1 Introduction

The county is the most complete micro-unit in China's national governance system. Within the urban system of China, county towns play a pivotal role in bridging urban and rural areas, serving as the primary vehicle for the urbanization of rural populations and as a significant breakthrough point for new urbanization. The rapid process of urbanization has brought about significant changes and impacts on county societies, particularly in terms of intensified population movement, increased pressure on urban infrastructure, and the complexity of social structures. As an essential component of the modernization of national governance, county towns also play a critical role in advancing the national emergency response system. Currently, out of China's over 2000 counties, the vast majority of the top 100 counties are concentrated in the economically developed eastern coastal areas, with the Yangtze River Delta region being the most prominent [1]. The Yangtze River Delta is also one of the most urbanized regions in China, where large-scale urbanization has led to varying degrees of changes and impacts on county societies [2].

Within county areas, weak links in various fields may couple together to form systemic risks, gradually leading county societies towards becoming risk-prone with diverse and frequent conflicts [3]. Systemic risks refer to complex risks formed by the coupling of weak links across different fields. These risks may be exacerbated by the interaction of multiple factors, increasing the overall system’s vulnerability [4]. For instance, in the economic domain, rapid urbanization may result in employment instability and widening income disparities; in the social domain, increased population mobility and cultural diversity may trigger social conflicts; and in the environmental domain, natural disasters such as floods and earthquakes may cause devastating impacts on fragile infrastructure. A risk society is characterized by increased instability due to diversity and conflicts. In county societies, the heterogeneity brought about by rapid urbanization complicates social structures and increases the frequency of conflicts and disputes. Heterogeneity refers to the diversity and complexity existing within a society [5]. As urbanization progresses, the differences among various population groups in county societies increase, including disparities in culture, economy, and social status. This heterogeneity may lead to a decline in social cohesion and an increase in social conflicts. In diverse urban areas such as London, New York, and Los Angeles, heterogeneity manifests as the coexistence of people from different ethnic, cultural, and economic backgrounds, which can lead to racial conflicts, social inequality, and uneven resource distribution [6, 7]. At the county level in the Yangtze River Delta region, these risks may manifest as infrastructure vulnerabilities, inadequate public services, and challenges in social governance.

To address the various risk events faced by county societies, conducting pre-assessment and early warning is not only significant but increasingly crucial in the current context. Therefore, this study delves deeply into the factors influencing social safety early warning in the counties of the Yangtze River Delta. By employing the grounded theory analysis method, we explore and understand the relationships and patterns among these factors. Based on this understanding, we establish a comprehensive risk early warning indicator system, which will provide strong support for policy formulation and risk management within counties. This system will also help better tackle the public safety challenges that county societies in the Yangtze River Delta face.

2 Literature review

Steven Fink’s [8] theory of the four stages of crisis and Robert Heath’s [9] “4Rs Model Theory” have provided theoretical support to the study of urban emergency management, serving as significant guidance for the contemporary research endeavors in this field and illuminating the path for scholarly inquiry.

The risks faced by county societies in the Yangtze River Delta region share similarities with urban risks, but they also have unique regional characteristics. These risks include natural disaster risks such as mountain floods, urban waterlogging, and geological hazards [10], as well as public health risks triggered by epidemics [11], and potential risks of mass incidents arising during the urbanization process [12]. Scholars have studied the causes, assessment systems, early warning mechanisms, and governance of county social risks from various perspectives, especially analyzing risk factors in the context of urbanization, including the environmental pressures brought by population migration and urban expansion, as well as challenges in social governance. The research mainly focuses on three aspects: Firstly, the investigation into the causes of social risks at the county level. Researchers have analyzed the causes of mass incidents in county-level societies through an examination of the foundation, key, and core of political identity [13], and explored the correlation between the influx of farmers into urban areas and the risks associated with the urbanization of counties [14]. In addition, they have probed into the factors influencing the vulnerability of populations to high temperatures at the county level [15], as well as the underlying logic behind social risks triggered by the entrepreneurial initiatives of returning young adults [16]. Secondly, research on the assessment systems for county-level social risks. The studies in this area include the employment of GIS (Geographic Information System) technology for the assessment of ecological risks within counties [17], and the use of Analytic Network Process (ANP), Probability Neural Network (PNN), DPSIR (Driver-Pressure-State-Impact-Response) Framework [18], and machine learning algorithms [19] for the assessment of natural disaster risks at the county level. Thirdly, exploration of the early warning systems for county-level social risks. Scholars have paid attention to the multi-dimensional aspects of county-level social risk early warning systems, including the economic, ecological, mass incident, and comprehensive factors. For instance, Li Meijuan [20] has put forward an early warning index system for social risks based on the fact of imbalanced economic development at the county level. With the advent of China’s national strategy for a new type of urbanization, research into the prevention and control of social risks at the county level has progressively deepened, with the perspective gradually shifting towards the governance of such risks. In this context, some scholars put forward to address the county-level social risks from the perspective of city-level security risk governance [21], and they also proposed effective measures to enhance the construction of county-level emergency management systems exemplified by the case study in the northwestern region [22].

In summary, scholars have explored the models, systems, and governance of county social public safety risk early warning from various perspectives, providing theoretical support and reference value for this field of research. However, despite numerous studies and government reports on this topic, existing research lacks systematic thinking in deeply integrating the grounded theory research method into the construction of county social public safety risk early warning systems. Specifically, both domestic and international literature generally recognize that the application of grounded theory in county public safety risk research is still at a preliminary stage. The use of open coding, axial coding, and selective coding to analyze the literature on county public safety risk early warning is not comprehensive enough, and the in-depth analysis of county social public safety issues is still insufficient. Particularly, there is a scarcity of studies that combine theoretical research methods with risk early warning systems.

3 Analysis of public security risk factors in county-level societies based on grounded theory

3.1 Research methodology

Grounded theory [23] is a qualitative research method originally proposed by Glaser and Strauss. It is a specialized methodology for constructing theories from data [24]. This theory encompasses three levels: the routinized techniques and strategies; the more abstract “logic of method”; and the most abstract issues related to epistemology, ontology, axiology, teleology, and validity [25]. The research process of grounded theory is characterized by a rigorous logical structure [26]. The purpose of this study is to explore the influencing factors of public security risks in the county-level societies of the Yangtze River Delta region in China and to establish a monitoring and early warning model based on these factors. Given that this study falls into the category of exploratory studies and currently there lacks a unified and clear understanding of public security risks in county-level societies, the grounded theory was selected as the methodological foundation of this study. Through grounded theory-based analysis of the data, the study aims to abstract and summarize the influencing factors for monitoring and early warning of public security risks in county-level societies, and thus lay the foundation for further exploration of a monitoring and early warning model for public security risks in the county-level societies of the Yangtze River Delta region.

4 Research data

To ensure the comprehensiveness and completeness of the data, this study obtained information from multiple sources, using triangulation through comparing interview results, literature review findings, and policy document requirements. Firstly, purposive random sampling was used to select 18 interviewees from different counties in the Yangtze River Delta region. These interviewees included safety field researchers and county government officials, comprising 5 urban management experts with professorial titles, 4 police officers, 4 directors and clerks from safety production supervision departments, 2 clerks from disease control agencies, and 3 clerks from environmental protection departments.

On one hand, these interviewees are knowledgeable about social safety issues, and on the other hand, they are familiar with the regulatory systems and operational status of social safety supervision, allowing them to quickly grasp the research theme and efficiently provide scientific insights. Through semi-structured interviews, in-depth understanding of the content and implementation processes of public safety risk early warning was obtained. These interviewees were able to accurately grasp the key points of the interview questions, providing comprehensive and detailed responses.

Considering the potential risks in county societies in the Yangtze River Delta, such as environmental, public health, and urban governance risks, an interview outline was designed (see Appendix). Each interview lasted approximately 40–60 min, with the interview texts numbered sequentially from F01 to F18.

Additionally, based on the specific content of the research questions and the availability of data, this study collected data by reviewing relevant policies and literature. These sources included documents such as the “Guidelines for the Construction of Urban Safety Risk Comprehensive Monitoring and Early Warning Platforms (Trial)” and the construction plans for urban safety risk comprehensive monitoring and early warning platforms in cities like Hefei, Nan**g, Chengdu, Foshan, and Shanghai.

Simultaneously, we retrieved and analyzed literature samples related to county social public safety risk early warning, totaling 67 literature samples. Throughout this process, 340 interview statements and 287 original text statements were organized. Of these, 15 randomly selected statements were used to test theoretical saturation, while other comments were used to construct the framework model.

4.1 Data analysis

4.1.1 Open coding

Open coding is the process of reading data that has been initially sifted sentence by sentence and paragraph by paragraph, extracting core information, and then labeling, organizing, summarizing, and categorizing this information. During this process, the extracted information is assigned category names, forming preliminary coding subcategories, so that it is able to conceptualize and categorize similar information within the same category. Herein, a concept refers to a conceptual label attached to individual things, events, or phenomena; a category is a combination of concepts, which can be clustered together when referring to the same phenomenon and are governed by a more abstract, higher-level concept [27]. Through open coding, interview records and literature materials can be conceptualized, allowing for the abstraction and generalization of information and the formation of new conceptual meanings. The subsequent process of categorization further refines and classifies these conceptual meanings, ultimately forming more complete categories. This coding process helps to clarify the main concepts in the data, enhance the ability to abstract and summarize information, and provide a clear and systematic analytical foundation for subsequent research.

Ultimately, this study obtained interview records, literature materials, and other data related to the urban safety risk assessment, totaling approximately 123,000 words. To ensure coding reliability, two researchers were assigned to independently code the textual data, thereby forming a horizontal comparison. Besides, each researcher performed dual rounds of coding on the same data to create a vertical comparison. The coding process yielded a high degree of reliability, evidenced by an inter-coder agreement rate exceeding 90%.

In order to eliminate personal preconceptions and biases as much as possible, 37 core concepts that appeared four or more times were extracted for the social security risk issues in the county domains of the Yangtze River Delta region. Through summarization, these core concepts were organized into 14 subcategories, as shown in Table 1.

Table 1 Concepts and subcategories formed by open coding
Full size table

4.1.2 Axial coding

During the open coding process, this study identified 20 initial categories through the analysis of textual data. However, these categories were independent of each other and lacked hierarchical causal relationships. To establish more logically coherent main categories, axial coding was conducted, and a cluster analysis of the subcategories was performed to clarify the logical relationships between categories. This systematic process revealed the interconnections among the various subcategories. Based on the relationships and logical sequence among different categories, similar thematic categories were grouped together, ultimately forming five main categories with clear logical structures: “Public Health Risks,” “Migratory Population Risks,” “Mass Incident Risks,” “Natural Disaster Risks,” and “Risk Management Mechanisms.” These categories are particularly significant in the context of rapid urbanization, as the environmental pressures, complexities in social governance, and unequal resource distribution brought about by urbanization further exacerbate these risks, as shown in Table 2.

Table 2 The correspondence between main categories and subcategories
Full size table

4.1.3 Selective coding

Selective coding is the refinement and integration of axial coding, which entails identifying the core category from the main categories, exploring the logical relationships between the main categories and subcategories, and ultimately being able to clearly describe the phenomenon. The purpose is to further reveal the connections between the core and main categories. This process that derives the conceptualized and cluster-generalized main categories and subcategories through the analysis of textual materials is used for the study of phenomena and issues. Selective coding connects the main categories through their intrinsic logical relationships, refining the categories to form an analytical framework. The core category identified in this study is “Public Safety Risks in County Societies of the Yangtze River Delta, China.” It consists of five core categories: public health risks, migratory population risks, mass incident risks, natural disaster risks, and risk management mechanisms, along with 20 subcategories. Based on the grounded theory analysis of the textual data, this study ultimately forms the framework for evaluating public safety risks in county societies in the Yangtze River Delta, as illustrated in Fig. 1.

Fig. 1
figure 1

Public security risk framework

Full size image

4.1.4 Saturation test

After the theory construction is completed, it is necessary to test the saturation of the theory, which is the signal for when to stop theoretical sampling [28]. The test of theoretical saturation specifically refers to the continued exploration of the characteristics of categories without acquiring additional data, serving as the basis for deciding whether to cease sampling [29]. If the analysis of newly acquired data does not yield any new theories or categories, then the theory is considered to be approaching saturation. The role of testing for theoretical saturation lies in the expansion of research materials and the refinement of theoretical construction. In this paper, coding analysis was conducted on the reserved original data of 15 case studies, and no new categories were discovered, which validates the theory effectively. Therefore, it can be considered that the mechanism model constructed in this study is saturated. As a result, the study ultimately forms 20 risk early warning indicators across five aspects: “Public Health Risks”, “Migratory Population Risks”, “Mass Incident Risks”, “Natural Disaster Risks” and “Risk Management Mechanisms”.

5 Monitoring and early warning index system for public security risks in county-level societies

5.1 Method for allocating weights to early warning indexes

Given that the factors influencing China’s urban safety are diverse, and these factors are interrelated and mutually constraining, they exhibit characteristics of complexity, dynamism, and uncertainty. Therefore, the use of purely quantitative methods to conduct a comprehensive assessment of urban safety has limited explanatory power. Additionally, the factors influencing urban safety have strong randomness, so relying solely on quantitative methods may lead to a significant discrepancy between the evaluation results and the actual situation. In view of the above circumstances, this study chooses to use the Analytic Hierarchy Process (AHP), which combines quantitative and qualitative methods, to determine the variable weighting and operational logic of the urban safety comprehensive evaluation and warning in China. This method can deal with the complex relationships between various factors in a more comprehensive manner. It not only makes full use of the information from quantitative data but also combines expert experience and qualitative analysis to improve the comprehensiveness and accuracy of the evaluation. This integrated approach helps to more comprehensively understand and warn of the risks to China’s urban safety.

AHP is a quantitative analysis method used to address multi-criteria decision-making problems. It breaks down a complex problem into manageable modules by structuring it into a hierarchy, consisting of the goal, the criteria, and the alternatives. By comparing and inducing, AHP can combine subjective judgment with objective data to obtain the relative weights of various factors, providing a scientific basis for decision-making [30].

The main idea of using AHP method to allocate weights to indexes in this paper is: Utilizing Saaty’s 1-9 scale method [31], to compare and score the importance of indexes within the same level two by two, and then construct the corresponding judgment matrix based on the scores obtained. Subsequently, conduct a consistency check of the judgment matrix. If it does not meet the consistency requirements, the judgment matrix needs to be reconstructed until the consistency requirements are satisfied. The specific steps are as follows.

5.1.1 Judgment matrix construction

The first step is to invite an expert panel to score the weights of the criteria layer and the index layer in accordance with the actual situation of early warning for public safety risks in county-level societies. The 1-9 scale method is used to determine the scores. For a given layer, when comparing the importance of the i-th element relative to the j-th element with respect to a factor of the previous layer, the quantified relative importance is represented by \({a}_{ij}\). If there are n elements involved in the comparison, the final Judgment Matrix A will be obtained as follows, with the corresponding elements being \({a}_{ij}\).

$$A=\left[\begin{array}{cccc}{a}_{11}& {a}_{12}& \cdots & {a}_{1n}\\ {a}_{21}& {a}_{22}& \cdots & {a}_{2n}\\ \vdots & \vdots & \ddots & \vdots \\ {a}_{n1}& {a}_{n2}& \cdots & {a}_{nn}\end{array}\right]$$

The Matrix A has the following characteristics:

  1. (1)

    \({a}_{ij}\) represents the importance of index i compared to index j.

  2. (2)

    When i = j, the two indexes are the same, and therefore equally important, denoted as 1, which means the main diagonal elements are 1.

  3. (3)

    \({a}_{ij}>0\) and satisfies the condition \({a}_{ij}\times {a}_{ji}=1\).

5.1.2 Calculation of single-layer weight ranking and consistency check

The calculation of single-layer weight ranking is based on the established judgment matrix to calculate the weights representing the order of importance of relevant indexes at the current layer relative to a certain index at the previous layer. The calculation involves normalizing all elements in the Judgment Matrix A and then determining the value of \({\overline{a} }_{ij}\) using Formula 1.

$${\overline{a} }_{ij}=\frac{{a}_{ij}}{{\sum }_{k=1}^{n}{a}_{kj}}$$
(1)

After normalizing the matrix, sum up the columns in the same row to obtain \({\overline{w} }_{i}\) using Formula 2.

$${\overline{w} }_{i}=\sum_{j=1}^{n}{\overline{a} }_{ij}$$
(2)

Divide the sum of the index vectors by n to get the weight vector, that is, \({w}_{i}=\frac{\overline{{w }_{i}}}{n}\), and determine the largest characteristic root \({\lambda }_{max}\) using Formula 3.

$${\lambda }_{max}=\frac{1}{n}\sum_{i=1}^{n}\frac{{(Aw)}_{i}}{{w}_{i}}$$
(3)

Perform a consistency check on the data to evaluate the rationality and reliability of the judgment matrix, ensuring that the decision-making results are reliable and credible. The Consistency Index (CI) is obtained using Formula 4.

$$CI=\frac{{\lambda }_{max}-n}{n-1}$$
(4)

CI = 0 indicates perfect consistency, and the larger the CI, the less consistent it is.

Finally, solve for the CR value using Formula 5 to determine if its consistency is acceptable.

$$CR=\frac{CI}{RI}$$
(5)

RI refers to the Random Consistency Index, obtained from a value table. When CR equals 0, the matrix has perfect consistency; when CR is less than 0, the matrix has satisfactory consistency; when CR is greater than 0.1, the matrix lacks consistency and should be adjusted.

5.1.3 Calculation of the total hierarchical ranking of weights

The calculation of the weights representing the relative importance of all factors at a certain layer with respect to the top layer is known as the total hierarchical ranking. To achieve the final objective of the AHP, the composite weights of the indexes at each layer with respect to the goal are calculated sequentially from the bottom up.

5.2 Determination of the weights of the early warning index system

5.2.1 Data source

The data for determining the weights of early warning indicators for public safety risks in county societies of the Yangtze River Delta, China, using the Analytic Hierarchy Process (AHP) were obtained through expert consultations. The Yangtze River Delta region encompasses Shanghai, Jiangsu, Zhejiang, and Anhui provinces. Based on the “2023 China County Economy Top 100 Research” report published by the China Center for Information Industry Development (CCID) [32], affiliated with the Ministry of Industry and Information Technology, Kunshan in Jiangsu Province, Cixi in Zhejiang Province, and Feixi in Anhui Province were selected as research subjects (see Fig. 2).

Fig. 2
figure 2

Location map of the Yangtze River Delta region [33]

Full size image

Experts were selected to score a survey questionnaire based on the public safety risk early warning indicators for these three counties. The collected data from these questionnaires were then gathered and organized for analysis.

5.2.2 Construction of judgment matrix

To reasonably evaluate the factors of public security risk early warning in a certain county society of the Yangtze River Delta region, 8 experts were invited, including 3 government department personnel, 3 university researchers, and 2 security assessment agency managers. These experts possess a high theoretical level and rich practical experience in the management of public security risk early warning at the county level. Therefore, their scores for the relative impact of the public security risk early warning indexes in the said county society can ensure representative assessment results.

Considering that the three counties are from different provinces within the Yangtze River Delta region, the principle of fairness was applied. The weights assigned to the scores for each of these three counties were set equally, with each county given a weight of 1:1:1. Using the expert evaluation method, the highest and lowest scores were discarded, and the average of the remaining scores was calculated to determine the value of each evaluation factor. Pairwise comparisons were then conducted among the factors at the same level to construct the target-level judgment matrix A.

$$A=\left[\begin{array}{ccccc}1& 3& 5& 2& 4\\ \frac{1}{3}& 1& 2& \frac{1}{2}& 2\\ \frac{1}{5}& \frac{1}{2}& 1& \frac{1}{4}& \frac{1}{2}\\ \frac{1}{2}& 2& 4& 1& 2\\ \frac{1}{4}& \frac{1}{2}& 2& \frac{1}{2}& 1\end{array}\right]$$

The importance scores between the indexes at the index layer were compared using the same method to construct the Comparative Judgment Matrices \({B}_{1}-{B}_{4}\).

$${B}_{1}=\left[\begin{array}{ccc}1& \frac{1}{2}& 3\\ 2& 1& 5\\ \frac{1}{3}& \frac{1}{5}& 1\end{array}\right] {B}_{2}=\left[\begin{array}{cccccc}1& \frac{1}{3}& 3& 4& 2& 5\\ 3& 1& 4& 5& 6& 7\\ \frac{1}{3}& \frac{1}{4}& 1& 2& \frac{1}{2}& 3\\ \frac{1}{4}& \frac{1}{5}& \frac{1}{2}& 1& \frac{1}{3}& 4\\ \frac{1}{2}& \frac{1}{6}& 2& 3& 1& 2\\ \frac{1}{5}& \frac{1}{7}& \frac{1}{3}& \frac{1}{4}& \frac{1}{2}& 1\end{array}\right]$$
$${B}_{3}=\left[\begin{array}{ccc}1& \frac{1}{3}& 2\\ 3& 1& 5\\ \frac{1}{2}& \frac{1}{5}& 1\end{array}\right] {B}_{4}=\left[\begin{array}{cccc}1& 2& 3& 4\\ \frac{1}{2}& 1& 2& 3\\ \frac{1}{3}& \frac{1}{2}& 1& 2\\ \frac{1}{4}& \frac{1}{3}& \frac{1}{2}& 1\end{array}\right]$$
$${B}_{5}=\left[\begin{array}{cccc}1& 3& 2& 5\\ \frac{1}{3}& 1& \frac{1}{2}& 3\\ \frac{1}{2}& 2& 1& 2\\ \frac{1}{5}& \frac{1}{3}& \frac{1}{2}& 1\end{array}\right]$$

5.2.3 Determination of weights and consistency check

Considering the large amount of expert subjective experience involved in establishing the judgment matrix, a consistency check is required to eliminate obvious logical errors due to subjectivity. The weights of each judgment matrix were calculated using Formulas 1 and 2, and then the consistency check was conducted using Formulas 3, 4, and 5. The weight vectors of each matrix and the results of the consistency check are shown in Table 3.

Table 3 Weight vector table
Full size table

5.2.4 Overall weight ranking

The final weights of the public security risk early warning index system for county-level societies in the Yangtze River Delta region were obtained by multiplying the weight values of the corresponding criteria and indexes, as shown in Table 4.

Table 4 Weighted overall ranking table
Full size table

According to the data in Table 4, within the dimension of public health risks, the factor with the highest weight is health services, with a weight value of 0.2459, followed by infectious disease control, and lastly, environmental sanitation. This indicates that the process of urbanization has a significant impact on health services, infectious disease control, and environmental sanitation in the counties of the Yangtze River Delta, making these factors more likely to trigger social risks in these areas.

It can be seen that health services, infectious disease control, and environmental sanitation are more significantly affected during the urbanization process. The reason for this is that rapid urbanization often leads to an increase in population density and pressure on public health infrastructure. Therefore, it is recommended to enhance the construction and maintenance of basic health facilities, improve the capacity for infectious disease control, and better manage environmental sanitation.

Within the dimension of migratory population risks, the factor with the highest weight is the economic income of the migratory population, with a weight value of 0.0669. This is followed by social status, education quality, employment situation, and governance policies for the migratory population. This indicates that during the urbanization process, the changes in economic income and social status of migrant workers returning home and integrating into county towns are the most sensitive factors. These two factors are also more likely to trigger social risks in the counties. During the urbanization process, it is essential to develop effective economic support policies and social integration measures to help the migratory population better adapt to urban life, thereby reducing social conflicts and disputes.

Within the dimension of mass incident risks, the factor with the highest weight is social opinion following mass incidents, with a weight value of 0.0443, followed by work efficiency and production accidents. This indicates that the impact of social opinion on the risk of mass incidents in county areas of the Yangtze River Delta is significant during the urbanization process, making it more likely to trigger social risks. The significant influence of social opinion on mass incidents suggests that timely, transparent, and effective public guidance and information dissemination are crucial after such incidents. Additionally, improving the efficiency of government response to mass incidents to prevent small conflicts from escalating into major events is also critical.

In the dimension of natural disaster risks, the factor with the highest weight is disaster resilience, with a weight value of 0.1152, followed by the types of disasters and the number of affected people and houses. This indicates that disaster resilience has the most significant impact on natural disaster risks in county societies within the Yangtze River Delta. It is recommended that county societies strengthen their disaster resilience, including raising the standards for disaster-resistant infrastructure, improving natural disaster emergency plans, and conducting regular disaster drills and training.

Within the dimension of risk management mechanisms, the factor with the highest weight is risk response, followed by the management system and coordination and collaboration, with early warning systems having the lowest weight. This indicates that in the construction and implementation of risk management mechanisms, the ability of county societies to respond to risks has the greatest impact on the effectiveness of these mechanisms. Therefore, during the construction and implementation of risk management mechanisms, the focus should be on enhancing risk response capabilities and ensuring a robust management system. It is also crucial to ensure effective coordination and collaboration among various functional departments when responding to risks. Additionally, the early warning system needs to be improved to increase the speed and accuracy of warning information dissemination, ensuring that residents can receive timely information.

6 Discussion

The rapid urbanization in the Yangtze River Delta region has brought about significant changes in county-level societies, leading to increased diversity, social contradictions, and the emergence of risk societies. This study addresses the urgent need for a robust system to manage these risks through early warning and emergency preparedness. By focusing on county-level societies, which serve as a bridge between urban and rural areas, this research ensures the stability and progress essential for the high-quality development of the region. Previous research has primarily focused on urban risks or risks at a national level, often overlooking the unique challenges faced by county-level societies during urbanization. For instance, Steven Fink’s theory of crisis stages [34] and Robert Heath’s “4Rs Model Theory” [9] have provided a foundation for urban emergency management but have not been specifically applied to the county level within the context of rapid urbanization. This study bridges that gap by applying grounded theory and the Analytic Hierarchy Process (AHP) to develop a comprehensive risk early warning index system tailored to the specific conditions of county-level societies in the Yangtze River Delta.

This research contributes to the theoretical understanding of public safety risks in county-level societies and provides practical tools for risk management. The comprehensive risk early warning index system developed offers concrete guidance for policymakers and local governments. By quantifying the weights of different risk factors, the system facilitates better resource allocation, policy formulation, emergency planning, and long-term risk monitoring and evaluation. The practical applications of the AHP-based early warning indicator system are extensive. It allows for the optimization of resource allocation by prioritizing high-weight risk factors, provides a scientific basis for formulating effective risk management policies, and offers clear guidelines for improving emergency response speed and efficiency. Additionally, it supports long-term monitoring and timely adjustments to risk management strategies, ensuring the safety and stability of county societies.

7 Conclusions

The weight of early warning indicators for public safety risks in county societies within the Yangtze River Delta region is a crucial component of the risk early warning indicator system. It also represents a key element of the risk early warning mechanism during the urbanization process in the region. This study utilized semi-structured interviews and literature collection, employing grounded theory with a three-level coding approach to analyze the factors influencing risk early warning. The study summarized five primary risk early warning indicators: public health risks, migratory population risks, mass incident risks, natural disaster risks, and risk management mechanisms, along with 20 secondary risk early warning indicators. Finally, the weight of each indicator was determined using the Analytic Hierarchy Process (AHP).

The research findings indicate the following: During the urbanization process, among the five primary risk early warning indicators—public health risks, migratory population risks, mass incident risks, natural disaster risks, and risk management mechanisms—the weight of public health risks is the highest, followed by natural disaster risks. Among the 20 secondary risk early warning indicators, the factor with the highest weight within the dimension of public health risks is the public health management system, which also has the most significant impact on society. Within the dimension of migratory population risks, the factor with the highest weight is the social status of the migratory population. Within the dimension of mass incident risks, the factor with the highest weight is social opinion following mass incidents. Within the dimension of natural disaster risks, the factor with the highest weight is disaster resilience.

The relatively high weights of these early warning indicators suggest that county social managers should pay close attention and take measures to prevent these factors from triggering social risks. Specifically:

Public Health Risks: Managers should prioritize improving the public health management system, enhancing the construction of basic health facilities, and increasing the coverage and quality of public health education to effectively address potential public health issues arising from rapid urbanization.

Natural Disaster Risks: Efforts should be made to enhance the disaster resilience of county societies, improve natural disaster emergency plans, and strengthen the monitoring and early warning systems for natural disasters. This will ensure a rapid and effective response when disasters occur.

Migratory Population Risks: Managers should focus on the social status and economic income changes of the migratory population. Relevant policies should be developed to help the migratory population better integrate into county society, thereby reducing risks associated with social status disparities and economic pressures.

Mass Incident Risks: It is essential to strengthen public guidance and information dissemination following mass incidents. Improving the efficiency of government departments in responding to mass incidents is crucial to prevent small conflicts from escalating into major events.

Risk Management Mechanisms: There is a need to improve risk response and management systems, ensuring effective coordination among various functional departments when responding to risks. Additionally, it is vital to enhance the speed and accuracy of early warning information dissemination and ensure proper maintenance of early warning equipment to keep residents well-informed.

In addition, the comprehensive risk early warning indicator system established using the Analytic Hierarchy Process (AHP) not only theoretically ranks the importance of risk factors but also demonstrates its value in practical application. By quantifying the weights of different risk factors, it provides concrete guidance for risk management in county societies. The practical applications are as follows:

Resource Allocation Optimization: Based on the AHP analysis results, the government and related agencies can allocate resources rationally, prioritizing the resolution of high-weight risk factors.

Policy Formulation Basis: It provides policymakers with a scientific basis, enabling them to formulate more precise and effective risk management policies.

Emergency Plan Development: It offers clear guidelines to improve the response speed and efficiency in the event of emergencies.

Long-term Monitoring and Evaluation: It is applicable for both short-term and long-term risk monitoring and evaluation, allowing for timely adjustments to risk management strategies to ensure the safety and stability of county societies.

Overall, the AHP indicator system has significant application prospects in the governance of public safety risks in county societies. Its scientific and practical nature provides a systematic and quantitative tool for risk management, hel** to enhance the level of risk governance in county societies. In future research and practice, this indicator system can be further improved and optimized by integrating big data analysis, intelligent monitoring, and other technologies, thereby enhancing its effectiveness in practical applications.