome resource constraints and achieve innovative development. By contrast, regions with richer resource endowments rely more heavily on existing natural and cultural assets, leading to relatively lower dependence on technological inputs and diminishing marginal returns to NQPF. In regions with weaker transportation infrastructure, the impact of NQPF is strongest, with a coefficient of 0.335. In such regions, inadequate physical accessibility constrains tourism development. However, digital platforms, online marketing, and smart service systems enabled by NQPF help overcome spatial limitations, reduce information asymmetry, and lower market entry barriers, thereby facilitating cross-regional dissemination and consumption. Conversely, in regions with well-developed transportation networks, physical infrastructure already supports efficient tourism flows and market connectivity, reducing the relative importance of digital substitution and weakening the marginal effect of NQPF. These findings provide strong empirical support for Hypotheses 4–6.

4.5. Tests of Robustness and Endogeneity

4.5.1. Endogeneity Tests

This study conducts a series of robustness checks by lagging the core explanatory variable, trimming the sample, and reconstructing the NQPF index within a two-way fixed-effects framework.

Table 9. Robustness test: adjusted sample interval.

Variable	(1) Baseline	(2) Lagged	(3) TWFE
L.NQPF	0.362 ***
	(8.923)
lngdp	0.033 ***	0.037 ***	0.038 *
	(11.216)	(13.334)	(1.808)
finance	0.013 ***	0.014 ***	0.005
	(6.298)	(6.894)	(0.979)
urben	−0.117 ***	−0.114 ***	0.359 ***
	(−5.625)	(−5.714)	(3.318)
open	−0.003	0.005	−0.038
	(−0.193)	(0.396)	(−1.422)
quality_w		0.321 ***
		(8.712)
newp			0.127 *
			(1.822)
Constant	−0.117 ***	−0.153 ***	−0.337
	(−4.787)	(−6.404)	(−1.624)
N	372	403	390
R2	0.821	0.808	0.953

Note: t-statistics for estimated coefficients are in parentheses; * p < 0.10, *** p < 0.01.

shows that the direction and statistical significance of the core explanatory variable remain consistent with the baseline findings, thereby confirming the robustness and reliability of the empirical results.

4.5.2. Robustness Checks

To address potential endogeneity arising from reverse causality and omitted variables, this study constructs a set of instrumental variables following the approach of Nunn et al. (2014) [45]. Specifically, three instruments are employed: (i) the logarithm of the interaction term between the number of fixed-line telephones per 100 people in 1984 and the number of Internet users in the previous year (IV1); (ii) the logarithm of the interaction term between the number of post offices per million people in 1984 and the number of Internet users in the previous year (IV2); and (iii) the first-order lag of provincial-level nighttime light intensity (IV3). In addition, lagged terms of the core explanatory variable (NQPF) are incorporated into the model, and a system GMM estimator is employed to further mitigate endogeneity concerns. Data are obtained from the Compilation of Statistical Data on New China’s First Fifty Years and nighttime light datasets (NPP-VIIRS and DMSP-OLS) provided by the U.S. National Oceanic and Atmospheric Administration (NOAA). The validity of the instrumental variables is supported by both theoretical and empirical considerations. Regions with stronger historical telecommunications infrastructure and faster recent internet diffusion tend to exhibit higher levels of NQPF, ensuring the relevance condition. Meanwhile, traditional telecommunications infrastructure has gradually declined in importance with economic development, reducing its direct influence on current SDCTI and thereby supporting the exclusion restriction. Moreover, the interaction terms generate panel instruments that vary across both regions and time, satisfying the requirements of panel data estimation. Regarding IV3, higher nighttime light intensity reflects a higher level of regional economic activity, which is closely associated with NQPF. At the same time, nighttime light intensity does not directly determine SDCTI. By introducing its first-order lag into the model, this study further strengthens the exogeneity of the instrument.

According to

Table 10. Endogeneity test: instrumental variables.

Variable	Quality (IV1)	Quality (IV2)	Quality (IV3)
L.SDCTI	0.634 ***	0.599 ***	0.759 ***
	(2.928)	(2.733)	(6.494)
NQPF	0.163 ***	0.170 ***	0.121 ***
	(3.053)	(3.506)	(3.189)
lngdp	0.006	0.012	0.017 *
	(0.727)	(1.573)	(1.694)
finance	0.004	0.011 ***	0.019 **
	(0.972)	(2.614)	(2.472)
urban	−0.146	−0.184	−0.134
	(−1.251)	(−1.467)	(−1.485)
open	0.056	0.046	−0.035
	(0.774)	(0.650)	(−1.029)
AR(1)	−2.94 ***	−2.92 ***	−3.06 ***
AR(2)	1.48	1.43	−0.40
Hansen test P	0.259	0.313	0.366
cons	0.047	−0.009	−0.135
	(0.922)	(−0.157)	(−1.548)
N	372	372	310

Note: t-statistics for estimated coefficients are in parentheses; * p < 0.10, ** p < 0.05, *** p < 0.01.

, From a dynamic perspective, the coefficient of the one-period lagged term of SDCTI is significantly positive, indicating strong path dependence in the integrated development of culture and tourism. In other words, the current level of SDCTI exerts a persistent influence on its future development. The diagnostic tests further support the validity of the model. The AR(1) test is statistically significant, while the AR(2) test is not, suggesting the absence of second-order serial correlation. In addition, the Hansen test statistic falls within the acceptable range of 0.1–0.9, indicating that the instrumental variables satisfy the exogeneity condition and are jointly valid. and, the sign and statistical significance of the core explanatory variable (NQPF) remain consistent with the baseline results. This indicates that, even after controlling for bidirectional causal relationships among the composite variables, the NQPF still has a significant positive effect on the SDCTI, confirming the robustness and reliability of the findings in this study.

5. Conclusions and Discussion

5.1. Conclusions

Using panel data for 31 provinces and municipalities in China from 2011 to 2023, this study examined the relationship between NQPF and SDCTI by employing the entropy method, a coupling coordination index, and system GMM estimation. The main findings are as follows. (1) SDCTI exhibits a steady upward trend across regions, gradually shifting from imbalance toward a more coordinated state. (2) NQPF has a significant positive effect on SDCTI. (3) Innovation, entrepreneurship, and labor mismatch act as important mediating channels through which NQPF influences SDCTI, whereas the mediating effect of capital mismatch is not statistically significant. (4) The impact of NQPF is more pronounced in regions with higher levels of artificial intelligence, lower tourism resource endowments, and weaker transportation infrastructure, suggesting the presence of technological synergy and resource transportation substitution effects.

5.2. Study Limitations

This study extends the application of NQPF to the cultural tourism service sector, thereby broadening the scope of existing research. However, several limitations should be acknowledged. The use of macro level data may obscure important micro level dynamics and heterogeneity. In addition, due to data availability constraints, the constructed indicators for NQPF may not fully capture its emerging dimensions, while the measurement of SDCTI may not adequately reflect intangible outputs, such as cultural value and visitor experience. Furthermore, the use of composite indicators to measure both NQPF and SDCTI may limit the ability to clearly identify the causal relationship between them. Although this study attempts to mitigate potential bidirectional causality through the instrumental variable approach, there remains room for further improvement in causal inference.

Future research could strengthen causal identification by exploiting exogenous policy shocks and adopting quasi-experimental approaches, such as event study designs or difference-in-differences methods. In addition, the use of micro level data at the city, firm, or scenic area level, combined with field surveys, would help reduce macro level aggregation bias and enable a more nuanced exploration of the micro level mechanisms through which NQPF influence the SDCTI.

5.3. Policy Implications

Given the significant role of NQPF in promoting SDCTI, policy design should be aligned with the identified mechanisms and heterogeneity patterns.

At the national level, governments should increase investment in digital technologies such as artificial intelligence, big data, and smart service systems within the cultural and tourism sectors. By leveraging digital platforms to integrate cultural resources with tourism products, a development pathway can be established that transforms cultural assets into tourism experiences and market value, thereby enhancing the overall synergy of SDCTI. At the cultural subsystem level, emphasis should be placed on promoting the digital transformation and dissemination of cultural resources, supporting the development of cultural and creative industries, and enhancing the innovation capacity and market appeal of cultural products, thereby achieving a dual improvement in cultural and economic value. At the tourism subsystem level, priority should be given to accelerating the development of smart tourism systems, upgrading digital management and service capabilities of scenic areas, and improving operational efficiency, so as to enhance the sustainability and competitiveness of the tourism sector.

From a mechanistic perspective, innovation and entrepreneurship serve as key mediating channels through which NQPF influences SDCTI, with entrepreneurial activity exhibiting a particularly strong mediating effect. Accordingly, enhancing R&D collaboration and fostering innovation are essential for advancing SDCTI. This requires promoting the development of frontier fields such as urban artificial intelligence, blockchain, and virtual reality, thereby providing sustained technological support for cultural and tourism enterprises. In addition, greater emphasis should be placed on entrepreneurial activity within the cultural and tourism sector. Policy instruments such as tax incentives and start-up subsidies can encourage firms to adopt digital technologies, innovate marketing models, and expand market reach. Given the significant mediating role of labor mismatch, it is also important to improve cross-regional talent mobility, strengthen workforce training in the cultural and tourism industries, and better align skills with market demand in order to alleviate labor misallocation. By contrast, as the mediating role of capital allocation is not statistically significant, policy design should avoid an excessive reliance on financial expansion alone.

Taking into account regional heterogeneity in NQPF, differentiated policy approaches are required. In regions with high levels of artificial intelligence, efforts should focus on deepening the integration of AI into the cultural and tourism industries, promoting digital sharing of resources, and enhancing technology-enabled user experiences. In contrast, regions with lower levels of AI development should prioritize the construction of digital infrastructure, including 5G networks, big data centers, and cloud computing platforms, while leveraging interregional cooperation and technical support to accelerate digital transformation. Furthermore, considering the resource and transportation substitution effects of NQPF, policy priorities should vary accordingly. In resource-scarce regions, emphasis should be placed on developing innovative tourism services, such as smart guided tours and virtual experiences, and on expanding the dissemination of cultural tourism products through digital platforms. In resource-rich regions, efforts should focus on upgrading the quality of cultural and tourism products and services, aligning them with evolving consumption patterns, and offering more personalized experiences. In areas with underdeveloped transportation infrastructure, big data and artificial intelligence can be leveraged to promote local tourism offerings and improve resource allocation. Conversely, in regions with well-developed transport systems, data platforms should be used to optimize visitor flows and service experiences, facilitating a shift from scale-driven expansion to quality-oriented development.