How Sustainable Are Chinese Cities? Empirical Insights from Eight Cities Using a Multidimensional Catastrophe Progression Model

Abstract

Sustainable development remains a crucial global priority. Despite significant progress at both the policy and technical levels, disparities in urban development and the absence of a comprehensive evaluation framework impede practical outcomes in China. While previous research has established the value of multidimensional frameworks and international indices for assessing urban sustainability, existing studies often lack the integration of local dynamics and rely on linear methods that cannot fully capture the complex, nonlinear changes in Chinese cities. This study proposes a four-dimensional indicator system and employs the catastrophe progression method to evaluate sustainable development levels. This study used ten years of panel data (2012–2022) from eight representative Chinese cities and normalized and analyzed 38 sub-indicators to derive membership values for each city and dimension. The findings reveal substantial disparities in sustainable development across the cities, with notable improvements in environmental indicators but persistent volatility in social welfare and resource efficiency. Technological innovation and education resource allocation emerge as management priorities for most cities. This methodological innovation fills a critical gap, offering a replicable framework for other developing countries and supporting the localization of global sustainability agendas. The study’s findings directly inform policy, advancing the achievement of the UN Sustainable Development Goals.

1. Introduction

Cities face acute challenges, including climate change, resource depletion, and worsening social inequities, that hinder progress toward the UN Sustainable Development Goals (SDGs) under accelerating urbanization [1,2,3], which are often intensified by deficient urban planning and restrained institutional capacity. As a result, there is a growing consensus that achieving resilient and sustainable cities requires integrated frameworks that holistically balance economic growth, ecological resiliency, and social equity while fostering effective governance and management [4,5,6].

Urban sustainable development indicators are increasingly pivotal in shaping global policy as cities now house over half of the world’s population; this is projected to reach 70% by 2050. International frameworks, such as the Sustainable Development Index, Urban Sustainability Assessment Framework, Green Development Index, Integrated Urban Sustainability Framework, and Multi-Criteria Decision Analysis (MCDA), have established broad and standardized frameworks for measuring urban sustainability [7,8,9]. A comparison of these indices is depicted in Table 1.

While existing research has advanced both the theory and practice of urban sustainable development evaluation, the current framework lacks the integration of dynamics in Chinese cities [10,11]. Despite making good progress in indexing urban sustainability globally, the most critical issues remained unsolved, such as methodological inconsistencies, insufficient local adaptation, the inadequate treatment of temporal dynamics, and the need for more participatory approaches [12]. Data gaps and inconsistencies in measurement frameworks hinder effective policy action, particularly in developing countries [13]. This paper adopted an approach that can support participatory governance and timely interventions, and is more locally adaptive than the other frameworks. A comparison between urban sustainability approaches is justified in Section 2.1. As a result, urban managers are better equipped with timely and actionable insights to formulate resilient, sustainable development strategies.

Table 1. Existing frameworks employed for urban sustainability measurement.

Framework/Method	Overview	Country/Region	Dimension 1	Ref.
Sustainable Development Index (SDI)	A composite index that aggregates multiple economic, social, and environmental indicators to evaluate a country’s overall progress toward sustainable development, with an emphasis on balancing human development and ecological sustainability.	China (Beijing, Shanghai), European Union	E, S, V	[14]
Urban Sustainability Assessment Framework (USAF)	A structured tool that uses specific indicators across economic growth, social equity, resource efficiency, and environmental quality to assess and guide the sustainability performance of urban areas, highlighting key areas for improvement.	China (Guangzhou, Shenzhen), Australia (Queensland), Iraq	E, S, R, V	[15]
Green Development Index (GDI)	A metric that compares the progress of cities or regions in environmentally sustainable development by integrating economic growth with environmental protection, focusing on resource efficiency and environmental quality.	Cities in developing countries (Central Asia, Southeast Asia, South Asia)	E, V	[16]
Urban Sustainability Framework (USF)	A unified guide developed by the World Bank to help cities assess, plan, and monitor sustainability across economic, social, and environmental dimensions.	Pilot applications in several developed and developing countries	E, S, V	[17]
City Resilience Framework (CRF)	A tool that emphasizes interconnected socio-technical–ecological systems and relies on participatory processes to guide city resilience.	100 cities globally (mostly in the US and Europe)	E, S, R, V	[18]
Multi-Criteria Decision Analysis (MCDA)	An approach that offers techniques for structuring and evaluating decisions in urban sustainability by simultaneously considering multiple criteria.	China, Turkey, Poland, Italy	E, S, R, V	[7]

¹ Note: Economic (E), Social (S), Resource (R), Environmental (V).

Table 1. Existing frameworks employed for urban sustainability measurement.

Framework/Method	Overview	Country/Region	Dimension 1	Ref.
Sustainable Development Index (SDI)	A composite index that aggregates multiple economic, social, and environmental indicators to evaluate a country’s overall progress toward sustainable development, with an emphasis on balancing human development and ecological sustainability.	China (Beijing, Shanghai), European Union	E, S, V	[14]
Urban Sustainability Assessment Framework (USAF)	A structured tool that uses specific indicators across economic growth, social equity, resource efficiency, and environmental quality to assess and guide the sustainability performance of urban areas, highlighting key areas for improvement.	China (Guangzhou, Shenzhen), Australia (Queensland), Iraq	E, S, R, V	[15]
Green Development Index (GDI)	A metric that compares the progress of cities or regions in environmentally sustainable development by integrating economic growth with environmental protection, focusing on resource efficiency and environmental quality.	Cities in developing countries (Central Asia, Southeast Asia, South Asia)	E, V	[16]
Urban Sustainability Framework (USF)	A unified guide developed by the World Bank to help cities assess, plan, and monitor sustainability across economic, social, and environmental dimensions.	Pilot applications in several developed and developing countries	E, S, V	[17]
City Resilience Framework (CRF)	A tool that emphasizes interconnected socio-technical–ecological systems and relies on participatory processes to guide city resilience.	100 cities globally (mostly in the US and Europe)	E, S, R, V	[18]
Multi-Criteria Decision Analysis (MCDA)	An approach that offers techniques for structuring and evaluating decisions in urban sustainability by simultaneously considering multiple criteria.	China, Turkey, Poland, Italy	E, S, R, V	[7]

¹ Note: Economic (E), Social (S), Resource (R), Environmental (V).

This research aims to assess the urban sustainability level of Chinese cities, enabling the identification of nonlinear dynamics and the establishment of an early warning management system for proactive urban sustainability governance. This study aims to achieve these objectives:

i.

To design and establish a multidimensional urban sustainability indicator system integrating economic, social, resource and environmental, and human capital dimensions specifically adapted to Chinese urban development characteristics.

ii.

To apply and validate the catastrophe progression method as a nonlinear analytical approach for modeling complex interactions and threshold effects in urban sustainability across eight representative Chinese cities using ten years of panel data (2012–2022).

iii.

To construct a dynamic early warning management system that translates sustainability assessments into actionable insights for urban policymakers.

This research assumes that the index system can effectively reveal multi-dimensional indicators’ synergy and mutation characteristics, and provide a theoretical basis for early warning management strategies. This study selects eight cities in China as the case study, namely Beijing, Shanghai, Guangzhou, Shenzhen, Hangzhou, Wuhan, Chongqing, and Xi’an. Their selection is justified in Section 3.1. This study employs CPM and membership functions to reflect the overall level of sustainable urban development, serving as the basis for the early identification of potential risks and formulating management recommendations.

2. Literature Review

The study of urban sustainability has existed for some time; however, integrating multiple dimensions into urban sustainability indicators in the Chinese context is a relatively new field. The pertinent literature can be categorized into five areas of methodological relevance and multi-dimensions (economic, social, resource and environmental, and human capital), in accordance with the Theory of Urban Complexity, which is outlined by Batty [19] and illustrates cities as complex systems by nature, where every stakeholder interacts across multiple dimensions and constantly makes decisions.

2.1. Methodological Relevance

The MCDA method is extensively applied in urban sustainability assessment, providing a structured approach to dealing with urban systems’ inherent complexity. MCDA encompasses a variety of methods, each tailored to address specific challenges in evaluating sustainability indicators. Among them, CPM is particularly effective at analyzing nonlinear relationships and potential risks, such as sudden environmental or infrastructural failures within urban sustainability indicators [20,21]. The Analytic Hierarchy Process (AHP) structures complex decision problems into a hierarchy of goals, criteria, and alternatives. It employs pairwise comparisons to derive the relative weights of each indicator, making it advantageous when expert judgment and qualitative analysis are essential [22,23]. Principal Component Analysis (PCA) is a statistical technique that reduces the dimensionality of large datasets by extracting principal components that capture the maximum data variance. This simplification distills complex sustainability datasets into a manageable set of key indicators, preserving essential information for analysis while minimizing redundancy [24].

However, the effective application of CPM demands high-quality data and expertise in model construction to capture these nonlinear behaviors accurately. Also, CPM was employed to evaluate the informatization level in four Chinese regions [25], and the results were found to be consistent with prior expectations, proving that the CPM could work well for Chinese cities. The details of CPM are further evaluated in Section 3.2.3. AHP facilitates stakeholder consensus building; however, its reliance on subjective assessments can introduce bias. PCA is useful when dealing with extensive quantitative data, but may be less effective in capturing qualitative or context-specific factors. These methods have proved to be a mixed blessing. Hence, selecting the appropriate method should be based on the specific objectives of the assessment goals, the nature and availability of data, and the contextual characteristics of the urban environment [26].

Due to the rapid development and transformation of Chinese cities, the indicators across the four dimensions exhibit dynamic and nonlinear changes. Traditional methods, such as AHP and PCA, are limited in their ability to effectively capture these complex dynamics. In contrast, as a dynamic analytical method, the CPM is well suited to identify sudden changes and potential risks within urban sustainability indicators. By leveraging CPM to establish a dynamic early warning system, urban managers and decision makers can proactively detect emerging risks and implement timely intervention measures.

2.2. Economic Dimension

From an economic perspective, this study highlights that economic coordination is an important foundation for achieving sustainability [27]. International scholars have explored the potential linkages among finance, trade, economic growth, and sustainable development. For example, Magazzino (2022) proposed the “the greener the richer” assumption, examining the potential relationship between economic development and sustainable development [28]. This assumption suggests that economic growth can coexist with environmental protection and resource conservation, while sustainable development practices can, in turn, foster economic prosperity. In addition to exploring macroeconomic factors, this study also considers the role of specific industries, such as green building technology, which are recognized as important drivers of sustainable urban development [29]. These technologies not only enhance energy efficiency and reduce environmental impact, but also create new economic opportunities and stimulate growth in related industries.

2.3. Social Dimension

The social dimension is an important factor affecting sustainable development [30]. Education is widely regarded as the cornerstone for providing and cultivating talents who can contribute to sustainable development [31]. By providing individuals with the knowledge and skills necessary for sustainable practices, education helps to build a workforce that can drive innovation and implement sustainable solutions. At the same time, physical education plays a crucial role in achieving sustainable development goals, as it promotes health and well-being, which are important components of sustainable development [32]. In addition, scholars believe that social support systems, including social services and community participation, are crucial for promoting social cohesion and equity. These systems ensure that different social classes share the benefits of development more fairly, reduce disparities, and improve overall social well-being. This study defines the social dimension as encompassing the broader concepts of social equity, inclusion, and community well-being, such as social security and social services. These indicators are intended to measure how well societies provide safety nets and enable full community participation, allowing governments to assess the effectiveness of policies aimed at reducing inequality and fostering social integration.

2.4. Resource and Environmental Dimension

This dimension is fundamental to achieving sustainable development, as it encompasses both the efficient use of resources and the protection of natural ecosystems. Research highlights the significant role of energy and resource productivity in enhancing environmental quality and advancing the SDGs [5]. Renewable energy consumption is another key factor, as it reduces the dependence on fossil fuels and lowers greenhouse gas emissions. The transition to renewable energy not only helps mitigate climate change but also creates new economic opportunities and enhances energy security. Improving productivity in these areas not only reduces environmental degradation but also contributes to economic growth and social well-being. A critical aspect of this dimension is the transition to renewable energy sources, which reduces the dependence on fossil fuels and lowers greenhouse gas emissions. This shift not only mitigates climate change but also fosters new economic opportunities and strengthens energy security. Equally important is the sustainable management of natural resources—such as water, land, and biodiversity—which is essential for maintaining ecological balance and supporting long-term development. Technological advancements, including green building technologies and digital solutions, play a pivotal role in promoting sustainable urban development [33], in minimizing the environmental footprint, and enabling efficient resource management and monitoring. This dimension serves as a cornerstone for sustainable development and improved quality of life.

2.5. Human Capital Dimension

The dimension of human capital stands as a pivotal factor in driving sustainable urban development. Within the framework of sustainable urban development, the realms of health, education, and technology are intricately intertwined. Public health is the foundation for adopting new technologies and pursuing innovative endeavors [34]. In sustainable urban development, thoughtful urban planning and establishing green spaces contribute to fostering healthy lifestyles among residents [35]. Education is a significant catalyst to enhance residents’ understanding of sustainability and motivate them to engage in environmental protection initiatives and green consumption practices [36]. Furthermore, digital technology could enhance the transparency of urban governance [37], which is a driving factor in sustainable urban development.

This study recognizes the interconnectedness of social and human capital. The human capital dimension is focused on the physical and mental well-being of individuals, as well as their knowledge, skills, and competencies. This approach aligns with established frameworks such as the World Bank’s Human Capital Index, which uses health and education as key metrics [38].

2.6. Research Gaps

Two important gaps persist. First, there is a lack of integrated, multidimensional assessment frameworks tailored to Chinese urban contexts. Existing research tends to either adopt international frameworks without adequate localization or address only isolated aspects of urban sustainability [39]. From a handful of urban sustainability indicators studies [14,15,16,17,40], rarely are there studies focusing on MCDA. This method is practical and feasible for policymakers and has the ability to reflect timely interventions for management efficiencies. The framework is empirically validated using real-time data, ensuring that the assessment is grounded in the local context and captures the unique development trajectories, spatial differences, and rapid industrialization characteristics of China’s urbanization. By using this framework, this study moves beyond the limitations of international frameworks, which often overlook China’s socio-economic and policy environment. Second, many existing evaluations suffer from the insufficient application of advanced and nonlinear analytical methods in urban sustainability evaluation. A linear and static perspective does not adequately capture the complex, nonlinear, and potentially abrupt changes inherent in urban systems. This study employs the CPM—a nonlinear analytical approach capable of capturing abrupt changes and complex interactions among sustainability indicators. Unlike traditional linear methods (such as AHP or PCA alone), CPM can model the threshold effects and nonlinearities prevalent in urban systems experiencing rapid change. The research applies CPM to real-world data and facilitates the early detection of emerging risks, supporting timely policy intervention, thus providing actionable management insights for city policymakers.

3. Materials and Methods

This paper aims to develop a multidimensional urban sustainability assessment framework tailored to the Chinese context, integrating four dimensions, covering 11 indicators and 38 sub-indicators. Eight Chinese cities were selected as case studies. This study first normalizes the data to eliminate the differences in scale and order of magnitude between different indicators. Then, a specific dataset of each indicator will be post processed using the CPM to derive the membership function value for each city to evaluate the urban sustainable indicators.

3.1. Overview of the Study Area

This study selected eight cities in China as the case study, including Beijing, Shanghai, Guangzhou, Shenzhen, Hangzhou, Wuhan, Chongqing, and Xi’an (see Figure 1). These cities were selected for their representativeness in economic performance, urbanization patterns, regional diversity, and development stages.

Economic performance-wise, the chosen cities are among the most economically influential in China, consistently ranking at the top in terms of GDP, industrial output, and innovation capacity. For example, Beijing, Shanghai, Shenzhen, and Guangzhou are recognized as the nation’s economic powerhouses, leading in technology, finance, and trade. Hangzhou and Wuhan are well known for their rapid economic growth and innovation ecosystems, while Chongqing and Xi’an serve as major financial centers in western China, reflecting the country’s efforts to balance regional development.

In the context of urbanization patterns, these cities exemplify diverse urbanization trajectories. Beijing and Shanghai, as traditional megacities, have experienced continuous high-density urban growth, while Shenzhen’s booming expansion is linked to its designation as the first Special Economic Zone of the nation. Chongqing and Guangzhou show different urban expansion models; Chongqing remains in a diffusion phase, and the latter demonstrates diffusion–coalescence patterns.

They also provide regional representation, spanning China’s major regions. Beijing, Shanghai, and Hangzhou represent the eastern side, and Chongqing and Xi’an represent the western part. Guangzhou and Shenzhen are in the south, and Wuhan is in the center. This choice allows for the analysis of how regional policies and resource allocations could shape urban sustainability.

These cities advance through diverse development stages. First, Beijing and Shanghai are metropolises with advanced infrastructure and higher international connectivity. Shenzhen and Hangzhou are emerging innovation hubs that lead in the technology and digital economy sectors. Wuhan, Chongqing, and Xi’an are focal points for national strategies to promote inland and western development. Their diversity enables the comparison of the challenges and opportunities at different development stages.

3.2. Data Sources and Methodology

3.2.1. Data Sources and Processing Method

Data sources. The primary data for this study are sourced from official publications, such as statistical yearbooks, government-published reports, and relevant online databases of municipal governments. The data sources are shown in Section S3.

Sub-indicator Selection and Weighting Scheme. In this analysis, all indicators are considered equally important. This study applies an equal-weighting approach, assigning identical weights to each indicator without further adjustments.

Processing of Missing Data. In cases where data are missing or incomplete for certain years or cities, a range of methods, such as predictive modeling, linear regression analysis, gray prediction models, exponential smoothing prediction method, and linear interpolation method, are employed to supplement datasets. The details of missing data processing methods for various data sources are shown in Section S3.

3.2.2. Research Design

This article mainly utilizes the catastrophe progression method, collects data from Beijing, Shanghai, Guangzhou, Shenzhen, Hangzhou, Wuhan, Chongqing, and Xi’an for the period between 2012 and 2022, and establishes a comprehensive sustainable indicator system in China to answer two research questions. First, how effectively does the proposed four-dimensional sustainability assessment framework capture and quantify nonlinear performance dynamics in Chinese cities? Second, which critical sustainability thresholds identified by the four-dimensional assessment can inform early warning indicators and guide proactive policy interventions in Chinese urban governance? Figure 2 depicts the process flowchart to illustrate the overall research ideas and clarify its structure. The CPM integrates fuzzy mathematics to develop a catastrophe fuzzy membership function [41,42].

Based on the literature [14,43] and the World Bank’s Human Capital Index [38], this paper develops a Sustainable Development Index system, containing 11 indicators and 38 sub-indicators under four dimensions. The process then proceeds with data normalization through the minimum–maximum method to normalize the data. Subsequently, based on the catastrophe theory, a fuzzy membership function of each indicator is constructed to reflect the overall level of sustainable urban development. The system then aggregates these membership values to derive membership values for each city. This value serves as the basis for the early identification of potential risks and formulating management recommendations. By leveraging the CPM methodology, this study calculates membership values to generate targeted policies, such as the China Blue Sky Action Policy, which is supported by providing a digitally driven pollution control schedule. Furthermore, the approach improves the local adaptability of the “dual carbon” strategy, aligning with China’s goals of ecological civilization and high-quality development. Cities are encouraged to adopt green building technology, aligning with the national strategy. Overall, this study provides a scientific basis for supporting green transformation policies.

3.2.3. Catastrophe Progression Method

The CPM is grounded in mathematical catastrophe theory, which analyzes how nonlinear systems can abruptly shift from one state to another when critical thresholds are crossed [42,44]. CPM is used to model sector-specific risks by quantifying how different subsystems interact. The method is highly quantitative, employing normalized indicators and objective weighting to feed into catastrophe progression formulas (like cusp, swallowtail, and butterfly models), which then yield security indices and risk classifications. CPM’s scope is tailored to technical systems with measurable variables and relies on the retrospective analysis of historical data to identify patterns of failure or instability. Its approach is largely data-driven, with minimal stakeholder input. While this makes CPM effective in data-rich environments for predictive early warning, it also introduces limitations: the method is highly dependent on the availability of detailed longitudinal data.

3.2.4. Application of Catastrophe Model

The $f\left(x\right)$ in

Table 2. Functions of the catastrophe model.

Catastrophe Model	Potential Function
Cusp Catastrophe	$f(x)=x^4+a{x}^2+bx$
Swallowtail Catastrophe	$f(x)={\displaystyle \frac{1}{5}}{x}^5+{\displaystyle \frac{1}{3}}a{x}^3+{\displaystyle \frac{1}{2}}b{x}^2+cx$
Butterfly Catastrophe	$f(x)={\displaystyle \frac{1}{6}}{x}^6+{\displaystyle \frac{1}{4}}a{x}^4+{\displaystyle \frac{1}{3}}b{x}^3+{\displaystyle \frac{1}{2}}c{x}^2+dx$

represents the potential function of a state variable $x$ in the system. In the formula, a, b, c, and d represent the control variables of the state variable $x$, where the control variables reflect the degree and direction of their influence on the state variable $x$ through the coefficients a, b, c, and d.

3.2.5. Establishment of Sustainable Development Index System

Table 3. Indicators and the numbers of sub-indicators adopted by each catastrophe model.

Catastrophe Model	Indicator	Number of Sub-Indicators
Cusp Catastrophe	Prices	2
	Social Services	2
Swallowtail Catastrophe	Foreign Trade	3
	Population	3
Butterfly Catastrophe	National Economy	4
	Social Security	4
	Resources	4
	Environment	4
	Health	4
	Education	4
	Technology	4

illustrates the three models and their indicators that are involved in the study, along with the number of sub-indicators under each indicator.

Table 4. Sustainable Development Index system.

2nd Level	3rd Level	4th Level	Unit	Attribute
Economy	National Economy A	Gross Domestic Product D1	RMB 100 MM	+
		Proportion of Tertiary Industry Output to GDP D2	%	+
		Per Capita Consumption Expenditure D3	RMB	+
		General Public Budget Revenue D4	RMB 100 MM	+
	Foreign Trade B	Total Import and Export Volume D5	USD 100 MM	+
		Actual Utilization of Foreign Direct Investment D6	USD 10,000	+
		Outward Foreign Direct Investment D7	USD 10,000	+
	Prices C	Consumer Price Index by Category D8	Preceding Year = 100	+
		Retail Price Index by Category D9	Preceding Year = 100	+
Social	Population D	Total Population D10	10,000 persons	+
		Total Dependency Ratio D11	%	+
		Natural Population Growth Rate D12	%	+
	Social Security E	Original Insurance Premium D13	RMB 100 MM	+
		Average Wage of Employed Persons D14	RMB	+
		Total Employed Population D15	10,000 persons	+
		Number of Urban Minimum Living Standard Guarantee Recipients D16	10,000 persons	+
	Social Services F	Number of Basic Endowment Insurance Participants at Year-end D17	10,000 persons	+
		Number of Social Organizations D18	unit	+
Resources and Environment	Resources G	Area of Park Green Space D19	Hectare	+
		Per Capita Water Resources D20	m3/person	+
		Per Capita Energy Consumption D21	kgce	+
		Electricity Consumption D22	100 MM kwh	+
	Environment H	General Industrial Solid Waste Generation D23	10,000 tons	−
		Days of the AQI meeting or exceeding the national grade II standard D24	day	+
		Industrial Wastewater Ammonia Nitrogen Discharged D25	ton	−
		Industrial COD Discharged D26	ton	−
Human Capital	Health I	Number of People Subsidized for Basic Medical Insurance D27	10,000 persons	+
		Number of Beds in Urban Medical and Health Institutions D28	unit	+
		Total Number of Health Personnel D29	10,000 persons	+
		Number of Health Institutions D30	unit	+
	Education J	Allocation in Education Funds D31	RMB 10,000	+
		Number of Institutions of Higher Education D32	unit	+
		Proportion of Local Financial Investment in Education Intensity D33	%	+
		Number of Faculty and Staff in Higher Education Institutions D34	10,000 persons	+
	Science and Technology K	Number of Units with R&D Activities D35	unit	+
		Internal Expenditure on R&D Funds D36	RMB 10,000	+
		Number of R&D Personnel D37	person	+
		Number of Patent Grants D38	unit	+

Note: MM—million, CNY RMB 1 = USD 0.13. Source: developed and compiled by the authors.

describes the four-level indicator system developed by the authors. The first level of the index system is dedicated to urban sustainable development. The second level consists of the four-dimensional framework. The third level includes 11 categories under the catastrophe models (labeled A to K), while the fourth level covers 38 sub-indicators (labeled D1 to D38).

3.2.6. Normalization of Indicators

Following the stepwise decomposition and quantification of the indicators, the indicators are further broken down into sub-indicators. This study employed dimensionless processing (min–max normalization) to standardize the data, rescaling the indicator values into positive and negative categories as follows:

i.

Positive indicators: GDP and Per Capita Expenditure.

ii.

Negative indicators: Common Industrial Solid Wastes Generated, Industrial COD Discharged.

(1)

Constructing the Evaluation Object

This study develops a multi-comprehensive evaluation system of sustainable development indicators to evaluate the impact of different indicators on different dimensions. The evaluated object is assumed to be $A=(A_1,A_2,A_3,\dots A_a)$ and the evaluation index as $B=(B_1,B_2,B_3,\dots B_a)$. The value of the evaluated object $A_m$ to the rating index $X_{mn}=$ ($m=1, 2, 3\dots a, n=1, 2, 3\dots a,),$ hence, the original index data matrix results as follows:

$$X_{mn}=\left(\begin{array}{ccc}{X}_{11}& \dots & X_{1b}\\ ⋮& \ddots & ⋮\\ X_{a1}& \dots & X_{ab}\end{array}\right)$$

(1)

where $X_{ab}$ represents the value of the $a$th indicator in the $b$th sample.

(2)

Dimensionless Transformation of Sustainability Indicators

The dimensionless processing for positive indicators is conducted as follows:

$$Y_{ab}={\displaystyle \frac{X_{ab}-X_{min(b)}}{X_{max(b)}-X_{min(b)}}}$$

(2)

The dimensionless treatment of negative indicators is as follows:

$$Y_{ab}={\displaystyle \frac{X_{max(b)}-X_{ab}}{X_{max(b)}-X_{min(b)}}}$$

(3)

In Formulas (2) and (3), $Y_{ab}$ denotes the initial value of the catastrophe fuzzy membership function after dimensionless processing. Each sample ${ X}_{ab}$ represents the original data, where $X_{ab}$ corresponds to the original data of item B for the $a$th evaluation sample within the system. Following dimensionless transformation, the data are normalized within the range of [0, 1].

3.2.7. Normalization Formulas for Catastrophe Models

The catastrophe fuzzy membership function value is quantified and analyzed in a hierarchical manner using the normalization formula, ultimately leading to the derivation of membership function. Normalization in this context involves the calculation of the first and second derivatives of the catastrophe function, which further infer the critical and bifurcation points within the potential function. The set of divergence points reflects the evolutionary path of the system state as the control variable of the catastrophe potential function changes. When the control variable enters the region defined by these divergence points, the system may experience an abrupt transition from one state to another, signifying a sudden change.

i.

Cusp Catastrophe Model

The following potential function represents this model:

$$f\left(x\right)=x^4+a{x}^2+bx$$

(4)

The first derivative of $f\left(x\right)$ is given by:

$$f^{\prime }\left(x\right)=4x^3+2ax+b$$

(5)

Setting the first derivative $f^{\prime }\left(x\right)$ = 0 yields:

$$f^{\prime }\left(x\right)=4x^3+2ax+b=0$$

(6)

Setting the second derivative $f^{″}\left(x\right)$ = 0 yields:

$$f^{″}\left(x\right)=12x^2+2a=0$$

(7)

By solving the above equations simultaneously, the bifurcation point can be expressed as:

$$a=-6x^2, b=8x^3$$

(8)

where $a=-{\displaystyle \frac{a}{6}}$, $b={\displaystyle \frac{b}{8}}$. Based on these results, the normalization formula for the cusp catastrophe model is derived as follows:

$$X_a=\sqrt{a}, X_b=\sqrt[3]{b}$$

(9)

ii.

Swallowtail Catastrophe Model

The potential function for this model is as follows:

$$f(x)={\displaystyle \frac{1}{5}}{x}^5+{\displaystyle \frac{1}{3}}a{x}^3+{\displaystyle \frac{1}{2}}b{x}^2+cx$$

(10)

The first derivative of $f\left(x\right)$ is given by:

$$f^{\prime }\left(x\right)=x^4+a{x}^2+bx+c$$

(11)

Setting the first derivative $f^{\prime }\left(x\right)$ = 0 yields:

$$f^{\prime }\left(x\right)=x^4+a{x}^2+bx+c=0$$

(12)

Setting the second derivative $f^{″}\left(x\right)$ = 0 yields:

$$f^{″}\left(x\right)=4x^3+2ax+b=0$$

(13)

By solving the above equations simultaneously, the bifurcation point can be expressed as:

$$a=-6x^2, b=8x^2, c=-3x^4$$

(14)

where $a=-{\displaystyle \frac{a}{6}}, b={\displaystyle \frac{b}{8}}, c=-{\displaystyle \frac{c}{3}}$, the normalization formula for the swallowtail catastrophe is derived as follows:

$$X_a=\sqrt{a}, X_b=\sqrt[3]{b}, X_c=\sqrt[4]{c}$$

(15)

iii.

Butterfly Catastrophe Model

The potential function for the model is as follows:

$$f(x)={\displaystyle \frac{1}{6}}{x}^6+{\displaystyle \frac{1}{4}}a{x}^4+{\displaystyle \frac{1}{3}}b{x}^3+{\displaystyle \frac{1}{2}}c{x}^2+dx$$

(16)

The first derivative of $f\left(x\right)$ is given by:

$$f^{\prime }\left(x\right)=5x^4+3a{x}^2+2bx+c$$

(17)

Setting the first derivative $f^{\prime }\left(x\right)$ = 0 yields:

$$f^{\prime }\left(x\right)=5x^4+3a{x}^2+2bx+c=0$$

(18)

Setting the second derivative $f^{″}\left(x\right)$ = 0 yields:

$$f^{″}\left(x\right)=20x^3+6ax+2b=0$$

(19)

By solving the above equations simultaneously, the bifurcation point can be expressed as:

$$a=-6x^2, b=8x^2, c=-3x^4, d=4x^5$$

(20)

where $a=-{\displaystyle \frac{a}{6}}, b={\displaystyle \frac{b}{8}}, c=-{\displaystyle \frac{c}{3}}$, $d={\displaystyle \frac{d}{4}}$, hence the normalization formula for the butterfly catastrophe is derived as follows:

$$X_a=\sqrt{a}, X_b=\sqrt[3]{b}, X_c=\sqrt[4]{c}, X_d=\sqrt[5]{d}$$

(21)

To sum up, the normalization formulas of the three catastrophe models are illustrated in

Table 5. Normalization formulas for catastrophe models.

Catastrophe Model	Normalization Formula
Cusp Catastrophe	$X_a=\sqrt{a}, X_b=\sqrt[3]{b}$
Swallowtail Catastrophe	$X_a=\sqrt{a}$, $X_b=\sqrt[3]{b},X_c=\sqrt[4]{c}$
Butterfly Catastrophe	$X_a=\sqrt{a},X_b=\sqrt[3]{b},X_c=\sqrt[4]{c},X_d=\sqrt[5]{d}$

Note: The catastrophe level values of control variables $X_a, X_b, X_c$, $X_d$. $x$ is the state variable, $f\left(x\right)$ is the potential function of x, where a, b, c, and d are the control variables of $x$.

.

3.2.8. Complementarity Analysis of Sustainable State

The membership function is first calculated based on multi-objective fuzzy decision theory. Suppose a given scheme includes objectives such as $A_1, A_2, A_3,$ etc., with corresponding membership functions $\mu A_1\left(\mathrm{x}\right), \mu A_2\left(x\right), \mu A_3\left(x\right),$ and so forth. The membership function $\mu \left(\mathrm{x}\right)$ in this scheme is defined as the minimum value among these objective membership functions, that is $\mu \left(x\right)=\mu A_1(x)\wedge \mu A_2\left(x\right)\wedge \mu A_3\left(x\right)$….

Those with higher membership function values are considered superior. The normalization is applied following the principle of selecting the minimum value from the normalized sets. This is then followed by assessing complementarity. If complementarity relationships exist, the arithmetic mean of the indicators is typically used as the final comprehensive score. If no complementarity is present, the principle of selecting the minimum values from the set, followed by the maximum and median, is applied. Lastly, when comparing target objects, the principle of ranking based on the highest comprehensive evaluation score is adopted. A higher score indicates a better evaluation outcome.

4. Results

4.1. Normalization of Specific Indicators

i.

Economic Dimension (GDP)

Figure 3 depicts the normalized GDP values across the case studies. Shanghai led in this dimension (1.000 in 2022), followed by Beijing, reflecting their strong economic foundations. Shanghai is a prominent global financial city with an open investment market environment, and Beijing benefits from its role as the capital in politics and culture.

The southern cities—Shenzhen and Guangzhou—recorded GDP values of 0.696 and 0.607, respectively, demonstrating significant economic vitality and competitiveness. Their economic structure, led by high-tech and modern service industries, enhances their regional influence and economic driving capability. Hangzhou (0.357) and Wuhan (0.360) also exhibited positive development trends. Hangzhou’s rapid growth in the digital economy and Wuhan’s advancements in emerging sectors such as intelligent manufacturing have steadily elevated their positions within the national economic landscape, progressively strengthening their economic influence. As a key city in western China, Xi’an recorded a normalized GDP of 0.177. It has a weaker economic influence; however, this is compensated by its scientific research and industrial foundation, particularly in aerospace and electronic information, alongside its strategic location on the Belt and Road Initiative, which provides future growth potential for its economic development. In summary, the GDP data reveals a relatively significant disparity in the economic development trajectories among cities.

ii.

Environment Dimension (Air Quality)

As shown in Figure 4, Shenzhen has the best air quality performance among the eight examined cities. This is evidenced by its comprehensive approach, which includes the continuous optimization of its energy structure, stringent industrial pollution controls, the vigorous promotion of electric vehicles, and enhanced ecological conservation efforts through habitat restoration and biodiversity protection.

Guangzhou demonstrated a relatively strong performance in 2018, which is probably attributed to the “2018 Guangzhou Plan for Total Emission Reduction of Major Pollutants”. Conversely, Beijing’s air quality performance is the worst among the cities, which is attributed to two factors. First, ozone remains a major pollutant despite the decrease in concentrations of traditional pollutants. Second, Beijing’s energy structure remains heavily reliant on coal. The southern cities, particularly Guangzhou and Shenzhen, have increasingly turned to other greener energy production.

Xi’an’s air quality is comparatively modest, likely influenced by its basin topography, resulting in a temperature inversion layer unfavorable for pollutant dispersion. However, Xi’an’s energy structure is transitioning from coal-dominated to diversified and clean, with increasing proportions of electricity, natural gas, and non-fossil fuel energy. Also, its industrial structure is centered around electronic information, aerospace, and biomedicine, promoting green and low-carbon economic development. In summary, energy policy and industrial structures remain the main driving factors for air quality in Chinese cities.

4.2. Calculation of Membership Values Using CPM

The membership values for indicators and the case study are calculated using the CPM, as shown in Section 4.2. The results and associated data are provided in Section S2, depicting the membership values of each third-level indicator for each city and year before proceeding to a comprehensive evaluation of the sustainable development level. The data helps reveal variations in sustainable development performance across cities.

The following example shows how membership value is derived. The data is modeled using the butterfly catastrophe model, specifically following the non-complementary criterion. Using Beijing’s national economy data from 2020 to 2022, the CPM function is expressed as follows. The method is similarly employed for other indicators as categorized in the previous Table 3.

$$\begin{array}{l}{x}_{2022}=\mathit{min}\left(x_{D1}^{{\displaystyle \frac{1}{2}}},x_{D2}^{{\displaystyle \frac{1}{3}}},x_{D3}^{{\displaystyle \frac{1}{4}}},x_{D4}^{{\displaystyle \frac{1}{5}}}\right)\\ =\left(1^{1/2}, 1^{1/3}, {0.944}^{1/4}, {0.917}^{1/5}\right)\\ =0.983\\ x_{2021}=0.827\\ x_{2021}=0.865\end{array}$$

(22)

The calculation showed that Beijing’s economic performance fluctuated between 2020 and 2022 but remained relatively high overall. The fluctuation was possibly affected by the city lockdown during the pandemic, and it rapidly recovered to a high level in 2022, reflecting its robust economic foundation and high adaptability.

4.3. Evaluation of Urban Sustainable Development Level

4.3.1. Economic Performance

In terms of economic dimension, Beijing, Shanghai, Shenzhen, and Guangzhou indicate strong economic operational efficiency with economic membership values of 0.937, 0.924, 0.802, and 0.763, respectively, in 2022. These cities are commonly known as “First-tier Cities” in China, and they are the main financial hub and economic barometer of the nation. Shenzhen and Guangzhou are in Guangdong Province, where the latter is the provincial capital. The reason Shenzhen overtook the provincial capital in terms of economic performance is probably due to Shenzhen’s autonomous status as a Special Economic Zone, granting it preferential policies in land, taxation, and foreign exchange, allowing it much greater financial and policy independence than most cities, including Guangzhou. Under this status, Shenzhen created an institutional openness and an international and cross-border platform, attracting a strong private sector.

Xi’an has a relatively weaker economic performance than Wuhan and Chongqing, indicating room for improvement despite the increasing priorities on high-end manufacturing in recent years. Chongqing remains one of the strongest cities in western China.

4.3.2. Social Welfare Provision

In the social dimension, significant disparities exist among cities regarding population, social security, and social services. Beijing, Shanghai, and Guangzhou exhibit relatively high membership values in social security and social services. There are two main reasons to explain this phenomenon. First, Beijing, Shanghai, and Guangzhou have the highest social insurance contribution rates and bases in China [45], which are linked to higher local wages. Second, the favorable demographics and urban appeal attract a vast portion of the young population, sustaining a broad contributor base in social insurance funds that further contribute substantial investments in social welfare, public service infrastructure, and well-established social service systems.

Surprisingly, although Shenzhen has a high economic performance, it has low social service provision. Several reasons explain this phenomenon. Shenzhen uses a much lower minimum wage as the contribution base, unlike other cities that use 60% of the average wage. Next, its high population mobility and short or interrupted contribution periods further reduced benefit levels. Shenzhen adjusted its policy to align with provincial standards in July 2024, eventually increasing labor costs in the city.

4.3.3. Resource and Environmental Management

Shenzhen stands out in the resource and environmental dimension with the highest membership value of 0.803 in 2022. Shenzhen’s environmental indicators are among the best due to green industrial transformation, strict regulation, and innovative governance, such as the “Blue Sky Action Plan” and “Zero-Waste City,” and the city implements technological and market-driven solutions (including carbon trading and green building assessment) and encourages community involvement in environmental protection.

Besides Shenzhen, Shanghai outperforms most Chinese cities in environmental protection due to: (i) industrial relocation and the closure of hundreds of heavy-polluting factories, especially near sensitive areas; (ii) mandatory citywide waste sorting and recycling since July 2019 for all residents and businesses, with strict fines for non-compliance; (iii) Shanghai’s ambitious 2024–2027 Green Action Plan to convert the majority of city transport to green vehicles and to implement massive energy-saving upgrades to building stock; (iv) large-scale urban greening by creating extensive green spaces and ecological corridors, and transformed former industrial riverbanks into parks.

Chongqing and Xi’an showed a relatively lower environmental performance than other cities. Besides their location on unfavorable terrain and climate that causes a temperature inversion layer, their concentration of heavy industries, a coal-dominated energy structure, accumulated historical pollution, and weak pollution control infrastructure lead to easy pollutant accumulation and great difficulty in governance.

4.3.4. Temporal Trends in Sustainable Development

China’s overall sustainable development levels exhibit an upward trend, although some cities experience fluctuations in specific dimensions. Beijing’s economic value showed considerable volatility between 2013 and 2016 but stabilized at a relatively high level from 2017 onwards. Similarly, Shanghai’s environmental values fluctuated significantly from 2012 to 2014, followed by a steady increase and stabilization since 2015. Guangzhou and Shenzhen experienced significant fluctuations in resource utilization efficiency between 2012 and 2014; however, both cities have gradually improved since 2015. In contrast, Wuhan, Chongqing, and Xi’an have exhibited significant fluctuations in normalized values across several dimensions.

The driving factors of these phenomena are as follows:

i.

Policy Cycles and Reforms. National initiatives, such as the 12th and 13th Five-Year Plans and the “Beautiful China” initiative, often lead to rapid policy shifts, causing short-term fluctuations followed by stabilization as policies take effect. Local pilot programs implemented in cities like Beijing and Shanghai also serve as pilots for new economic, environmental, or social policies, resulting in initial volatility as systems adapt.

ii.

Economic Restructuring and Upgrading. Beijing’s economic volatility (2013–2016) aligns with efforts to phase out low-end industries and promote high-value sectors, causing temporary disruptions but yielding long-term stability. The fluctuations in Guangzhou and Shenzhen’s resource utilization (2012–2014) reflect the challenges of transitioning from extensive to intensive growth models, with improvements as industrial upgrading and technological innovation take hold.

iii.

Environmental Regulation and Investment. Shanghai’s environmental fluctuations (2012–2014) correspond to the introduction of stricter pollution controls and large-scale environmental investments, which initially disrupt but ultimately improve environmental indicators. Also, ecological improvements often lag behind policy implementation, leading to short-term volatility before realizing steady gains.

iv.

Urbanization and Demographic Shifts. Wuhan, Chongqing, and Xi’an have experienced rapid urbanization. High rates of migration and demographic change can temporarily disrupt social, economic, and environmental balances.

v.

External Shocks and Global Influences. Global economic shifts (for instance, the 2015–2016 stock market correction and 2018 trade tensions) and public health events such as the 2013 H7N9 outbreak and the COVID-19 pandemic cause short-term disruptions across multiple dimensions of sustainable development.

4.3.5. Key Additional Indicators (Foreign Trade, Social Security, and Innovation)

As illustrated in Figure 5, Shenzhen has consistently maintained a high value in foreign trade. Shenzhen’s dominance in foreign trade is underpinned by its strong high-tech manufacturing base and rapid product innovation. Favorable policies, such as free trade agreements in RCEP and customs facilitation, have significantly reduced the trade barriers and costs for Shenzhen.

Beijing and Shanghai also demonstrate considerable strengths in this domain, catalyzed by their role as the capital city and a financial hub. Hangzhou shows promising potential in foreign trade, leveraging its digital economy strengths and cross-border e-commerce to boost trade.

Despite the recent improvements from Belt and Road strategies, Wuhan, Chongqing, and Xi’an have lower foreign trade values due to less developed export-oriented industries and less favorable geographic positions.

As illustrated in Figure 6, Beijing, Shanghai, Hangzhou, Chongqing, and Xi’an have significantly developed their social security systems, resulting in continuous improvements in residents’ social security level. Cities with older populations (Beijing and Shanghai) have prioritized pension and healthcare system improvements to address demographic challenges. Guangzhou, Shenzhen, and Wuhan have exhibited an upward trend during this period. Guangzhou and Shenzhen’s large migrant populations and flexible labor markets have made achieving stable social security coverage more challenging. In contrast, Wuhan has faced institutional and fiscal constraints due to rapid urban expansion and industrial transformation.

Shenzhen, Beijing, Hangzhou, and Shanghai have consistently maintained high levels in this domain, exhibiting a steady upward trend in Figure 7 due to strong innovation ecosystems, high R&D investment, concentrated talent, supportive policies, and thriving high-tech industries. In contrast, Chongqing and Wuhan are improving but show more fluctuations because of their industrial legacy, lower R&D intensity, and less mature innovation systems, though national initiatives are helping to boost their progress.

4.4. Early Warning Management Strategies

This paper proposes several forewarning response systems as part of early warning management strategies.

4.4.1. Economic Forewarning Response System

Real-Time Economic Monitoring and Early Warning Mechanisms. Powered by big data and AI, this helps cities track economic trends, quickly identify risks like inefficient resource use or sectoral imbalances, and enable timely interventions [46]. For instance, Beijing’s economic membership value reached 0.937 in 2022, reflecting a strong economic foundation; however, its resource and environmental membership values were 0.606 and 0.241. These systems support better decision-making, enhance urban resilience, and are essential for managing the complexity of modern city economies.

City Linkage and Regional Integration. Beijing led in the economic membership value (0.937) in 2022. Shanghai (0.924) and Shenzhen (0.802) also demonstrate strong economic performance, while Guangzhou (0.763) and Chongqing (0.690) are relatively lower. Coordinated development policies, improved infrastructure, innovation spillovers, and complementary industries drive linkage and integration. These factors enable leading cities like Beijing, Shanghai, and Shenzhen to boost regional competitiveness, while integration helps hubs like Chongqing drive economic growth in surrounding areas.

Economic Resilience and Innovation. Shenzhen and Hangzhou, with high economic membership values of 0.802 and 0.598, are driven by substantial R&D investment, advanced high-tech industries, the effective commercialization of research, and active foreign trade. Strengthening foreign trade by leveraging Special Economic Zones’ strategic advantages can enhance cooperation with international cities. Such efforts improve economic resilience [47,48] and advance China’s competitive position within the global industrial value chain.

4.4.2. Social Forewarning Response System

Optimizing Social Services. Beijing and Shanghai demonstrate a strong performance in social security, with values of 0.783 and 0.736, respectively. Optimizing social services requires targeted policies, integrating elderly care, encouraging multi-sector participation, raising social security standards, and using big data for more effective support. The active involvement of all sectors of society is encouraged to support the development of charitable organizations and social welfare initiatives [49,50].

Addressing Social Security Gaps. There remains considerable potential for improving social services in Guangzhou and Shenzhen. Notably, Shenzhen’s education membership value in 2022 was 0.363, significantly lower than its economic (0.802) and environmental (0.803) membership values. The social security and education gaps in Guangzhou and Shenzhen stem from rapid population growth, uneven resource allocation, and institutional barriers, leaving services behind economic progress. To address these issues, cities should enhance social security and public service access for migrant workers and the floating population; strengthen government supervision and improve labor laws to protect vulnerable groups, especially where services are market-driven; expand the coverage and inclusiveness of social welfare programs; and lastly, improve employment and housing security and safeguard the rights of freelancers. These measures will help build a more comprehensive and equitable social security system for all urban residents.

4.4.3. Resources and Environment Forewarning Response System

Ecological Conservation. Shenzhen demonstrated an outstanding environmental membership value of 0.803, while Shanghai, Guangzhou, and Hangzhou also recorded high values of 0.843, 0.823, and 0.513, respectively. Conversely, Beijing, Xi’an, and Wuhan exhibited moderate environmental membership values of 0.241, 0.322, and 0.464, respectively. To strengthen ecological protection, cities should define and plan conservation areas, integrate these with urban and land use planning, and create ecological corridors linking urban areas to natural ecosystems. Prioritizing ecological restoration, expanding green spaces, and enhancing green infrastructure are essential. Including ecological indicators in government evaluations and establishing compensation mechanisms will further encourage sustainable environmental management and long-term resilience.

Ecological Economy Development. The key drivers of ecological economy development include: (i) Reducing automobile and industrial emissions through green transportation and stricter pollution control. (ii) Using big data and remote sensing for intelligent ecological space management and real-time monitoring. (iii) Promoting eco-friendly industries and ecotourism to boost economic and social value. (iv) Adopting low-impact development (LID) technologies and improving waste classification and recovery rates for better resource efficiency. Shenzhen’s resource utilization membership value of 0.271 indicates considerable potential for improvement in this area [51]. (v) Optimizing resource structures and intensifying pollution source control to support sustainable urban growth.

4.4.4. Forewarning Response System in Human Capital Dimensions

Strengthening Human Capital Development. Cities such as Shanghai, Shenzhen, and Hangzhou excel in human capital development due to substantial investments in education, healthcare, and innovation infrastructure, with membership values of 0.649, 0.637, and 0.640, respectively. Their strong performance is supported by top universities, advanced medical facilities, robust R&D ecosystems, and effective policies for attracting and retaining skilled talent.

In contrast, cities like Guangzhou (0.313) and Chongqing (0.454) show moderate results, mainly due to uneven resource allocation, hospital overcrowding, and challenges in attracting high-level R&D personnel. To enhance human capital, cities should prioritize expanding and equitably distributing educational and healthcare resources, implement hierarchical healthcare systems, and foster technological innovation by increasing R&D staff and patent output. These measures will help narrow regional disparities and support sustainable urban development.

5. Discussion

This study contributed to the innovation of a locally tailored methodological framework that could capture urban systems’ complex, nonlinear dynamics. The findings from this paper are largely consistent with the existing literature, indicating that urban sustainability in major Chinese cities is improving, yet some drawbacks have been identified. The cities that perform well in economic and environmental performance still reveal persistent weaknesses in social service provision. This research extends traditional disaster-focused warning systems to encompass socioeconomic and human capital risks.

The findings of this study are consistent with Michalina et al. [52], who also highlighted the need for standardized frameworks in urban sustainability research. The methodological relevance of this study is further supported by Rezvani et al. [53], who demonstrated that MCDA approaches are particularly effective in capturing the complexity of urban systems. The four-dimensional CPM methodology represents theoretical contributions that extend beyond the Chinese context. Recent research increasingly recognizes the inadequacy of traditional three-pillar approaches and the need for frameworks that can capture the complex, nonlinear dynamics of urban systems [54]. The integration of human capital indicators in this study addresses a gap in Chinese urban sustainability research, where societal aspects have been historically underemphasized [55].

Regarding the economic dimension, recent research has found that cities with high economic performance do not necessarily achieve high eco-efficiency, with some economically advanced cities showing poor effectiveness in terms of human well-being elements [56]. This phenomenon resonates with this study’s finding that despite strong economic performance, Shenzhen exhibits relatively low social service provision, highlighting the importance of multidimensional assessment approaches.

This study’s identification of significant temporal fluctuations in sustainability performance, particularly during 2012–2016, aligns with broader national patterns identified in recent research, where comprehensive urbanization quality increased alongside significant regional disparities and coordination challenges [57]. The policy-driven volatility identified in this study corresponds to the broader patterns of policy implementation cycles and structural economic transitions documented in the recent literature.

The spatial clustering of high-performing cities in the Yangtze River Delta and Pearl River Delta regions identified in this study is consistent with Liu et al., where urban ecological efficiency forms continuous spatial clustering patterns with significant spatial spillover effects [58], strengthening the evidence for regional sustainability patterns in Chinese urban development.

Recent global assessments of urban early warning systems found that many large cities, particularly in developing countries, lack adequate early warning capabilities [59]. This study’s integration of sustainability assessment with early warning functionality addresses a gap identified in the recent literature. While traditional early warning systems focus primarily on meteorological and disaster risks, the multidimensional approach developed in this study enables the proactive identification of sustainability risks across economic, social, environmental, and human capital dimensions. This innovation aligns with recent calls for more integrated urban governance approaches that can address complex, interconnected sustainability challenges.

Lastly, the early warning system developed in this study provides a framework for similar proactive policy interventions, demonstrating a substantial alignment with recent policy analysis and implementation studies. Research on China’s Low-Carbon City Pilot policy has found significant positive effects on urban ecological efficiency through green technology innovation and energy transition mechanisms [58]. Also, the recent analysis of Beijing and Shenzhen’s different ESG strategies illustrates the practical relevance of this study’s findings [60]. Beijing’s emphasis on regulatory frameworks and standard setting versus Shenzhen’s focus on technological innovation reflects the differentiated policy approaches identified in this research. The early warning system’s ability to provide city-specific guidance supports more targeted policy development based on local sustainability performance patterns.

6. Conclusions

This study reaffirms the main research idea that a multidimensional, dynamic, and locally adapted evaluation framework is essential for accurately assessing and advancing urban sustainable development in China. By applying the CPM methodology to eight representative Chinese cities, this research systematically quantified sustainability across economic, social, resource, environmental, and human capital dimensions, revealing both the progress made and the persistent disparities among the cities. This paper also presents a range of forewarning response systems applicable to the case study cities, which can serve as valuable references for policymakers seeking to implement timely interventions. In summary, this study successfully achieved the three research objectives outlined earlier and demonstrated the value of CPM in capturing nonlinear dynamics and abrupt changes within urban systems—capabilities that traditional linear methods often lack, thus enabling more responsive early warning and management strategies.

This study acknowledges several limitations. Notably, the sample size was limited and the regional focus may restrict the applicability of findings to broader contexts. These limitations could potentially bias the results and highlight the need for caution when extrapolating the findings to other urban settings. Future studies should address these limitations by expanding sample sizes and including more diverse urban regions to enhance the generalizability of findings, and developing and testing unified methodological frameworks that can be applied consistently across different urban contexts.

The significance of this research lies in its methodological innovation and practical relevance. By integrating multidimensional indicators and employing a catastrophe-based approach, this study provides a more nuanced understanding of urban sustainability and offers a replicable model for other developing countries seeking to localize global sustainability agendas. The findings directly support the achievement of the SDGs, notably SDG 11 (Sustainable Cities and Communities) and SDG 8 (Decent Work and Economic Growth), and reinforce China’s role as a methodological contributor to global sustainability governance.

Ultimately, the path to sustainable urban development in China—and globally—requires integrated, evidence-based approaches that bridge policy, technology, and community engagement. By advancing such frameworks and sharing empirical insights, Chinese cities can accelerate their sustainable transformation and serve as models for urban resilience and innovation worldwide.