A New Framework for Evaluating City–Industry Integration in New Urban Districts: The Case of Xixian New Area, China

Abstract

Assessing city–industry integration levels is a critical diagnostic approach for promoting sustainable urban development. However, existing evaluation frameworks are mainly based on overlaying the level of development of individual systems and rely on statistical data, lacking analysis of spatial attributes. This study addresses these gaps by constructing an “industry–city–population” (I–C–P) evaluation system based on the interaction mechanisms among industry (I), city (C), and population (P), viewed through the lens of spatial correlation. Focusing on Xixian New Area and using 2022 sectional data, the study applies the CRITIC method to calculate the overall level of city–industry integration and the interaction levels across different dimensions in the district, and the Entropy Method (EM) is used to validate the results. The findings indicate the following: (1) The overall level of city–industry integration in Xixian New Area remains relatively low, with Fengdong and Fengxi significantly outperforming the other three new cities. (2) The interactions between “P–I” and “C–P” exhibit lower levels compared to the “I–C” interactions. Additionally, the spatial characteristics of the dimensional levels reveal both variability and consistency. The integrated indicator system, incorporating both spatial big data and traditional statistical data, significantly expands the data sources and dimensions for evaluating city–industry integration, which helps to provide a reference for the assessment of the potential for high-quality sustainable development in the new district and other regions.

1. Introduction

During the phase of rapid urbanization, driven by objective demand and subjective land financial dividends, numerous new cities and districts (hereinafter referred to as “new districts”) have emerged. These new districts play a crucial role in optimizing spatial structures, alleviating environmental pressures, and coordinating urban functions [1,2,3]. However, rapid spatial expansion has given rise to challenges such as low land utilization efficiency, inadequate infrastructure, and misaligned urban functions [4]. On the one hand, under the influence of differential land rent, substantial amounts of industrial land have been relocated to new urban districts. Due to the agglomeration effects of production factors, the reconcentration of employment tends to outpace the reconcentration of residential areas [5]. Meanwhile, land in central cities is predominantly converted into residential use, resulting in spatial mismatches between employment and housing. This misalignment has resulted in urban problems, including housing vacancies, traffic congestion, and air pollution [5,6]. On the other hand, the internal functionality of new districts tends to be relatively limited, with service quality falling short of that in central cities due to constraints imposed by the threshold scale of facilities. This exacerbates the separation between jobs and housing [5,7], leading to a “city–industry separation” dilemma. As a result, significant economic and social costs are incurred, resources and energy are wasted, and the high-quality development of urbanization is severely hindered, ultimately constraining the vitality of sustainable urban development. The concept of “city–industry integration” is vital to the development of new urbanization, offering a direction for addressing the challenges of city–industry separation (The National New Urbanization Plan (2014–2020)). Recognized as a critical issue in the urbanization process, city–industry integration has garnered increasing attention from the government and academia. Over time, it has evolved into a key metric for evaluating the comprehensive integration of cities, population, and industries within the framework of new urbanization [8]. Currently, urbanization is transitioning from a phase of rapid growth to one focused on improving quality [9]. Urban development is also shifting from incremental expansion to stock renewal. In this context, promoting city–industry integration and enhancing the quality of urban development have emerged as key priorities. A rigorous scientific evaluation of the current level of city–industry integration is essential for advancing high-quality urban development, identifying deficiencies in urban development, addressing uncertainties in urban progression, and underpinning the formulation of high-quality, sustainable development policies and plans for cities and regions.

Academia has engaged in extensive discussions about city–industry integration, encompassing conceptual connotation, evaluation systems, evaluation methods, etc. However, the concept of “city–industry integration” remains ambiguously defined, with significant variations in interpretation among scholars. Zhang, the first scholar to propose this concept, believed that the essence of “two-way integration” between industry and cities lies in attaining “balance” [10]. The interpretations of city–industry integration have become increasingly diversified with the advancement of the urbanization process and evolution of the city–industry relationship. The main perspectives include the following: (1) City–industry integration represents the fusion of industrialization and urbanization [11,12]. (2) It indicates the integration of industrial parks and urban areas, essentially aligning residential and employment areas [13,14]. (3) The key to achieving city–industry integration lies in the coordinated development of industry, city, and population [15,16,17]. (4) City–industry integration is a multidimensional and complex system requiring synergy and mutual promotion among various elements, including industry, city, population, land, and transportation [18,19].

The inconsistent understanding of city–industry integration has made it challenging for scholars to reach a consensus on the selection of evaluation dimensions and indicators. Existing evaluation dimensions can be broadly categorized as follows:

(1) Two dimensions encompassing “Industrialization–Urbanization” [20,21,22] and “Industrial function–Urban function” [20,23,24].

(2) Three dimensions such as “Population–Industry–City” [16,18,25,26,27], “Humanistic–Function–Spatial” [28] and “Urbanization–Industrialization–Ecological condition” [29].

(3) Multiple dimensions include Industrial economy, Population, Employment, Income, Land, and Social development [30].

It can be found that the existing evaluation system primarily focuses on the three aspects of industry, city, and population with the development of new urbanization. However, indicators are selected to reflect only the independent development level of these dimensions, and most indicators are less representative, limiting the reflection of the interaction mechanisms between dimensions, the spatial attributes of elements, and the spatial correlation between industry and city. With regard to data type, most studies have utilized socio-economic statistics from statistical yearbooks. To address the limitation of a single evaluation system and the poor timeliness and weak representativeness of the statistical yearbook data, some scholars have employed multisource geospatial data, including mobile phone signaling, Baidu heat map, points of interest (POI), and other big data [31,32], as well as subjective data such as questionnaires [33]. Therefore, aimed at the above limitations, this study constructs the “I–C–P” evaluation framework, with the following contributions: (1) Compared with evaluating the level of industry, city, and population separately, the framework evaluates the interaction level of “industry and city”, “population and industry”, and “city and population”, respectively, based on the interaction mechanism among industry, city and population, and the evaluation dimensions are more in line with the connotation of “integration” rather than simple combination. (2) Compared with simple socio-economic indicators, the evaluation indicators selected based on each dimension can more accurately reflect the spatial characteristics and the “interaction” of each system. Furthermore, existing studies largely focus on national scales [18,34], urban agglomerations [27,35], provinces [20,25], cities [29,36], and industrial parks [32,37], and insufficient focus has been placed on new districts, which are distinct, as they are both connected to and relatively independent from the central cities. The phenomenon of city–industry separation is more pronounced in the western region due to disparities in regional resources and economic development. Thus, there exists a need for evaluation research on city–industry integration, utilizing the new districts in the western region as a representative case study [38].

Commonly used evaluation methods for city–industry integration include factor cluster analysis [12], the fuzzy analytic hierarchy process [39], grey relational analysis [40], combination weighting and the four-quadrant method [41], principal component analysis [27], the entropy method and coupling coordination model [42], and the analytic hierarchy process [43]. This study employs the CRITIC method proposed by Diakoulaki for objective weight assignment and composite index calculation [44,45,46]. Unlike conventional objective weighting approaches that typically consider either information entropy or indicator independence in isolation, the CRITIC method systematically integrates both dimensions through combining standard deviation measurements and correlation matrix analysis, which ensures mathematically robust weight allocation that effectively prevents subjective bias.

The expansionary incremental development of new districts conflicts with the current urban inventory planning stage in terms of spatial development [47]. City–industry integration offers a promising approach to balancing this contradiction, playing a crucial role in mitigating spatial development conflicts, optimizing urban spatial layouts, and fostering the organic integration of new districts with main cities. A scientific evaluation of city–industry integration is essential for achieving sustainable development in this area. Building on previous research, this study constructs the industry–city–population (I–C–P) evaluation system from the perspective of spatial correlation. Combining multisource geospatial data and statistics, the study comprehensively assesses the level of city–industry integration and analyzes the spatial characteristics of the levels of each dimension, offering innovative insights for optimizing evaluation systems and providing theoretical references to address the challenges of city–industry separation in the development of new districts.

2. Theoretical Framework

2.1. “I–C–P” Interaction Theory

First, new urbanization prioritizes humanistic orientation, recognizing that population is the central element of industry and city development. As such, I–C–P integration should be the primary focus of the research on city–industry integration in the context of new urbanization [48]. Second, the ultimate objective of city–industry integration is to achieve a balance of social, economic, and ecological benefits. Solely relying on industrial development or urban construction often results in short-term efficiency gains, making sustained long-term development challenging. Only through the organic interaction of the systems of industry, city, and population can new districts achieve high-quality, sustainable development. Accordingly, this study proposes a theoretical framework for “I–C–P” interaction (Figure 1):

Industry–city (I–C) interaction: Industrial development offers economic support for city construction. The city serves as the material carrier and guarantor of industrial activities by offering a spatial environment and functional services for various industries. From the perspective of city–industry integration, the organic alignment of industrial development and urban functions is essential for the transformation and upgrading of industries. This requires the comprehensive development of functional and spatial integration, ultimately aiming to maximize economic benefits [49].

Population–industry (P–I) interaction: The population, as the primary productive force, provides labor and intellectual capital for industrial development. The concentrated development of various industries generates abundant job opportunities for the workforce. The organic interaction between population and industry fosters the transformation of the industrial structure, further enhancing employment structure and labor income. From the perspective of city–industry integration, the core of population–industry interaction lies in aligning employment structure with the population structure. The real integrated development of city and industry can be promoted only when industrial, employment, and consumption structures align with each other [21,49].

City–population (C–P) interaction: A sound urban service system attracts talents with diversified services, while the demands of different population groups drive the creation of varied urban functions. The spatial mismatch between supply and demand may result in low functional utilization and limited population agglomeration. Therefore, the interaction between city and population in new districts depends not only on balancing the quantity of supply and demand but also on ensuring spatial adaptation and livability quality.

In summary, the evaluation system for city–industry integration should extend beyond macro-level assessments of urbanization and industrialization coordination or micro-level evaluations of industrial development and functional construction in industrial parks. In other words, it must reflect the combined I–C–P interaction. Specifically, “I–C” interaction focuses on spatial connections, functional integration, and economic benefits; “P–I” interaction is mainly reflected in the population, industrial, consumption, and job-housing structures; “C–P” interaction is mainly characterized by population vitality and seeks to achieve a balance between urban supply and population demand in terms of quantity, quality, and space. All dimensions interact organically, ultimately driving the high-quality sustainable development of industry, city, and population.

2.2. Evaluation System of City–Industry Integration

Based on the theoretical foundation of “I–C–P” interaction and addressing the shortcomings of the existing evaluation system, this study establishes an evaluation framework for city–industry integration through three dimensions: “I–C”, “P–I”, and “C–P”.

(1) “I–C” interaction: Regarding the industry–city interaction, spatial connection is fundamental, functional integration forms the cornerstone, and economic benefits serve as the direct expression of its effectiveness. The development of new districts often relies on the radiation-driven influence of the central cities on which they are based, including industrial agglomeration in adjacent industrial parks, population spillover from the central city, and shared functional services. It is necessary to subdivide the spatial connection into “spatial connection between new district and central cities” and “spatial connection within the new district”. To represent the connection between the new district and the central city, spatial distance and the number of bus routes are utilized [50], thus capturing the new district’s spatial connectivity from aspects of static geographical layout and dynamic traffic linkages. Specifically, spatial distance can mirror the population’s commuting burden as well as the agglomeration degree of industrial and service resources. The number of bus routes directly influences the population’s inter-regional commuting efficiency and can also reflect the effectiveness of the TOD model implementation in Xixian New Area. The coverage of bus stations within 500 m, subway stations within 800 m [33], and road network density are selected to represent the convenience of spatial connectivity within the new district. On the one hand, these three indicators are commonly used to evaluate spatial integration. On the other hand, the coverage rate of public transportation focuses on reflecting accessibility, the road network density can reflect the convenience of various types of traffic connection, and the combination of these three indicators can fully reflect the spatial connection degree within the new district. Functional integration captures the level of integration between production and living functions [51]. Conventional assessments of functional development have predominantly relied on statistical data such as the number of hospital beds and teachers, which inadequately capture spatial multifunctionality and suffer from poor data timeliness. Functional mixing degree is extensively applied to evaluate the spatial characteristics and land use degree [52,53]. Employing spatial big data can more directly depict the integration of diverse functions. Additionally, this methodology particularly aligns with the “compact town” development concept in Xixian New Area. Moreover, to comprehensively measure the economic benefits, this study incorporates not only commonly used statistical data such as regional output value per unit area [18] but also spatial economic vitality, which uses nighttime light remote sensing data to reflect the spatial attributes of economic development [6].

(2) “P–I” interaction: Population, employment, and consumption structures are the key elements of population–industry interaction. The job–housing spatial relationship influences population employment, and innovation potential is crucial for city–industry integration efficiency. Considering the job–housing balance, income and consumption levels, and innovation development capacity as the primary indicators of “I–C” interaction can adequately reflect the population–industry interaction. Using the ratio of the number of jobs to the number of dwellings in the region to measure the level of job–housing balance (JHB) overlooks individual choices. In this study, we use the job–housing balance index to approximate the distribution trends of individuals in geospatial space using big data [54,55]. Furthermore, the ratio of the growth rate of total retail sales of consumer goods to the GDP growth rate is used to measure the structural balance between industrial production growth and population consumption in the new districts. Total retail sales of consumer goods reflect the population’s purchasing power and consumption levels, while the GDP growth rate reflects the speed of total economic development and the vitality of the regional economy [56]. Considering that Xixian New Area, as a new district in the western region, has a relatively weak development foundation compared with the indicators reflecting the total economic volume, this indicator can better eliminate the impact of regional scale differences and truly identify the strengths and weaknesses of the income and consumption structure. In addition to the labor force, the population also offers intellectual and technical support for the industrial development of new districts and fosters innovation for the efficient development of the new districts. Meanwhile, high-tech enterprises serve as economic entities driving industrial transformation and development but also as key providers of knowledge-based and technical positions for highly educated talents. Consequently, this study uses the ratio of the population with higher education to the total employed population, along with the annual incremental volume of the high-tech enterprises, to reflect innovative development capacity [24]. These two indicators can reflect not only the cultivation ability of talents but also the policy effect and innovative growth potential of Xixian New Area under the concept of innovative development.

(3) “C–P” interaction: Population agglomeration is a key driver of city–industry integration, with the degree of matching between the supply and demand of facilities in terms of quantity and space, as well as the level of environmental construction, significantly influencing subjective choices. Accordingly, the primary indicators of “P–I” interaction encompass population vitality, facility supply and demand, and environmental construction. Urban vitality is derived from the interaction between human activities and urban space [57]. Population vitality is a visual indication of the degree of population agglomeration, reflected by population density [58,59]. The extent of spatial matching between functional supply and population demand effectively reflects the “C–P” interaction level. This study adopts the Facility-Population Matching Index (FPMI) for evaluation [60,61], which can not only directly reflect the quantity of facilities and the spatial supply and demand relationship but also accurately adjust the supply of facilities according to the population demand to put forward targeted policy recommendations. The ecological environment underpins the spatial carrying capacity essential for the sustainable development of “industry”, “city”, and “population”. This study reflects the level of environmental construction through green coverage (expressed as NDVI [62,63]) and PM2.5 concentration.

Ultimately, the 16 evaluation indicators selected are as follows (

Table 1. Table of Evaluation Indicator System for City–Industry Integration.

Dimension	Primary Indicator	Secondary Indicator
I–C	Spatial connection between new district and central cities (I–C-1)	X1_Spatial distance from new district to central cities
		X2_Bus routes from new district to central cities
	Spatial connection within the new district (I–C-2)	X3_Bus station coverage ratio
		X4_Subway station coverage ratio
		X5_Road network density
	Functional integration (I–C-3)	X6_Functional mixing degree
	Economic benefit (I–C-4)	X7_Regional output value per unit area
		X8_Space economic vitality
P–I	Job–housing balance (P–I-1)	X9_JHB index
	Income and consumption (P–I-2)	X10_Total Retail Sales of Consumer Goods growth rate/Regional GDP growth rate
	Innovation and development (P–I-3)	X11_Higher education population/Total employed population
		X12_Annual increment of high-tech enterprises
C–P	Population vitality (C–P-1)	X13_Population density
	Facilities supply and demand (C–P-2)	X14_Facility-Population Matching Index (FPMI)
	Environment construction (C–P-3)	X15_Green coverage
		X16_PM2.5

):

3. Materials and Methods

3.1. Research Area

Xixian New Area is located between the built-up areas of Xi’an City and Xianyang City in Shaanxi Province. It encompasses 23 townships and streets in 7 counties (districts) under the jurisdiction of Xi’an and Xianyang City. The planned control area is 882.32 square kilometers. The existing permanent population of Xixian New Area is 1,010,200, including Airport New City, Fengdong New City, Qinhan New City, Fengxi New City, and Jinghe New City. The detailed location of the study area is presented in Figure 2. On 6 January 2014, Xixian New Area was officially established as the first state-level new area with the theme of innovative urban development. In 2015, Xixian New Area was listed as the second batch of national comprehensive pilot areas for new urbanization. On 16 May 2016, it was designated as a comprehensive pilot test area for building a new system of open economy.

3.2. Data Sources

This study utilized data encompassing basic geography, statistics, LBS data, and remote sensing imagery. The main data types and data sources are shown in

Table 2. Data types and sources of the index system of city–industry integration.

Type	Data	Source
Basic Geography	Administrative boundaries	National Catalogue Service For Geographic Information (https://www.webmap.cn/, accessed on 7 March 2024)
	Network	OSM website (https://www.openstreetmap.org/, accessed on 8 March 2024)
	Bus routes	91weitu (https://www.91weitu.com/default.htm, accessed on 25 November 2024)
Statistics	Population	Seventh National Population Census
	Economy	Xixian New Area Construction Administrative Committee of Shaanxi province (http://www.xixianxinqu.gov.cn/, accessed on 10 April 2024) Xixian New Area Statistical Survey Center
	High technology expertise	Science and Technology Bureau of Xixian New Area
LBS data	POI	Amap (https://lbs.amap.com/, accessed on 25 February 2024) Baidu Maps (https://lbsyun.baidu.com/, accessed on 25 February 2024)
	Heatmap (200 m × 200 m)	Baidu Maps (https://lbsyun.baidu.com/, accessed on 12 March 2024)
Remote Sensing Image	NPP—VIIRS nighttime lights (1 km)	National Oceanic and Atmospheric Administration (https://www.noaa.gov/, accessed on 5 June 2024)
	NDVI (30 m)	National Science & Technology Infrastructure (http://www.nesdc.org.cn/, accessed on 5 June 2024)
	PM2.5 (1 km)	National Tibetan Plateau Data Center (https://data.tpdc.ac.cn/home, accessed on 5 June 2024)

:

3.3. Data Processing and Indicator Calculation

The data processing methods and calculation steps for each indicator are as

Table 3. Data processing and calculation methods.

Indicator	Calculation Method
X1_Spatial distance from new district to central cities	The average distance from the center of mass of each New City to the center of mass of Xi’an and Xianyang City
X2_Bus routes from new district to central cities	Total number of bus routes from each New City to Xi’an and Xianyang City
X3_Bus station coverage ratio	Make 500 m buffers centered on the bus stations and calculate: The area of buffers/The total area of New City
X4_Subway station coverage ratio	Make 800 m buffers centered on the subway stations and calculate: The area of buffers/The total area of New City
X5_Road network density	The total length of road network at all levels/Total area of New City
X6_Functional mixing degree	1. The 18,339 POI data points for Xixian New Area obtained from screening and reclassification were categorized into eight types and assigned weights according to public awareness [64]. The classification and weighting are shown in Table A1. 2. Based on the ArcGIS 10.7 platform, according to the administrative division boundaries of Xixian New Area, 200 m × 200 m grids were created by using the tool “Create Fishnet” as the base units for analysis and calculation of the spatial evaluation values [52]: $$N_{ij}={\displaystyle \frac{{\sum }_{k=1}^n{D}_k\times w_k}{{\sum }_{j=1}^m{\sum }_{k=1}^n{D}_k\times w_k}}\times 100\%$$ $\mathrm{where}{N}_{ij}$ $,D_k$ $,w_k$ $\mathrm{are}\mathrm{the}\mathrm{evaluation}\mathrm{value}\mathrm{of}\mathrm{the}{j}_{th}$ $\mathrm{function}\mathrm{within}\mathrm{the}{i}_{th}$ $\mathrm{evaluation}\mathrm{unit}\mathrm{and}\mathrm{the}\mathrm{number}\mathrm{and}\mathrm{weights}\mathrm{of}\mathrm{POI}\mathrm{points}\mathrm{of}\mathrm{category}k$$.n$ $\mathrm{is}\mathrm{the}\mathrm{number}\mathrm{of}\mathrm{POIs}\mathrm{of}\mathrm{the}{j}_{th}$ $\mathrm{function}\mathrm{within}\mathrm{the}\mathrm{evaluation}\mathrm{unit},\mathrm{and}m$ is the number of POIs of all functions within the evaluation unit. 2. Calculate the spatial functional information entropy index based on the spatial evaluation values to reflect the functional mixing degree [53,65]: $${EI}_i=-{\sum }_{j=1}^j\left(N_{ij}\right)\times \mathit{log}\left(N_{ij}\right)$$
X7_Regional output value per unit area	Calculate the GDP/Total area and apply dimensionless processing.
X8_Space economic vitality	Based on the ArcGIS platform, the “Zonal Statistics” tool was utilized to count the average value of nighttime light [66,67], and the data were dimensionless.
X9_JHB index	1. According to the eight-hour working system, 09:00–12:00 and 14:00–17:00 were selected as the rest periods, and 23:00–05:00 was selected as the working period. Based on the ArcGIS platform, the POIs of companies, factories, scientific research institutions, and residential areas were selected as the places of work and residence respectively, and 300 m buffer zones were established by using the “buffer zone” tool. The population intensity point elements of each time were extracted according to the buffer zones. 2. Calculating the JHB index:   Ratio of the mean value of the share of the population activity intensity at the workplaces during working period to the share at the places of residence during rest period: $$I_{JHB}={\displaystyle \frac{{\sum }_{t_1}{a}_{t_{1}n}/\sum t_1}{{\sum }_{t_2}{a}_{t_{2}n}/\sum t_2}}$$ $\mathrm{where}{I}_{JHB}$ is the JHB index $\mathrm{of}\mathrm{the}{n}_{th}$ $\mathrm{area};t_1$ $\mathrm{is}\mathrm{the}\mathrm{working}\mathrm{period};t_2$ $\mathrm{is}\mathrm{the}\mathrm{rest}\mathrm{period};a_{t_{1}n}$ $\mathrm{is}\mathrm{the}\mathrm{ratio}\mathrm{of}\mathrm{the}\mathrm{number}\mathrm{of}\mathrm{active}\mathrm{populations}\mathrm{in}\mathrm{the}\mathrm{workplace}\mathrm{of}\mathrm{the}{n}_{th}$ $\mathrm{New}\mathrm{City}\mathrm{in}\mathrm{the}\mathrm{working}\mathrm{period}\mathrm{to}\mathrm{the}\mathrm{total}\mathrm{number}\mathrm{of}\mathrm{active}\mathrm{populations}\mathrm{in}\mathrm{the}\mathrm{new}\mathrm{district}.a_{t_{2}n}$ $\mathrm{is}\mathrm{the}\mathrm{ratio}\mathrm{of}\mathrm{the}\mathrm{number}\mathrm{of}\mathrm{active}\mathrm{populations}\mathrm{in}\mathrm{the}\mathrm{residence}\mathrm{of}\mathrm{the}{n}_{th}$ New City in the rest period to the total number of active populations in the new district.
X10_Total Retail Sales of Consumer Goods growth rate/Regional GDP growth rate	Dimensionless processing of the data
X11_Higher education population/Total employed population	Dimensionless processing of the data
X12_Annual increment of high-tech enterprises	Dimensionless processing of the data
X13_Population density	Calculate the Number of permanent population/Total area of New City, and perform dimensionless processing of the data.
X14_Facility-Population Matching Index (FPMI)	1. Network analysis based on GIS: The GIS network analysis method was used to calculate the maximum walking range of each residential area for 15 min based on the actual vector road network with 723 residential areas in the study area as the center. The spatial connectivity tool was used to count the number of facilities and the intensity of population activity in each 15 min life circle separately. 2. Correlation analysis: First, based on SPSS27.0 software, a normal distribution test was conducted; the results are shown in Table A2, with p < 0.001 indicating that the data do not meet the criteria for normality. Second, a monotonicity test was performed where the number of facilities served as the dependent variable (Y) and population activity intensity as the independent variable (X). The findings revealed a monotonic relationship between X and Y (Figure A1a). Finally, Spearman correlation analysis was carried out, with results detailed in Table A3. 3. Linear regression model: $$Y_i={\beta }_0+{\beta }_1{X}_i\left(i=1,2,3,\dots \dots ,723\right)$$ $\mathrm{where}{Y}_i$$,X_i$ $\mathrm{are}\mathrm{the}\mathrm{number}\mathrm{of}\mathrm{facilities}\mathrm{and}\mathrm{population}\mathrm{activity}\mathrm{intensity}\mathrm{within}\mathrm{the}15\min \mathrm{life}\mathrm{circle}\mathrm{of}\mathrm{the}{i}_{th}$ $\mathrm{neighborhood};{\beta }_0$ $\mathrm{and}{\beta }_1$ are regression coefficients. The linear regression model was solved by ordinary least squares (OLS), and the results of the model were tested. The parameters of the model are presented in Table A4, while the standardized residual plots are shown in Figure A1b. 4. Calculate FPMI by substituting the number of facilities in the life circle of the $i_{th}$ neighborhood fitted by the regression equation in step 3 into the following equation: $$M_i=Y_i-\widehat{Y_i}\left(i=1,2,3,\dots \dots ,723\right)$$ $\mathrm{where}{M}_i$$,Y_i$$,\widehat{Y_i}$ $\mathrm{are}\mathrm{the}\mathrm{FPMI},\mathrm{the}\mathrm{number}\mathrm{of}\mathrm{fitted}\mathrm{facilities},\mathrm{and}\mathrm{the}\mathrm{number}\mathrm{of}\mathrm{actual}\mathrm{facilities}\mathrm{in}\mathrm{the}{i}_{th}$ neighborhood.
X15_Green coverage	Based on the ArcGIS10.7 platform, the “Zonal Statistics” tool was utilized to count the average NDVI of each New City, and the data were dimensionless.
X16_PM2.5	Based on the ArcGIS platform, the “Zonal Statistics” tool was utilized to count the annual average PM2.5 concentration in each New City, and the data were dimensionless.

:

3.4. Evaluation Model Based on Improved CRITIC Method

The conventional CRITIC method contains two limitations, which are improved in this study: (1) While the original CRITIC framework utilizes standard deviation (an absolute measure of dispersion) to measure data dispersion, this study replaces this with the coefficient of variation (CV). This modification prevents the evaluation process from being affected by the differences in the magnitude of the indicator while preserving essential data characteristics, thereby enhancing methodological robustness and result interpretability [68,69,70]. (2) The conventional conflict quantification method interprets only positive correlations as strong inter-indicator relationships, disregarding the equivalent significance of negative correlations, so the improved CRITIC method employs absolute values of correlation coefficients.

The calculation steps and formulas are as follows:

Construct the evaluation matrix: Assuming that the number of evaluation objects of city–industry integration is m, and the number of evaluation indicators is n. Then, the rating $X_{ij}$ of the $j_{th}$ evaluation index of the $i_{th}$ evaluation object constitutes the original evaluation matrix $X={\left(x_{ij}\right)}_{m·n}$.

Standardization: The collected data are standardized to eliminate the effects of different units and dimensions. X₉ and X₁₄ are considered moderate indicators, with values of 1 and 0, respectively. X₁ and X₁₆ are negative indicators, while the rest are positive indicators. $X_{ij}$ denotes the $j_{th}$ value of the $i_{th}$ indicator. The standardized formula for various indices is as follows:

$$\mathrm{Positive}\mathrm{indicators}:X_{ij}={\displaystyle \frac{X_{ij}-\mathit{min}\left(X_{ij}\right)}{\mathit{max}{X}_{ij}-\mathit{min}{X}_{ij}}}$$

(1)

$$\mathrm{Negative}\mathrm{indicators}:X_{ij}={\displaystyle \frac{\mathit{max}\left(X_{ij}\right)-X_{ij}}{\mathit{max}{X}_{ij}-\mathit{min}{X}_{ij}}}$$

(2)

$$\mathrm{Moderate}\mathrm{indicators}:X_{ij}=1-{\displaystyle \frac{\left|X_{ij}-k\right|}{max\left\{\left|X_{ij}-k\right|\right\}}}$$

(3)

where k is the moderate value. In this study, 0.01 is selected as the leveling variable to level the standardized zero and negative values.

Calculate the coefficients of variation of all indicators:

$$V_j={\displaystyle \frac{S_j}{\overline{X_j}}},\overline{X_j}={\displaystyle \frac{1}{m}}\sum _{i=1}^m{X}_{ij},S_j^2={\displaystyle \frac{1}{m}}\sum _{i=1}^m{\left(X_{ij}-\overline{X_j}\right)}^2$$

(4)

Calculate the independence coefficients of all indicators:

$$C_j=\sum _{i=1}^m\left(1-\left|r_{jl}\right|\right),(j=1,2,\dots ,n)$$

(5)

where $r_{jl}$ is the correlation coefficient between the indicators j and l.

Compute the information content:

$$P_j=V_j\times C_j$$

(6)

Calculate the weights of each indicator:

$$w_j={\displaystyle \frac{P_j}{{\displaystyle \sum _{j=1}^n}{P}_j}}$$

(7)

Calculate the comprehensive evaluation index of the $i_{th}$ evaluated object:

$$U_i=\sum _{j=1}^n{w}_j{X}_{ij}$$

(8)

4. Results

4.1. City–Industry Integration Level in Xixian New Area

Currently, the comprehensive score of city–industry integration in Xixian New Area is 0.4779, reflecting a low level of integration. There is obvious spatial heterogeneity, with the city–industry integration level of Fengdong and Fenxi far exceeding Airport, Jinghe, and Qinhan (

Table 4. The level of city–industry integration in Xixian New Area and each New City.

Area	Fengdong New City	Fengxi New City	Jinghe New City	Airport New City	Qinhan New City	Xixian New Area
City–Industry Integration	0.7693	0.6762	0.3308	0.3343	0.2789	0.4779

). Generally, new cities with better development foundations and location conditions show a higher level of city–industry integration.

4.2. Spatial Differentiation of Each Dimension Level

Overall, the contribution of the three dimensions to the level of city–industry integration varies significantly, with the level of “I–C” interaction much higher than that of “P–I” and “C–P”. Spatial differences also exist in the interaction level across dimensions in each new city (Figure 3). Specifically, the interaction level of “I–C” is higher than that of “P–I” and “C–P” in Fengdong and Fengxi, the interaction level of “C–P” surpasses that of both “I–C” and “P–I” in Airport and Qinhan, and the interaction levels of “I–C” and “P–I” are comparable and exceed that of “C–P” in Jinghe.

There is variability and consistency in the scores for evaluation indicators in each new city. In terms of “I–C” interaction, as shown in Figure 4, significant spatial variability exists in the levels of spatial connectivity (“I–C-1” and “I–C-2”) and economic efficiency (“I–C-4”), while the level of functional integration (“I–C-3”) is spatially consistent. From the secondary indicators, Fengdong and Fengxi are far ahead of other new cities in all indicators, while Jinghe, Airport, and Qinhan are lagging behind to different degrees due to various factors. Regarding spatial connectivity, Airport and Jinghe exhibit the lowest levels because of location disadvantages and lagging public transportation development. Qinhan has excellent location advantages but inadequate transportation infrastructure. As for the general phenomenon of low functional integration, Figure 5 visualizing the functional mixing degree (X₆) further reveals that the Xixian New Area exhibits a low degree of functional mixing, with a dominance of single functions, and the spatial distribution follows the pattern of “overall low-level mixing and local high-level mixing”, indicating a spatial mismatch exists between the industrial and urban functions, as well as a low land-use efficiency. According to economic efficiency, while Jinghe’s economic output remains relatively low, its economic vitality is showing positive trends. In contrast, Qinghan and Airport demonstrate low economic output and vitality levels, indicating an urgent need to optimize their industrial structures, enhance industrial diversity, and strengthen economic vitality.

In terms of “P–I” interaction, the overall job–housing structure (P–I-1) and income–consumption structure (P–I-2) are relatively stable, but the spatial differences in innovation development (P–I-3) are significant (Figure 6). As shown in

Table 5. JHB index and balance type in Xixian New Area and each New City.

Area	Fengdong	Fengxi	Jinghe	Airport	Qinhan	Xixian New Area
JHB index	0.87	0.85	1.18	1.11	2.01	1.21
Balance type	Relatively balanced	Relatively balanced	Relatively balanced	Relatively balanced	Employment-oriented	Employment-oriented

, Xixian New Area shows an employment-oriented job–housing relationship. With the exception of Qinhan, which is employment-oriented, the other four new cities are relatively balanced [54]. Figure 7 reveals that the employment and residential spaces are highly consistent, and Qinhan shows low heat values in both periods, with the higher value area concentrated around Zhouling Emerging Industrial Park. In addition, the level of income–consumption matching (P–I-2) is similar in Fengxi, Jinghe, Airport, and Qinhan, but Fengdong exhibits a clear mismatch. As shown in

Table 6. Regional GDP growth rate and total retail sales of consumer goods in each New City.

New City	Fengdong	Fengxi	Jinghe	Airport	Qinhan
Total Retail Sales of Consumer Goods	−7.20%	5.90%	−0.30%	2.00%	1.30%
Regional GDP Growth Rate	−1.10%	9.20%	9.20%	4.40%	5.40%

, Fengxi, Airport, and Qinhan show positive retail sales and GDP growth rates, indicating upward economic dynamism and consumption rates in these regions. Jinghe shows less decline in the growth rate of total retail sales of consumer goods and a notable GDP growth rate. Although Fengdong ranks highest in economic output, both indicators are negative, and the values are highly variable. Furthermore, the level of innovation development (P–I-3) in Fengdong and Fengxi far surpasses that of the other three new cities. From the perspective of secondary indicators, the proportion of the higher education population and the annual increment of high-tech enterprises in Fengdong and Fengxi are much higher than in other new cities, reflecting the accumulation of high-quality educational and innovative resources.

In terms of “C–P” interaction, Figure 8 shows significant spatial differences in population vitality (C–P-1) and environmental construction (C–P-3), and a basic alignment between the supply and demand for facilities (C–P-2). Except for Fengdong, the population vitality of other new cities is generally low. Concerning facilities supply and demand, visualizing the FPMI (X₁₄) found that the imbalance of supply and demand in Jinghe is mainly due to the overallocation of facilities, with the overmatched facilities concentrated in Jingyang County, and the other new cities are generally balanced in supply and demand (Figure 9). Regarding environmental construction, further analysis of secondary indicator scores reveals that Fengdong’s green coverage (X₁₅) and PM2.5 (X₁₆) scores are both low, which can primarily be attributed to the high proportion of industrial and residential land development, the limited number of green spaces and parks, as well as the comparatively severe air pollution resulting from construction activities. Fengxi has the lowest score for PM2.5 (X₁₆), indicating that while green construction is good, air pollution significantly impacts environmental construction. The better air quality in Jinghe and Airport, as well as the high green coverage (X₁₅) in Qinhan, have a positive effect on environmental construction levels.

4.3. Evaluation Verification Based on Entropy Weight Method

In order to verify the stability of the evaluation results, this study employs the EM to assign the indicators and calculate the scores of each dimension of each new city to compare the consistency of the calculation results of the two methods. The EM is also used as an objective assignment method, which is assigned according to the discrete degree of the indicator data, which can objectively reflect the amount of information of the indicators, and is widely used in the comprehensive evaluation process. The calculation steps are as follows:

Normalization: The original data have been standardized and non-negative shifted in the process of CRITIC method calculation; therefore, the data need to be further normalized.

$$Y_{ij}=X_{ij}/\sum _{i=1}^m{X}_{ij}$$

(9)

Calculate the entropy:

$$e_j=-k\times {\sum }_{i=1}^m{(Y}_{ij}\times \mathit{ln}{Y}_{ij}),k={\displaystyle \frac{1}{\mathit{ln}m}}$$

(10)

Calculation the information entropy redundancy:

$$d_j=1-e_j$$

(11)

Calculate the weights of each indicator:

$$W_j=d_j/\sum _{j=1}^n{d}_j$$

(12)

Calculate the “I–C”, “P–I”, “C–P” and City–Industry Integration scores:

$$U_i=\sum _{j=1}^n{W}_j{X}_{ij}$$

(13)

Table 7. Results of indicator weighting based on the EM.

Indicator	X1	X2	X3	X4	X5	X6	X7	X8
Entropy	0.0414	0.0491	0.0735	0.0677	0.0561	0.0541	0.1024	0.0544
CRITIC	0.0669	0.0439	0.0618	0.0597	0.0441	0.0462	0.0555	0.0619
Indicator	X9	X10	X11	X12	X13	X14	X15	X16
Entropy	0.0327	0.0323	0.0829	0.1127	0.1081	0.0323	0.0402	0.0601
CRITIC	0.0685	0.0487	0.1074	0.0578	0.0841	0.0901	0.0408	0.0626

compares the weights of each indicator calculated based on the entropy value method and the CRITIC method, and it can be found that there is a certain difference between the weights of the indicators calculated by the improved CRITIC weight method and the entropy value method. In order to further compare whether the weight differences on the evaluation results lead to low stability of the evaluation results, and considering the small number of samples, this paper adopts Kendall’s tau-bx correlation analysis to test the consistency of the calculation results of the level of industry–city integration and the scores of each dimension of the five new cities.

Table 8. Correlation coefficients between the results of the CRITIC method and the EM.

Dimension	I–C–P		I–C		P–I		C–P
Method	CRITIC	Entropy	CRITIC	Entropy	CRITIC	Entropy	CRITIC	Entropy
CRITIC	1 (0.019 **)	0.738 (0.077 *)	1 (0.017 **)	1 (0.017 **)	1 (0.017 **)	1 (0.017 **)	1 (0.017 **)	0.8 (0.083 *)
Entropy	0.738 (0.077 *)	1 (0.019 **)	1 (0.017 **)	1 (0.017 **)	1 (0.017 **)	1 (0.017 **)	0.8 (0.083 *)	1 (0.017 **)

**, * represent the correlation significance at the level of 0.01, 0.05 and 0.1, respectively.

shows that the correlation coefficient between the scores of the level of city–industry integration of the five new cities calculated by the entropy value method and the CRITIC method is 0.738, the correlation coefficient of the score of the level of I–C interaction is 1, the correlation coefficient of the score of the level of P–I interaction is 1, and the correlation coefficient of the score of the level of C–P interaction is 0.8, with a p-value of 0.083. It can be seen that the calculation results of the two methods present significant consistency.

Table 9. City–industry integration degree in Xixian New Area and New Cities based on EM.

Area	Fengdong New City	Fengxi New City	Jinghe New City	Airport New City	Qinhan New City	Xixian New Area
City–Industry Integration	0.8217	0.6467	0.2959	0.2468	0.2088	0.4440

shows the city–industry integration scores of Xixian New Area and each new city calculated by the entropy method. By comparing them with Table 4, the consistency of the results of the two methods can be visualized. Firstly, the comprehensive scores of city–industry integration in Xixian New Area are all within the range of 0.4–0.5. Second, Fengdong and Fengxi rank in the top two, Qinhan ranks at the bottom, and the scores of Airport and Jinghe are very similar. Figure 10 visualizes the results calculated by the entropy method, and it can be found that the spatial variability characteristics of the interaction level of each dimension and the scores of each indicator are consistent with the results calculated by the CRITIC method.

5. Discussion

5.1. Reasons for the Low Level of City–Industry Integration

The results indicate that the level of city–industry integration in Xixian New Area is generally low, which aligns with the findings of many scholars [18,23,24,71]. The reasons for this outcome can be attributed to several factors:

(1) Lack of Focus on “Population”: The average score of the “I–C” interaction is much higher than the average scores of “P–I” and “C–P” interactions. This imbalance is the primary reason for the overall low level of city–industry integration in Xixian New Area. It reflects a neglect of the role of population in the development process. Previous studies assessing industry, city, and population individually have shown that while industrial development and urban construction levels are relatively high, population development lags significantly [16,26]. This phenomenon arises from the prioritization of individual system development at the cost of fostering the necessary interactions, particularly overlooking the population as a crucial intermediary between industrial growth and urban development.

(2) Significant Differences in Resource Endowments: There are notable disparities in resource endowments within Xixian New Area, leading to an imbalance in the development of the five new cities. The highest score among the five cities is 0.7693, while the lowest score is only 0.2789. Such uneven spatial development is common at both the city and national levels and is primarily due to variations in regional location conditions, resource endowments, and economic foundations, which affect population attraction and investment potential [18,71]. For instance, Fengdong stands out significantly from the other four New Cities due to its proximity to Xi’an City, with its overall development level far surpassing that of the other districts. Therefore, it is essential to explore strategies for improving city–industry integration by considering the perspectives of location characteristics and structural attributes [72].

5.2. Analysis of the Spatial Differentiation in Interaction Levels of Dimensions and the Rationality of the Evaluation Indicators

“I–C” interaction: The spatial differences in the “I–C” interaction are mainly in terms of spatial connectivity and economic benefits. On the one hand, differences in location and the level of public transportation are the main reasons for differences in spatial connectivity. On the other hand, numerous studies using economic panel data have shown that the industrial economy plays a dominant role in promoting city–industry integration [17,30]. However, the economic performance of Xixian New Area remains relatively low and has significant spatial variations. As indicated in Figure 4, the regional output per unit area (X₇) has the highest score, while the score for spatial economic vitality (X₈) is low. This disparity suggests that while the area shows high production output, the economic benefits have not been fully realized in terms of urban vitality. This indicates that traditional economic indicators alone are insufficient to fully capture the interaction between industrial development and urban spatial optimization in the new district. In addition, Xixian New Area shows relative spatial consistency in terms of functional integration. Figure 11 shows that areas with a concentration of industrial land have relatively little public service and residential space, and a considerable portion of the region is marked as “no data”, indicating that certain areas are dominated by the primary industry and have failed to form an effective link with new industries. A similar situation can be seen in Lanzhou New Area, which also suffers from a low degree of functional mixing [73]. In addition, Milton Keynes New Town, as one of the most successful cities in the UK, initially exhibited a single-function nature and over-reliance on the central city during its development. Its successful practices in industrial agglomeration and humanistic orientation offer valuable lessons for developing countries.

“P–I” interaction: The results of this study show significant spatial differences in innovation development. Innovation capability is the primary driver of city–industry integration, as confirmed by numerous studies [12,74]. In practice, Fengdong and Fengxi have actively attracted and cultivated high-tech enterprises since 2017, while Jinghe, Airport, and Qinhan are primarily focused on the manufacturing, service, and logistics industries and only began developing high-tech enterprises after 2020. This delay has led to a significant gap in innovation scores among the new cities, contributing to insufficient self-driven momentum in these areas. This challenge is common in many new districts with a weak industrial base [75]. Meanwhile, job–housing is one of the key issues in developing new districts. The job–housing structure is generally employment-oriented in Xixian New Area. Relevant studies have pointed out that the limited and single-type jobs available have led to a tendency for the local workforce to migrate to Xi’an and Xianyang cities, while the foreign workforce tends to adopt a “same-day commuting” mode of employment and residence [76]. This study further explored the factor of spatial mismatch. As shown in Figure 7 and Figure 11, the mismatch between the supply and demand for employment space and residential space is a major factor influencing the choices of the population. Furthermore, while the levels of industrial development and population consumption show a general upward trend, the total amounts remain low. Industrial development creates jobs, attracts population concentration, drives consumption, and in turn promotes further industrial upgrading. Xixian New Area is subject to the radiation-driven effect of Xi’an City. However, the area is also affected by the siphoning effect of Xi’an City, particularly in Fengdong New City. Without unique employment advantages, Fengdong may struggle to maintain population concentration, which could negatively impact overall consumption and growth rates.

“C–P” interaction: Differences in population vitality and environmental construction are the main characteristics of the “C–P” interaction. One of the primary constraints on consumption and population agglomeration in new districts is the small resident population, which contributes to the city–industry separation [77,78]. Some scholars have argued from the perspective of population inflow into new districts that lower income levels compared to the central city, a “crowding-out effect” on labor-type populations, and a poor industrial structure are key reasons why the population struggles to concentrate [79]. This study further explores how the future development focus of each New City, regional spatial constraints, and demographic differences between old and new urban areas contribute to the disparities in population vitality. Moreover, research has suggested that a favorable ecological environment directly supports population and industrial agglomeration [74]. However, this study reaches the opposite conclusion: a negative correlation between the levels of environmental construction and population vitality. This issue essentially stems from the conflict between the traditional “industry-oriented” development model and the “human-oriented” goal of new urbanization, which reflects the contradiction and policy paradox between ecological protection and short-term economic development. For instance, Jinghe set the dual goals of industrial clustering and green-low-carbon development. However, large-scale industrial investment may intensify short-term carbon emission pressure. Moreover, it reveals a deviation between the new district’s planning guidance, policy formulation, and actual development. The spatial layout concept of “open and harmonious” in Xixian New Area emphasizes ecological isolation, which is intended to realize intensive land use and avoid rapid space expansion. However, it also leads to the job–housing separation, increases the commuting cost, and inhibits the population vitality. Similar problems also existed in the “sleeping city” phenomenon in the early days of Tama New Town in Tokyo. Additionally, the utilization of spatial data on facilities and population reveals spatial imbalances, compensating for the limitations of statistical data that make it difficult to compare spatial differences when results are similar. This study indicates that the allocation of public services is primarily based on meeting standardized numbers, often disregarding the specific needs of the population, and areas with lower population densities may experience mismatches due to noncompliance with facility configuration standards.

5.3. Limitations

Although this study has achieved some results, it still has the following shortcomings. (1) This study only evaluates the city–industry integration in Xixian New Area, but lacks a comparative study between Xixian New Area and other new districts in China. (2) This study evaluates the current level of city–industry integration in Xixian New Area and analyzes the possible reasons for it, but does not explore in depth the specific factors affecting the level of city–industry integration. (3) This study only evaluates the level of city–industry integration in Xixian New Area in 2022, focusing more on the spatial correlation of city–industry integration without combining the time series for dynamic assessment, which makes it difficult to reflect the historical trend of the level of city–industry integration in Xixian New Area. Therefore, in future research, we can try to expand the analysis of the historical trend of city–industry integration from the perspective of time series and combine cutting-edge nonlinear machine learning to reveal the driving factors of regional internal differences [80,81].

5.4. Policy Recommendations

Based on the findings from the previous study and discussion, the following suggestions are proposed for improving city–industry integration in Xixian New Area, focusing on the three aspects of “I–C”, “P–I”, and “C–P” interactions:

“I–C” Interaction:

(1) Strengthen the construction of various public transport systems, such as subways and bus lines, to improve the connectivity between different transport modes. Notably, Jinghe New City, located in the north, has cultural and tourism resources like Fu Tea Culture and Lewa City, but is distant from Xi’an and Xianyang. Future planning should consider extending subway routes and increasing bus lines to the central cities, enhancing the attraction for consumers, promoting cultural and tourism industries, and boosting regional appeal. In addition, Fengdong and Fengxi advocate the TOD development model, which can refer to the successful experience of Tama New Town in Japan in the traffic mode of human–vehicle separation and TOD-oriented development and construction.

(2) Promote the mixed use of land, transitioning from a single-function to a mixed-function new district. The focus should be on strengthening living functions and avoiding the separation of production and living spaces. Specifically, increasing land use efficiency in the Energy Financial Trade Zone, the core area of Fengxi New City, and the residential areas along the Weihe River will help maximize location advantages. Additionally, the relocation of industrial functions from the urban center to the periphery will create a radiation effect, fostering regional synergies.

“P–I” Interaction:

(1) Leverage resource endowments for the new city’s differentiated industrial development. Fengdong and Fengxi possess distinct geographical advantages, diverse industrial types, and concentrated scientific and educational resources. Drawing on the industrial development experience of Milton Keynes New City, Fengdong and Fengxi can achieve industrial complementarity and dislocation competition with Xi’an and Xianyang, offer differentiated job opportunities, reduce commuting and population spillovers, and attain job–housing balance. Moreover, emulating the experience of Berlin’s Adlershof Science City, integrating universities, research institutions, and high-tech enterprises in the new city can foster industry–university–research integration, stimulate innovation potential, and attract talent. Regarding the development of Jinghe, Airport, and Qinhan, they should focus on refining advantageous industries to prevent homogenized competition from weakening the industries’ self-driven force.

(2) To address weak income and consumption, it is recommended to spearhead new consumption initiatives, foster new technologies and industries, and enhance the competitiveness of local industries. Additionally, collaboration with neighboring regions can help cultivate innovation potential, linking Fengdong with Xi’an Hi-Tech Industries Development Zone to achieve industry, population, and resource agglomeration, thus stimulating regional economic growth and boosting overall consumption levels.

“C–P” Interaction:

(1) Based on the service demands of various population groups, improve public service facility construction, particularly in Jinghe and Airport, where industrial development is weak. Address mismatches between supply and demand for facilities, such as in the Energy Financial Trade Zone and Kunming Pool Plate in Fengdong, the Innovation Port and the first phase of the Core Zone in Fengxi, and the Cultural Tourism and Commerce Plate in Jinghe.

(2) Strengthen the focus on green development by incorporating environmental protection into industrial growth. Economic benefits should be pursued alongside social and ecological benefits to ensure the sustainable, long-term development of the new districts. In particular, Fengdong and Fengxi should continue to encourage green enterprises, boosting the value of urban land and consolidating the results of sponge city construction to attract industrial investment and foster population agglomeration.

6. Conclusions

This study aims to construct an evaluation system from the interaction of “industry” “city” and “population”, and to analyze the city–industry integration level at the new district scale by combining spatial big data and statistical data. Using the Xixian New Area in western China as a case study, CRITIC method was applied to determine the weights and calculate the city–industry integration level based on 2022 data. The findings are highlighted as follows:

(1) The comprehensive level of city–industry integration in Xixian New Area is relatively low, just 0.4779, and the rankings of the new cities from high to low are as follows: Fengdong > Fengxi > Airport > Jinghe > Qinhan. Significant differences were found in the development levels of the dimensions, with the “I–C” interaction higher than the “P–I” and “C–P” interactions.

(2) Low attention to population and unbalanced internal development are the primary reasons for the low level of city–industry integration. Therefore, to realize high-quality sustainable development in the new districts, it is essential to pay more attention to the needs of all kinds of people and promote the improvement of urban functions, attract population agglomeration, and diversify the industrial structure with a humanistic orientation. Meanwhile, new cities should pay attention to the use of resource endowments to cultivate differentiated competitive advantages and avoid homogenized development.

(3) There is both spatial variability and consistency in the levels of the dimensions:

“I–C” interaction: Greatest spatial differences exist in spatial connection and economic benefits, which are closely related to location conditions and development foundations. Additionally, the low degree of functional integration and inefficient land use emerge as common challenges in the development of new districts.

“P–I” interaction: Jinghe, Airport, and Qinhan should accelerate the cultivation of high-tech enterprises in the future, establish competitive industrial chains, and strengthen the self-sustaining momentum of their industries. Job–housing and income–consumption structures are relatively stable but still need to be upgraded.

“C–P” interaction: Substantial spatial heterogeneity in population vitality and environmental construction, while the supply and demand for facilities exhibit relatively minor overall variations but demonstrate significant localized disparities.

In summary, this study establishes an “I–C–P” evaluation framework, integrating spatial big data with traditional socio-economic statistics, to comprehensively evaluate the degree of city–industry integration and the level of interaction among various dimensions within the Xixian New Area from a spatial perspective. Furthermore, it identifies the primary factors contributing to the low level of city–industry integration and highlights the spatial disparities in the development levels across dimensions and the limitations in the growth of each new city. These findings offer valuable insights for promoting high-quality city–industry integration and sustainable development in the Xixian New Area, as well as providing a reference for other regions, particularly cities in the West.