[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Next Article in Journal
A Geographic Information System-Based Model and Analytic Hierarchy Process for Wind Farm Site Selection in the Red Sea
Previous Article in Journal
TravelRAG: A Tourist Attraction Retrieval Framework Based on Multi-Layer Knowledge Graph
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJGI
Volume 13
Issue 11
10.3390/ijgi13110415
ijgi-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Integrated Assessment of Security Risk Considering Police Resources

by
Jieying Chen
1,2,
Weihong Li
1,2,3,
Yaxing Li
4 and
Yebin Chen
4,*
1
School of Geography, South China Normal University, Guangzhou 510631, China
2
SCNU Qingyuan Institute of Science and Technology Innovation, Qingyuan 511500, China
3
Guangdong Shida Weizhi Information Technology Co., Ltd., Qingyuan 511500, China
4
Research Institute of Smart City, School of Architecture and Urban Planning, Shenzhen University, Shenzhen 518060, China
*
Author to whom correspondence should be addressed.
ISPRS Int. J. Geo-Inf. 2024, 13(11), 415; https://doi.org/10.3390/ijgi13110415
Submission received: 27 August 2024 / Revised: 5 November 2024 / Accepted: 15 November 2024 / Published: 16 November 2024
Browse Figures
Figure 1
<p>Theoretical framework of the integrated assessment of security risk.</p> "> Figure 2
<p>Study area. Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.</p> "> Figure 3
<p>Research process of the integrated assessment of security risk.</p> "> Figure 4
<p>Distribution characteristics of public premises and their security risks in JY City: (<b>a</b>) political units; (<b>b</b>) commercial establishments; (<b>c</b>) transportation hubs; (<b>d</b>) community facilities; (<b>e</b>) public premises.</p> "> Figure 5
<p>Distribution characteristics of the security situation of security risk in JY City (partial). Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.</p> "> Figure 6
<p>Distribution characteristics of the urban morphology and their security risk in JY City: (<b>a</b>) political units; (<b>b</b>) commercial establishments; (<b>c</b>) transportation hubs; (<b>d</b>) community facilities.</p> "> Figure 7
<p>Distribution characteristics of the police resources of security risk in JY City (partial). Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.</p> "> Figure 8
<p>Hierarchical distribution map of integrated security risk points in JY City. Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.</p> "> Figure 9
<p>Hierarchical distribution map of the integrated security risk areas in JY City. Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.</p> ">
Versions Notes

Abstract

:
The existing research on security risk often focuses on specific types of crime, overlooking an integrated assessment of security risk by leveraging existing police resources. Thus, we draw on crime geography theories, integrating public security business data, socioeconomic data, and spatial analysis techniques, to identify integrated risk points and areas by examining the distribution of police resources and related factors and their influence on security risk. The findings indicate that security risk areas encompass high-incidence areas of public security issues, locations with concentrations of dangerous individuals and key facilities, and regions with a limited police presence, characterized by dense populations, diverse urban functions, high crime probabilities, and inadequate supervision. While both police resources and security risk are concentrated in urban areas, the latter exhibits a more scattered distribution on the urban periphery, suggesting opportunities to optimize resource allocation by extending police coverage to risk hotspots lacking patrol stations. Notably, Level 1 security risk areas often coincide with areas lacking a police presence, underscoring the need for strategic resource allocation. By comprehensively assessing the impact of police resources and public security data on spatial risk distribution, this study provides valuable insights for public security management and police operations.

1. Introduction

Cities are confronting the dual challenge of escalating security risk and constrained police resources. The strategic allocation of police stations or patrols to finely delineated geographical units is imperative. Security risk is a collective term for a series of potential factors that may cause public security issues that endanger social order and public safety [1]. These issues include criminal incidents, public security mishaps, mass disruptions, sudden security events, and conflicts. Such concerns are a priority for police in their daily defense and control efforts. While the causes of public security incidents vary, they often share common drivers, including socioeconomic conditions, demographics, urban planning, and infrastructure. The primary purpose of security risk assessment is to evaluate the likelihood of such incidents occurring.
Precisely targeting police resources within areas of heightened security risk significantly reduces the risk of incidents escalating and helps mitigate major criminal cases and illegal activities. Police resources comprise staffing, equipment, training, and management [2], with staffing levels being particularly critical. However, China’s police-to-civilian ratio ranks at the middle to lower level globally, with its police force density notably lagging behind other countries. Meanwhile, China’s social security is among the highest in the world, which indicates that the per capita workload of Chinese police is greater than the world average. Due to factors such as institutional frameworks, prevention experiences, and resource allocation within law enforcement agencies, the spatial distribution of police resources often overlooks an adequate consideration of security risk. Consequently, limited police resources struggle to effectively prevent and control complex security risk issues. Optimizing police efficacy requires strategically deploying finite resources to security risk hotspots. This urgent priority addresses the tension between public security demands and limited police resources, necessitating a swift resolution.
Here, our aim is to identify spatial hotspots with a high security risk and insufficient police resources to offer law enforcement agencies a practical methodology to delineate targeted security risk areas, optimizing the allocation of police resources and interventions. This approach addresses the challenges of increasing security risk and limited police resources effectively.

2. Literature Review

2.1. Qualitative Research on Security Risk

Security risk, a concept within the realm of public security, has been subject to qualitative research methods primarily focusing on the development of a public security evaluation index system. This system entails a range of indicators that holistically depict the public security landscape, serving as a pivotal foundation for crime forecasting and public security management [3]. There is no consistent program for the design of public security assessment indicators at home or abroad. For example, in the main policing indicator system introduced in the United States, the first-level indicators are divided into five categories: crime and victims, public safety management and control, public perception, resource assurance, and international comparison. Japan’s policing assessment is based on four components: fire management, traffic safety management, security management, and crime situation. China’s Ministry of Public Security issued the public security industry standard “Evaluating Specifications for Social Security Warning Levels” in 2013, which categorizes the nature of policing indicators as changing trends, status, and control, while also taking into account the public’s tolerance of crime. By leveraging such an index system, dynamic evaluations of public security entities, processes, and outcomes can be conducted [4], facilitating early warning systems for security risk [5].

2.2. Quantitative Research on Security Risk

Quantitative research in the realm of crime geography constitutes a significant component of security risk analysis. Crime geography has emerged as a pivotal methodology for objectively delineating the state of security risk, discerning patterns of risk development and prognosticating future trends. Scholars in this field often initiate spatiotemporal investigations of security risk from diverse angles, typically leveraging crime occurrences as their focal point. Such research exhibits several key characteristics.
Uncovering Spatial Distribution Patterns: Some scholars delve into the spatial distribution patterns of crime, unveiling phenomena such as heightened robbery rates in residential areas, urban villages, and commercial zones [6]. Notably, locations like shopping malls, railway stations, hospitals, and residential communities are susceptible to theft [7]; criminals often target low-density residential communities for their illicit activities [8]. Drug-related crimes tend to concentrate in commercially developed areas and urban village regions [9], with the “Thunder Anti-drug” campaign shifting drug-dealing crimes from semi-public spaces to outdoor and private areas [10]. Correlations between establishments like hotels, KTVs, and foot massage parlors and sexual crimes have been observed [11].
Analyzing Environmental Characteristics: These studies scrutinize the characteristics of criminal environments, identifying factors such as the co-location of hospitals, fire stations, and areas with high ESL (locations where there are a greater number of people with English as a second language) populations that elevate the risk of robbery [12]. Positive correlations have been noted between the number of retail stores, street network density, and residential burglaries [13]. Additionally, the interplay between population mobility and housing types (e.g., self-built, affordable, and public housing) significantly influences burglary rates [14]. Amid the pandemic, changes in land-use types have emerged as a primary driver of crime rate fluctuations [15]. Risk Terrain Modeling (RTM) is a typical method for analyzing the relationship between crime and environment, allowing an assessment of the extent to which environmental factors contribute to crime risk [16,17]. There are quite a number of research results regarding the use of RTM to mine crime and environmental characteristics. For example, past homicides, drug trafficking, confiscated assets, and rivalries among groups create an environment conducive to mafia-related killings [18].
Introducing Novel Spatiotemporal Prediction Methods: Some researchers propose innovative approaches to crime prediction, often integrating multiple models or employing machine learning algorithms. For instance, some studies advocate for combining LSTM (long short-term memory network) and ST-GCN (spatial–temporal graph convolutional network) models for daily crime prediction [19], while others advocate for a crime-prediction model based on HDBSCAN (hierarchical density-based spatial clustering of applications with noise) and SARIMA (seasonal auto-regressive integrated moving average) approaches [20]. Machine learning algorithms such as support vector machines, logistic regression, and decision trees are also employed to enhance prediction accuracy, improve efficiency, and mitigate the impact of data sparsity and uncertainty on prediction outcomes [21,22,23,24,25,26,27,28].

2.3. Research on Security Risk Considering Police Resources

In recent years, the attention of scholars to police resources has steadily grown, leading to their inclusion in studies on security risk assessment. The public security organ, the functional departments of the Chinese Government, serving as a pivotal entity in upholding public order and executing prevention and control measures, has emerged as a focal point in this discourse [29]. Its spatial distribution is instrumental in mitigating social security risk. For instance, increased police resources through the Street Crime Initiative correlated with a notable decrease in robberies in the targeted areas [30]. The proximity to patrol stations serves as a metric for gauging accessibility and the level of criminal oversight, exerting a significant influence on nocturnal street crimes [31]. Diverse patrol strategies yield varying impacts on property crimes, with police car patrols, for instance, proving effective in curbing such offenses [32]. There is a scarcity of research on the spatial location of police institutions and the allocation of police force, such as utilizing the spatial heterogeneity of crime hotspots and building distribution to conduct research on optimizing the space of police stations [33]. Urban security extensively relies on closed-circuit television (CCTV), emerging as a cornerstone in crime prevention and control efforts [34]. The quadrant statistical method facilitates the analysis of crime displacement and the diffusion of benefits stemming from police CCTV surveillance based on crime data [35].

2.4. Research Gap

The existing research has established a framework for security risk and provided foundational references. However, it has three main shortcomings: (1) it lacks an integrated perspective that combines subjective policing experience and objective spatial distribution patterns to accurately assess the impact and scope of risk factors; (2) the research is predominantly focused on specific crime types, neglecting broader security risk, which limits an integrated perspective; (3) and there is insufficient study on the spatial distribution of police resources, with most recommendations stemming from theoretical analysis rather than systematic quantitative research. To address these issues, this paper integrates theories from crime geography and practical public security experience. We select risk indicators from multiple dimensions to construct an integrated security risk assessment index system. Using quantitative methods, this study determines the spatial impact range and degree of each indicator. Such an approach effectively identifies areas with high security risk and inadequate police resources, offering a scientific basis for optimizing police deployment. This research is helpful for shifting from reactive to proactive strategies in public security control, enhancing the modernization of social governance.

3. Theory Framework

Security risk encapsulates diversity, linkage, interweaving, and ambiguity [36]. The differentiation and interweaving of material space and social space aggravate the universality and complexity of the causes of security risk. The sources of security risk include personnel, goods, abnormal environments, and information and involve different elements of crime geography theory. Public safety theories on security risk are limited, whereas crime geography offers a more developed framework. Given the similarities between crime and security risk mechanisms, we propose creating a security risk indicator system based on crime geography theory. The crime pattern theory divides criminal activity into crime generators (places with many people, like schools and shopping centers) and crime attractors (hidden places like bars and banks) [17,37]. Social disorganization theory points to chaotic community environments as a cause of increased crime [38]. Routine activity theory emphasizes that crime occurs when motivated offenders, suitable targets, and the absence of capable guardians against crime converge in time and space [39]. The criminal near repeat principle refers to the increased likelihood of similar crimes occurring near the original crime location within a certain period [40]. The above theories have different emphases, but the causes of security risk are complex. Selecting just one theory makes it challenging to construct a framework that fully captures these characteristics. Therefore, this paper combined the core concepts of crime pattern theory, social disorganization theory, routine activity theory, and the criminal near repeat principle to build an integrated indicator system (Figure 1).
In China’s daily security control, the police pay attention to specific places, including political units, commercial establishments, transportation hubs, and community facilities. These areas, due to their political sensitivity or high population density, are more susceptible to security incidents. This paper uses these four types of places as measures of crime generators and crime attractors. In summary, political units include government agencies, foreign institutions, and social organizations; commercial establishments encompass landmark buildings, large supermarkets, cultural squares, and commercial pedestrian streets; transportation hubs consist of airports, train stations, passenger stations, and bus stops; and community facilities comprise hospitals, schools, and sports venues.
In terms of people management, police focus on floating populations and key people. Population instability, cultural heterogeneity, and income differences caused by the migration of floating populations are prone to bias or criminal behavior [41]. Therefore, floating population was selected as a measure of population mobility in the social disorganization theory. Key personnel include those prone to illegal activities, the more typical types being individuals struggling with substance abuse and those with psychopathic traits, represented in this paper as motivated offenders in the routine activity theory. Additionally, the nighttime light and impervious surface variables serve as indicators of economic development, reflecting regional economic disparities and urban spatial characteristics [10]. The likelihood of security incidents is greater in urban areas with higher levels of economic development.
According to the routine activity theory, public security organs and police facilities act as capable guardians against crime, while the population represents suitable targets. Public security organs are special organs tasked with ensuring societal safety and defense, equipped with a dedicated police force. Comprising various police departments and stations, these organs deter motivated offenders. Police facilities, including CCTV and traffic surveillance, refer to the various equipment used by police for upholding public order. These facilities play a crucial role in crime detection and social stability. The spatial distribution of the population influences the opportunity to commit offenses.
In addition, street crime is the focus of public security prevention and control [42]. Street crime, which occurs in public areas like urban transport, streets, parks, and markets, includes offenses such as robbery, theft, and assault [43]. These open spaces facilitate both the commission of and escape from crimes, making the streets a focal point for public security issues. Police resources, such as patrol officers, are crucial in mitigating street crime through rapid response, investigation, and control measures. Based on the criminal near repeat principle, we use street crime as an example of public security issues to measure security risk in historical public security issue locations.
These indicators are potential drivers of public safety issues and are categorized into four dimensions: public premises, security situation, police resources, and urban morphology. Together, they constitute an integrated security risk indicator system that provides a macro-level overview, supporting police strategists and managers in assessing broader security risk.

4. Study Area, Data, and Methods

4.1. Study Area

The study area of this research is JY City (Figure 2), located in the southeast of mainland China, which has 2 municipal districts and 2 counties and manages 1 county-level city. According to the 2023 Statistical Yearbook of JY City, the total land area of the city is 5265.84 km2, with a permanent population of 5.65 million and a total economic production of approximately CNY 244.50 billion, an increase of 7.50 percent year-on-year. Statistically, the ratio of police to citizens in JY City is only 6:10,000. Currently, JY City is in a period of rapid economic development, but police resources are relatively scarce. Therefore, choosing JY City as the research area is considered representative for the aims of this study.

4.2. Data and Processing

The research data utilized in this paper comprise micro-assessment, macro-assessment, and basic geographic data. The micro-assessment data encompass point of interest (POI), street crime, floating population, key population, and police facility measures. These datasets consist of point data with latitude and longitude values. Geographical coordinates were assigned to the micro-assessment data, with manual corrections conducted to eliminate entries outside the study area or with incomplete information. The macro-assessment data encompass population distribution, nighttime light intensity, and land use patterns, all of which are in raster format. We used the ‘Raster Resampling’ tool in ArcGIS to process the above three variables to unify their raster size. The basic geographic data consist of administrative boundary delineations and road networks. The details of the data are shown in Table 1.

4.3. Methods

To investigate the influence of various indicators on security risk, this study employed a grid method to analyze indices within the study area at the microscale. Following the norms of daily public security prevention and control and drawing from the existing research, a 100 m × 100 m grid was adopted as the fundamental unit for an integrated assessment of security risk. Taking into account the significance of each indicator and its spatial influence, an integrated security risk map was generated. Subsequently, a hotspot detection model was applied to extract security risk points and generate security risk areas. The research process is illustrated in Figure 3.

4.3.1. Indicator Weight Calculation

Considering the specificities of public security prevention and control in JY City, this study combined the analytic hierarchy process (AHP) and entropy weight method (EWM) approaches to determine the weights of each measurement indicator.
(1)
Analytic Hierarchy Process
AHP, a method of multi-factor decision analysis, combines both qualitative and quantitative analysis. In this study, we employed AHP to facilitate the quantification of empirical judgments from field practitioners and criminal geography experts. AHP was utilized to determine the weights of the three representation dimensions of public premises, security situation, and urban morphology and the categories of political units, commercial establishments, transportation hubs, and community facilities in public premises. The main steps in the calculation of indicator weights using the AHP method are as follows:
(i)
Constructing the hierarchical model: the security risk, representation dimensions, and measurement indicators shown in Figure 1 correspond to the target, criteria, and indicator layers of the hierarchical model, respectively.
(ii)
Constructing the judgment matrix: Scoring scales were designed, and practitioners and experts scored the security risk indicators using two-by-two comparisons. The scoring scales were recovered and collated into a judgment matrix.
(iii)
Consistency test: The consistency of the judgment matrix was evaluated by calculating the CI (consistency index) and CR (consistency ratio) in Matlab. The smaller the CI, the greater the consistency. Meanwhile, if CR < 0.1, the judgment matrix was considered to have passed the consistency test.
(iv)
Calculating the weights: weights were calculated in Matlab to determine the weights of the risk indicators in each layer.
(2)
Entropy Weight Method
EWM determines objective weights based on indicator variability. A greater indicator dispersion indicates a larger impact on the comprehensive evaluation. In this study, EWM was employed to determine the weights of secondary indicators in the security situation, urban morphology and police resources variables. All the steps were performed in Matlab except for the first. The details are as follows:
(i)
Indicator layer calculation: the research area was divided into 1000 m × 1000 m grids, and the sum of the calculation indices or grid values in each grid was determined;
(ii)
Standardization: the values of indicators were standardized to eliminate the effect of scale;
(iii)
Information entropy computation: the smaller the entropy value, the greater the amount of valid information provided by the indicator and the greater the weight;
(iv)
Weight assignment: the weight of each indicator was calculated based on the entropy value.

4.3.2. Kernel Density Estimation

KDE serves as a two-dimensional visualization tool for understanding point distribution patterns and is widely utilized in predictive crime mapping within criminal geography [44]. By applying KDE to street crime, floating population, and key personnel, we can discern the spatial distribution trends of their associated risks. The calculation is represented as follows:
F x = 1 n τ ∑ i = 1 n K x − x i τ
where n represents the number of samples within the bandwidth range; x − x i denotes the distance from the estimated point x to the sample point x i ; and τ refers to the bandwidth. Given the scope of the activities of key personnel and the heterogeneity of the spatial distribution among different indicators, the bandwidth range should encompass various indicators as comprehensively as possible. Following experimental analysis, a bandwidth of 500 m was deemed to be appropriate.
Conventional KDE treats all point data with the same density distribution within the bandwidth range, thereby failing to reflect differences in their impact intensity [45]. Weighted kernel density addresses this limitation. The weighted KDE of public premises unveiled spatial disparities in the impact of different types and levels of public premises on security risk. The calculation is outlined below:
D x = ω · 3 π τ 2 ∑ i = 1 n 1 − x − x i τ 2 2
where n , x − x i , and τ retain their previous interpretations, and ω represents the weight coefficient.

4.3.3. Density-Field-Based Hotspot Detector (DF-HD)

The integrated security risk grid, derived from the synthesis of the four dimensions of the security risk grid surface, offers a summary-level representation of the spatial distribution characteristics of integrated security risk. However, pinpointing the exact peak values and areas of risk points remains challenging. To address this, the DF-HD proposed by Zhang Haiping [46,47] was introduced in this study to detect extreme points on the integrated security risk surface. The specific process is delineated in Figure 3.
Initially, focus statistics were conducted on the integrated risk grid of public security, identifying the maximum risk value within the neighborhood to generate the domain extremum surface. Subsequently, this extremum surface was subtracted from the original security integrated risk grid, yielding a nonnegative surface where pixels with a value of 0 denote the location of the extremum point. Zero-value pixels were then extracted using the reclassification method. Following pixel transfer, risk values corresponding to the integrated security risk grid were assigned. Ultimately, all risk points were graded based on the magnitude of the risk value.

5. Results and Analysis

5.1. Determination of the Weight of Security Risk Indicators

This study acknowledges the deterrent effect of public security organs and police facilities on security risk. Areas near existing police resources were deemed redundant for security control, resulting in assigning negative weights to police resources and positive weights to other factors. Following previous work [48], weight proportions were set with a total positive weight of 1.1 and a total negative weight of −0.1. Table 2 presents the calculated weights for security risk indices, illustrating their significance in influencing social order and reflecting the level of risk.
In the calculation results of AHP, the CI value is 0.042, and the CR value is 0.080, computed for the third-order matrix. For the fourth-order matrix, the CI value is 0.022, and the CR value is 0.025. Both CR values fall below 1, indicating that the judgment matrices in this study pass the consistency test.
Upon examining the weight calculations, it is evident that the security situation poses the highest security risk in the criteria layer, weighing 0.504, followed by public premises at 0.179. The security situation directly correlates with security risk, serving as a key metric. Public premises, as persistent risk factors, serve as supplementary indicators. Urban morphology provides a contextual backdrop for security risk assessments at the macro-level. Within the indicator layer, the top influences on security risk include public security issues, key personnel, floating population, political units, and transportation hubs.
In public premises, political units rank highest in the category weights, followed by transportation hubs, commercial establishments, and community facilities. Political units are highly sensitive politically, with security incidents here having significant societal repercussions. Transportation hubs pose a heightened security risk due to their diverse populations and essential role in city transportation. Commercial establishments and community facilities, with their dense populations, increase the likelihood of public security issues. Public security issues hold the highest weight within the security situation, underscoring their critical role in mitigating security risk by addressing unlawful activities such as street crime. Key personnel concentration and floating population positively correlate with crime rates. In urban morphology, population distribution and impervious surfaces share similar weights, while nighttime light has a comparatively lower weight, likely due to varying data resolutions. Regarding police resources, centralized police forces within public security organs exert a more significant deterrent effect on security risk compared to facilities like CCTV and traffic surveillance. As a result, they carry a greater weight in assessments than the latter.

5.2. Single Criteria Assessment Analysis of Security Risk

5.2.1. Analysis of Security Risk in Public Premises

The spatial distribution of various public premises is depicted in Figure 4, revealing consistent and stable trends of concentration in both the central and suburban areas of JY City. These hotspots predominantly cluster within the main urban area (B4–C4, E5–E6) and its suburban extensions (E8, G3–H3), with fewer occurrences in peripheral zones. Political units exhibit spatial clustering, primarily concentrated in the northwest quadrant of the study area (Figure 4a). Similarly, commercial establishments mirror this distribution pattern, displaying pronounced aggregation characteristics (Figure 4b). Transportation hubs, however, diverge outward from the urban center, displaying distinct banding features along the periphery (Figure 4c). Community facilities share spatial distribution traits akin to political units and commercial establishments, albeit with a more diffuse spread (Figure 4d). While public premises collectively exhibit a propensity to aggregate in urban zones, there exists a notable spatial heterogeneity.
Utilizing the category weights of public premises, weighted KDE was employed for spatial analysis, normalizing the density grid to generate the security risk grid map of public premises (Figure 4e). Overall, public premises demonstrate evident spatial aggregation tendencies, with high nuclear-density areas primarily concentrated in the main urban zone. These areas form a “high–low–high–low–high” circular distribution radiating from the urban center toward the northwest–southeast axis. Featuring a multi-center configuration, the main urban and suburban areas host four prominent hotspots. Additionally, scattered high kernel density regions align along the major urban thoroughfares. High-scoring areas signify not only a substantial quantity and importance of public premises but also serve as focal points for political, economic, and cultural activities. Consequently, the flow and congregation of people, resources, and commerce within these hubs engender a heightened security risk. For instance, political centers may experience conflicts detrimental to social stability, while economic hubs are susceptible to property crimes and financial disputes. Moreover, cultural centers, with their high accessibility, population density, and variable security supervision, are prone to theft and other offenses.
Conversely, smaller hotspots predominantly located in suburban, town, or central village areas often rely on main transportation arteries or intersections for their formation. These sporadic locales, characterized by lower road and population densities, exhibit a less pronounced security risk.

5.2.2. Analysis of Security Risk in Security Situation

KDE analysis was conducted on public security issues, floating population, and key personnel. The resultant density grids were weighted and aggregated based on the weights assigned to their respective indicators. Following normalization, the grid map of the security situation security risk was derived. As this part of the data is confidential and not easy to show globally, Figure 5 showcases a representative area selected for detailed analysis.
Upon inspection, it is evident that regions exhibiting high security situation densities predominantly cluster within the urban expanse, displaying a gradient decline from the urban core toward the periphery. Additionally, scattered hotspots are observed around the outskirts of the urban area. In terms of numerical values, the overall security risk values pertaining to the security situation dimension tend to be relatively low. Specifically, the risk value spans from 1 to 0.353 for the first level and 0.353 to 0.173 for the second level, with the coverage of areas within these levels constituting only a minimal portion. This indicates that these areas are likely to have a concentration of public security issues, floating population, and key personnel at the same time and also underscores a significant spatial heterogeneity in the distribution of these three risk indicators.
Regions characterized by frequent public security incidents such as street crime, as well as those with heightened concentrations of floating population and key personnel, often exhibit an elevated criminal and security risk. Consequently, these areas necessitate a focused attention in daily security prevention and control efforts. Allocating police resources to such high-risk zones with elevated security situations holds the potential to substantially mitigate security risk.

5.2.3. Analysis of Security Risk in Urban Morphology

Under the dimension of urban morphology, Figure 6a–c illustrates the distribution of population, nighttime light, and impervious surface, respectively. These three measurement indicators were weighted and summed according to their respective weights, yielding the urban morphology security risk grid map after normalization (Figure 6d). Overall, the spatial distribution pattern of population, nighttime light, and impermeable surface exhibits a consistent trend, characterized by aggregation in urban and peripheral areas, with a gradual decrease from the center outward.
Population distribution hotspots appear relatively diffuse (Figure 6a), reflecting the complexity of social relations and the heightened likelihood of disputes in densely populated areas, thereby amplifying the occurrence of public security issues. Nighttime light delineates urban outlines more distinctly (Figure 6b), while the distribution of impermeable surfaces displays fragmentation characteristics (Figure 6c). The higher scores for impermeable surfaces and nighttime light indicate a greater urban development and commercial activity, thereby providing increased space and opportunities for activities that disrupt social order.
The distribution of urban-scale security risk integrates the spatial characteristics of the aforementioned three indicators (Figure 6d). It primarily manifests two major hotspots and several sporadic ones. The broadest range of hotspots is concentrated in B4–C4, the primary urban area of JY City, with risk values gradually diminishing outward from the central urban core. The second-largest hotspot is situated in E5–E6, gradually merging with hotspot B4–C4. Sporadic hotspots are primarily distributed in the southwest and east of the study area.
Urban morphology, comprising population distribution, nighttime light, and impervious surface, dictates a varied security risk across different urban spatial configurations. Crimes, public security incidents, residential activities, and police prevention and control are inherently linked to the overall constraints, regular influence, and basic spatial characteristics of urban morphology. Urban areas characterized by dense populations and high economic development exhibit a heightened probability and risk of public security incidents, necessitating enhanced daily public security prevention and control measures in these regions.

5.2.4. Analysis of Security Risk in Police Resources

The analysis conducted involved buffer zone assessments based on the categories of public security organs and police facilities. The buffer zone delineation signifies the extent of policing supervision, reflecting the degree of deterrence exerted over potential offenders and the control of emergencies. Specifically, the buffer ranges for each indicator category were established as follows: 300 m from the public security bureau, 200 m from police stations, and 100 m from police offices, CCTV, and traffic surveillance points. After converting these buffer zone ranges into a grid format, the weighted aggregation of each indicator weight yielded a grid map representing the police resources of security risk. As with Figure 5, this part of the data is sensitive, so a representative region was selected for a zoom-in analysis (Figure 7). As illustrated in Figure 7, in the visualization, darker shades of blue denote a lower security risk within the covered area.
Broadly, the spatial distribution of police resources is more fragmented than the indicators in the other criteria. Among them, various public security organs predominantly inhabit urban zones, displaying notable clustering tendencies, whereas police facilities exhibit a comparatively scattered distribution. Areas encompassed by police resources pose a heightened risk and cost for potential criminals or troublemakers, thereby exerting a reduction effect on public security incidents. When establishing daily security prevention and control checkpoints, it is imperative to avoid redundant deployment in these zones to prevent wastage of police resources.
As urban development progresses, buildings and thoroughfares become increasingly dense. This evolution may inadvertently create blind spots in public security control, fostering spaces conducive to illegal activities or providing refuge for perpetrators post-offense, as illustrated by the light blue area in Figure 7. Such areas are prone to frequent illegal activities due to inadequate public security oversight, underscoring the need for the judicious allocation of police resources commensurate with the prevailing security risk level.

5.3. Integrated Assessment and Analysis of Security Risk

5.3.1. The Spatial Distribution of Security Risk Points

Based on the weights assigned to the criteria layer, the score grids of the four dimensions were aggregated and weighted to generate an integrated grid map of security risk, encompassing a range of integrated risk values from 0.6658 to −0.0992 (Figure 8). Following extraction by the DF-HD method, security risk points were categorized into four grades based on their scores, ranging from high to low, corresponding to four distinct colors: red, orange, yellow, and blue. Table 3 outlines the number, proportion, and risk value range of the security risk points at each level.
To elucidate the distribution of security risk across different high-value regions, this study zooms in on four representative areas: main urban area 1 (C3–C4), main urban area 2 (E6–F5), sub-urban area 1 (E9), and sub-urban area 2 (H3).
Observing Figure 8, the spatial distribution of security risk points exhibits a multi-center circular structure overall, with a concentration of points in the central zones of JY City’s urban areas. These areas are characterized by robust economic development, high road density, extensive road network accessibility, well-developed urban amenities, diverse urban functions, and dense pedestrian flow. However, the uneven distribution of public security measures and intensity contributes to a high security risk, facilitated by chaotic road traffic and pedestrian flow, providing cover and refuge for potential offenders and troublemakers.
Furthermore, the different levels of security risk points display distinct spatial differentiation characteristics on the map. Within each security risk hotspot, security risk points are distributed in a gradient manner from the center outward and from high to low levels. As indicated in Table 3, Level 1 security risk points, totaling 15.49%, are concentrated in bustling urban areas with high pedestrian flow, influenced by dimensions including public premises, security situation, urban morphology, and police resources. Level 2 security risk points, numbering 14 and accounting for 19.72%, are distributed in proximity to Level 1 security risk points. Level 3 and Level 4 security risk points exhibit a relatively scattered distribution on the periphery of Level 1 and Level 2 security risk points.
The uneven distribution of security risk across the city reflects the differentiation of the urban spatial environment and economic development. Comparing the risk value intervals of different security risk points reveals a diminishing trend as the level decreases, indicating significant disparities between security risk levels. This underscores both a risk hierarchical formation trend and the rationality of security risk point extraction through integrated indicators. Deploying corresponding scales of prevention and control points and police resources at security risk points can effectively combat and deter security incidents and illegal activities, thereby enhancing the security level of local risk areas and improving citizens’ sense of security and satisfaction.

5.3.2. The Spatial Distribution of Security Risk Areas

Based on daily public security prevention and control response times and road accessibility, security risk areas were delineated for each level of security risk points, utilizing the road network (Figure 9). The coloration of the security risk areas in the figure corresponds to the security risk points depicted in Figure 8. The service radius for Level 1 security risk points is 300 m, achievable by a minute’s walk; Level 2 security risk points extend to a 1000 m radius, accessible within a 3 min urban drive; Level 3 security risk points encompass a 2500 m radius, reachable within a 5 min drive; and Level 4 security risk points cover a 5000 m radius, representing the longest distance for daily patrols.
Table 4 illustrates the proportion of public security organs and police facilities within each level of the security risk area. Broadly, as the security risk area level decreases, the proportion of police resources increases. In Level 1 security risk areas, police resources only account for 1.26% of the city, with public security organs representing 2.12% and police facilities 1.24%. Conversely, in the remaining areas, police resources constitute a substantial 65.77%, with public security organs and police facilities accounting for 43.22% and 66.07%, respectively. This indicates a considerable consumption of police resources in low-risk or risk-free areas, necessitating an urgent reallocation toward security risk areas to address resource scarcity and security risk discrepancies.
Observing Figure 9, each level of the security risk areas achieves a comprehensive coverage of areas prone to security risk hazards, reflecting JY City’s urban spatial development pattern. Within security risk areas, crimes, disturbances, and other emergencies impacting urban public security and social order may proliferate. Notably, significant differences exist in the predominant types of establishments within the security risk areas at different levels. Level 1 security risk areas tend to encompass or be closely located to government buildings while avoiding the PSB, owing to the political sensitivity of government facilities. Meanwhile, Level 2 security risk areas extend their coverage to hospitals, and Level 3 security risk areas expand to include schools. These locations, characterized by high pedestrian traffic and insufficient crowd supervision, coupled with adjacent amenities like restaurants and supermarkets, are attractive targets for theft and similar incidents, necessitating strengthened daily patrol and prevention measures.
Passenger stations, airports, and train stations primarily fall within Level 3 and Level 4 security risk areas, often located in peripheral regions with a relatively underdeveloped infrastructure and lower security risk. The categorization of security risk areas facilitates a responsive police force capable of promptly addressing security issues and reaching incident scenes efficiently, effectively combating and deterring security incidents and criminal activities and enhancing the city’s overall public security situation.

6. Discussion

6.1. Composition of Security Risk Hotspots

A security risk area is defined by a high public security issue incidence, concentrations of dangerous individuals, key unit clusters, and police resource limitations. Public security issues carry the highest weight in the indicator layer, indicating a heightened risk in historical incident hotspots, consistent with the existing research [6,49]. These incidents are direct manifestations of security risk, and security risk assessments aim to mitigate their occurrence and minimize their impact. The distribution of key personnel and their daily activities significantly explain security risk [50], while the impact of the floating population, although smaller, remains noteworthy. Individuals struggling with substance abuse and those with psychopathic traits pose an added risk to public safety, as they may be unable to control their behavior, potentially threatening the community. The floating population, often consisting of rural-to-urban migrant workers without urban residency benefits, tends to congregate in areas with lower socioeconomic status and poor living conditions, which increases their vulnerability to both victimization and criminal activity [51].
Security risk is influenced by time, space, population dynamics, and socioeconomic development. In addition to the security situation and police resources, other indicators play supportive roles. Political units and transportation hubs increase policing complexities due to large crowds and traffic. While indicators like commercial establishments, community facilities, economic development, and urban areas have a minimal direct impact, they contribute positively to overall security risk [7,9,12,13]. Areas with advanced urban infrastructure and high levels of economic development tend to increase the volatility of social security, as the movement and concentration of people, property, and goods heighten the risk. The interaction of various risk factors and limited police oversight can escalate public security issues.
The weights assigned to public security organs and police facilities (−0.080 and −0.020, respectively) are relatively small based on the established literature [48], policing practitioners, and crime geography experts, suggesting the need for further exploration. Incorporating police resources helps avoid high-risk areas already covered by police efforts.

6.2. Comparison of the Spatial Distribution of Security Risk and Police Resources

Both police resources and security risk are concentrated in urban areas, though security risk is more sporadic at the urban periphery. As noted by the existing research [31], areas farther from police stations present a weaker deterrence, leading to increased crime and a higher security risk. Public security organs, mainly in densely populated urban zones, maintain public security. Video surveillance helps enhance security by reducing crime opportunities [35], with greater deployment in areas of high economic and social activity, such as malls, hospitals, and schools. The spatial distribution of police resources aligns with the security risk patterns, deterring crime and reducing public security issues.
Relevant studies show high-risk areas are often located in commercial zones and urban villages [6,9], confirming this study’s evaluation method. An identified new trend is the significant security risk found at the urban–rural fringe. These fringes, with heavy traffic and diverse populations, face challenges in security coverage and become hotspots for unlawful activities due to insufficient measures. The distribution of security risk often exceeds that of police resources, necessitating timely adjustments in police resource allocation to address fluctuating risk.

6.3. The Spatial Relationship Between Security Risk and Police Resources

The security risk areas are primarily in urban zones, with higher-risk areas strategically avoiding primary public security organ locations. As risk levels decrease, more police resources are allocated to these areas. While the size of security risk areas influences police resource distribution, lower-risk areas still receive the majority of resources. Research shows that security risk declines from city centers outward [6,7,9,14,31,50]. The hotspots identified here align with previous studies but differ from some that suggest a negative correlation between crime and the proximity to patrol points [24,52,53]. This discrepancy may be due to public security organs being located in high-activity areas or possible flaws in crime supervision assessments. This paper develops a security risk index incorporating police resources, addressing various risk factors and their impact. As a result, the identified security risk areas avoid primary public security organ locations, avoiding conflicting conclusions.

6.4. Expectations and Deficiencies

This paper addresses the limitations in the existing research and offers key contributions. We integrate the proven crime risk factors with police resources to create an integrated indicator system for assessing security risk, accounting for both the characteristics and mechanisms of public security incidents. This system aids in the development of cross-sectoral prevention strategies and optimizing resource allocation. Additionally, by combining the expertise of public security practitioners with the spatial patterns of objective indicators, we enhance the rigor, interpretability, and practical relevance of our approach. Our methodology has been recognized by police departments and crime geography experts, and its findings are already being applied in practice.
Our results herein offer valuable insights for police daily security deployment by recommending improved supervision and resource allocation, such as increased video surveillance, new police stations, and proactive patrols in high-risk areas. This approach aims to strengthen public security and deter public security issues. However, this study has some limitations. Currently, there is no clear evidence confirming that the selected indicators are directly linked to security risk. Future research should explore additional indicators and quantitatively analyze their correlation with security risk. Moreover, using data from the second year on public security incidents, such as crimes, conflicts, and security emergencies, to assess whether the results contribute to reducing incidents and improving policing efficiency would further enhance the scientific rigor and practical applicability of this methodology.

7. Conclusions

Drawing on crime pattern theory, social disorganization theory, routine activity theory, and the criminal near repeat principle, this paper integrates police resources into an integrated evaluation index system for security risk on the basis of the existing research. Through methods such as AHP, EWM, KDE, weighted KDE, and buffer zone analysis, this study focuses on JY City to conduct a spatial analysis of security risk factors. Utilizing DF-HD, security risk points are extracted, and security risk areas are delineated based on the road network, revealing the spatial distribution pattern of security risk points and areas. The key findings include the following:
(1) The top five influential indicators of security risk encompass public security issues, key personnel, floating population, political units, and transportation hubs. These factors collectively contribute to security risk areas characterized by dense populations, diverse urban functions, high probabilities of illegal activities, economic development, and inadequate supervision, emphasizing the need for enhanced security measures.
(2) Both police resources and security risk exhibit concentrations in urban areas, with decreasing aggregation toward the urban periphery. However, security risk displays a more scattered distribution around the outskirts, highlighting the importance of extending police resources to hotspots lacking patrol stations to effectively combat public security issues and optimize resource utilization.
(3) Security risk points at different levels exhibit a concentric spatial distribution pattern, gradually decreasing from the center outward. While security risk areas largely cover urban zones, Level 1 security risk areas notably lack a public security organ distribution. Strategic police resource allocation based on security risk levels can facilitate the establishment of rapid response mechanisms and enhance overall security and public confidence.
This integrated assessment of security risk considering police resources enables a holistic understanding of spatial security risk distribution, guiding targeted measures for police prevention and control to mitigate security incidents and criminal activities. The proposed evaluation index system and methodology offer valuable insights applicable to space–time–based security risk research across various scales and contexts. The categories and weights of the indicators at each layer and the parameters in the experimental process can be adjusted according to the actual situation.

Author Contributions

Conceptualization, J.C., Y.C. and W.L.; methodology, J.C. and Y.C.; software, J.C.; validation, W.L.; formal analysis, J.C.; investigation, J.C., Y.C. and W.L.; resources, J.C. and W.L.; data curation, J.C.; writing--original draft preparation, J.C.; writing--review and editing, J.C., Y.C. and Y.L.; visualization, J.C.; supervision, Y.C. and W.L.; project administration, Y.C. and W.L.; funding acquisition, Y.C. and W.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Guangdong Province Marine Economic Development (Six Major Marine Industries) Special Fund Project, grant number [GDNRC[2023]25], the Ministry of Public Security Science and Technology Project, grant number [2022JSYJD05], the Key-Area Research and Development Program of Guangdong Province, grant number [2020B0101130002], and the Guangzhou Municipal Science and Technology Program, grant number [2024B03J1377].

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Acknowledgments

The authors would like to thank the China Centre for JY City Public Security Bureau for providing us with public security business data. The authors also thank the Department of Public Security of Guangdong Province for providing valuable suggestions in this research.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Gong, Z. Zhi’anxue Daolun; China Renmin University Press: Beijing, China, 2015. [Google Scholar]
  2. Huang, X. The Scientific Allocation and Use of Police Resources. Polic. Stud. 2009, 2, 64–74. [Google Scholar] [CrossRef]
  3. Shi, J. Research on the Historical Evolution and Changing Rules of Public Order Evaluation Index System. Master’s Thesis, People’s Public Security University of China, Beijing, China, 2020. [Google Scholar]
  4. Shi, Q. Index System Research on Grassroots Social Public Order Governance. J. Zhejiang Police Coll. 2021, 1, 47–53. [Google Scholar]
  5. Peng, Y. Study on Intelligent Prevention of Community Security Risks. Master’s Thesis, People’s Public Security University of China, Beijing, China, 2020. [Google Scholar]
  6. Guo, Y.; Chen, P. Formation and Feature Analyses of Crime Hotspots Using Near Repeat Principle: A Case Study of Robbery in Beijing. Prog. Geogr. 2020, 39, 804–814. [Google Scholar] [CrossRef]
  7. Zheng, Z.; Jiang, C.; Wang, J.; Liu, L.; Chen, P.; Dong, Q. Spatiotemporal Evolution of Urban Theft Crimes and Mechanism in the Context of Regular COVID-19 Pandemic Prevention and Control: A Case Study of Haining, Zhejiang. Prog. Geogr. 2023, 42, 341–352. [Google Scholar] [CrossRef]
  8. Vandeviver, C.; Bernasco, W. “Location, Location, Location”: Effects of Neighborhood and House Attributes on Burglars’ Target Selection. J. Quant. Criminol. 2020, 36, 779–821. [Google Scholar] [CrossRef]
  9. Liu, Y.; Li, W.; Wang, X. Spatiotemporal Distribution of Drug Crimes at Micro-Scale: Taking NH and DM Residential Communities in SZ City as An Example. Prog. Geogr. 2020, 39, 841–852. [Google Scholar] [CrossRef]
  10. Liu, L.; Liu, H.; Chen, J.; Xiao, L.; Zhu, W.; Sun, Q. The Impact of “Thunder Anti-Drug” Operation on Drug Dealing Crime: A Case Study of the Main Urban Area of ZG City. Acta Geogr. Sin. 2022, 77, 1461–1474. [Google Scholar]
  11. Zheng, Z.; Lin, Z.; Jiang, C.; Chen, P.; Zhou, J.; Liu, L. Comparing Spatial-Temporal Patterns and Influencing Factors of Rape and Indecency Crime in Haining City, Zhejiang Province. Sci. Geogr. Sin. 2023, 43, 2208–2216. [Google Scholar] [CrossRef]
  12. He, Z.; Deng, M.; Xie, Z.; Wu, L.; Chen, Z.; Pei, T. Discovering the Joint Influence of Urban Facilities on Crime Occurrence Using Spatial Co-Location Pattern Mining. Cities 2020, 99, 102612. [Google Scholar] [CrossRef]
  13. Mahfoud, M.; Bernasco, W.; Bhulai, S.; van der Mei, R. Forecasting Spatio-Temporal Variation in Residential Burglary with the Integrated Laplace Approximation Framework: Effects of Crime Generators, Street Networks, and Prior Crimes. J. Quant. Criminol. 2021, 37, 835–862. [Google Scholar] [CrossRef]
  14. Song, G.; Li, X.; Xiao, L.; Liu, L.; Chen, J. Effects of Population Mobility and Housing Type on Spatial Pattern of Urban Burglary from the Perspective of Interaction. Geogr. Res. 2022, 41, 2897–2911. [Google Scholar]
  15. Wu, Y.; Li, Y. “Hot Street” of Crime Detection in London Borough and Lockdown Impacts. GEO-Spat. Inf. Sci. 2023, 26, 716–732. [Google Scholar] [CrossRef]
  16. Caplan, J.M.; Kennedy, L.W.; Drawve, G.; Baughman, J.H. Data-Informed and Place-Based Violent Crime Prevention: The Kansas City, Missouri Risk-Based Policing Initiative. Police Q. 2021, 24, 438–464. [Google Scholar] [CrossRef]
  17. Kinney, J.B.; Brantingham, P.L.; Wuschke, K.; Kirk, M.G.; Brantingham, P.J. Crime Attractors, Generators and Detractors: Land Use and Urban Crime Opportunities. Built Env. 2008, 34, 62–74. [Google Scholar] [CrossRef]
  18. Dugato, M.; Calderoni, F.; Berlusconi, G. Forecasting Organized Crime Homicides: Risk Terrain Modeling of Camorra Violence in Naples, Italy. J. Interpers. Violence 2020, 35, 4013–4039. [Google Scholar] [CrossRef]
  19. Han, X.; Hu, X.; Wu, H.; Shen, B.; Wu, J. Risk Prediction of Theft Crimes in Urban Communities: An Integrated Model of LSTM and ST-GCN. IEEE Access 2020, 8, 217222–217230. [Google Scholar] [CrossRef]
  20. Butt, U.M.; Letchmunan, S.; Hassan, F.H.; Ali, M.; Baqir, A.; Koh, T.W.; Sherazi, H.H.R. Spatio-Temporal Crime Predictions by Leveraging Artificial Intelligence for Citizens Security in Smart Cities. IEEE Access 2021, 9, 47516–47529. [Google Scholar] [CrossRef]
  21. Bilen, A.; Ozer, A.B. Cyber-Attack Method and Perpetrator Prediction Using Machine Learning Algorithms. PeerJ Comput. Sci. 2021, 7, e475. [Google Scholar] [CrossRef]
  22. Cichosz, P. Urban Crime Risk Prediction Using Point of Interest Data. ISPRS Int. J. GEO-Inf. 2020, 9, 459. [Google Scholar] [CrossRef]
  23. Aldossari, B.S.; Alqahtani, F.M.; Alshahrani, N.S.; Alhammam, M.M.; Alzamanan, R.M.; Aslam, N.; Irfanullah. A Comparative Study of Decision Tree and Naive Bayes Machine Learning Model for Crime Category Prediction in Chicago. In Proceedings of the 2020 6th International Conference on Computing and Data Engineering, Deqing, China, 24–25 July 2020; ACM: Sanya, China, 2020; pp. 34–38. [Google Scholar]
  24. Zhang, X.; Liu, L.; Lan, M.; Song, G.; Xiao, L.; Chen, J. Interpretable Machine Learning Models for Crime Prediction. Comput. Environ. Urban Syst. 2022, 94, 101789. [Google Scholar] [CrossRef]
  25. Wheeler, A.P.; Steenbeek, W. Mapping the Risk Terrain for Crime Using Machine Learning. J. Quant. Criminol. 2021, 37, 445–480. [Google Scholar] [CrossRef]
  26. Lamari, Y.; Freskura, B.; Abdessamad, A.; Eichberg, S.; de Bonviller, S. Predicting Spatial Crime Occurrences through an Efficient Ensemble-Learning Model. ISPRS Int. J. GEO-Inf. 2020, 9, 645. [Google Scholar] [CrossRef]
  27. Zhang, Y.; Siriaraya, P.; Kawai, Y.; Jatowt, A. Predicting Time and Location of Future Crimes with Recommendation Methods. Knowl.-Based Syst. 2020, 210, 106503. [Google Scholar] [CrossRef]
  28. Liu, E.; Chen, K.; Shi, Y.; Deng, M. A Hot Spot Detection Method of Criminal Events under Data Uncertainty. Geomat. Inf. Sci. Wuhan Univ. 2023, 1–14. [Google Scholar] [CrossRef]
  29. Gong, Z.; Li, X. Research on Several Basic Issues of Social Security Prevention and Control System. J. People’s Public Secur. Univ. China Soc. Sci. Ed. 2014, 30, 107–115. [Google Scholar]
  30. Machin, S.; Marie, O. Crime and Police Resources: The Street Crime Initiative. J. Eur. Econ. Assoc. 2011, 9, 678–701. [Google Scholar] [CrossRef]
  31. Liu, L.; Liang, S.; Song, G. Explaining Street Contact Crime Based on Dynamic Spatio-Temporal Distribution of Potential Targets. J. Geo-Inf. Sci. 2020, 22, 887–897. [Google Scholar]
  32. Ratcliffe, J.H.; Taylor, R.B.; Askey, A.P.; Thomas, K.; Grasso, J.; Bethel, K.J.; Fisher, R.; Koehnlein, J. The Philadelphia Predictive Policing Experiment. J. Exp. Criminol. 2021, 17, 15–41. [Google Scholar] [CrossRef]
  33. Sun, L.; Wang, Z.; Sun, L.; Xu, Z.; Zhu, Y.; Wang, R. Spatial Location Optimization of Urban Police Stations and Police Allocation. J. Geo-Inf. Sci. 2019, 21, 346–358. [Google Scholar]
  34. Circo, G.; McGarrell, E. Estimating the Impact of an Integrated CCTV Program on Crime. J. Exp. Criminol. 2021, 17, 129–150. [Google Scholar] [CrossRef]
  35. Liu, L.; Li, L.; Zhou, H.; Jiang, C.; Lan, M. The Effects of Police CCTV Camera on Crime Displacement and Diffusion of Benefits: A Case Study from Gusu District in Suzhou, China. Sci. Geogr. Sin. 2020, 40, 1601–1609. [Google Scholar] [CrossRef]
  36. Li, J.; Chen, R. Risk and Regulation of Big Data Policing. J. People’s Public Secur. Univ. China Soc. Sci. Ed. 2023, 39, 127–143. [Google Scholar]
  37. Brantingham, P.J.; Brantingham, P.L. The Geometry of Crime and Crime Pattern Theory. In Environmental Criminology and Crime Analysis; Routledge: London, UK, 2016; pp. 117–135. [Google Scholar]
  38. Errol, Z.; Madsen, J.B.; Moslehi, S. Social Disorganization Theory and Crime in the Advanced Countries: Two Centuries of Evidence. J. Econ. Behav. Organ. 2021, 191, 519–537. [Google Scholar] [CrossRef]
  39. Cohen, L.E.; Felson, M. Social Change and Crime Rate Trends: A Routine Activity Approach (1979). In Classics in Environmental Criminology; Routledge: London, UK, 2010; pp. 203–232. [Google Scholar]
  40. Townsley, M.; Homel, R.; Chaseling, J. Infectious Burglaries—A Test of the near Repeat Hypothesis. Br. J. Criminol. 2003, 43, 615–633. [Google Scholar] [CrossRef]
  41. Xu, C.; Yang, Y.; Song, G.; Liu, L.; Lan, M.; Chen, X. The Impact of Civil Registration-Based Demographic Heterogeneity on Community Thefts. Habitat Int. 2022, 129, 102673. [Google Scholar] [CrossRef]
  42. Zeng, M.; Mao, Y.; Wang, C. The Relationship between Street Environment and Street Crime: A Case Study of Pudong New Area, Shanghai, China. Cities 2021, 112, 103143. [Google Scholar] [CrossRef]
  43. Yue, H.; Xie, H.; Liu, L.; Chen, J. Detecting People on the Street and the Streetscape Physical Environment from Baidu Street View Images and Their Effects on Community-Level Street Crime in a Chinese City. ISPRS Int. J. GEO-Inf. 2022, 11, 151. [Google Scholar] [CrossRef]
  44. Gu, H.; Chen, P.; Li, H. Overview and Prospect for Spatial-Temporal Prediction of Crime. J. Geo-Inf. Sci. 2021, 23, 43–57. [Google Scholar]
  45. Kang, L.; Liu, H.; Cheng, W.; Chen, X.; Li, J. A Method of Urban Facility Hot Spot Recognition Considering Attribute Characteristics. Bull. Surv. Mapp. 2022, 1, 8–14. [Google Scholar] [CrossRef]
  46. Zhang, H.; Zhou, X.; Tang, G.; Ji, G.; Zhang, X.; Xiong, L. Inference Method for Cultural Diffusion Patterns Using a Field Model. Trans. GIS 2020, 24, 1578–1601. [Google Scholar] [CrossRef]
  47. Zhang, H.; Zhou, X.; Tang, G.; Zhou, L.; Ye, X. Hotspot Discovery and Its Spatial Pattern Analysis for Catering Service in Cities Based on Field Model in GIS. Geogr. Res. 2020, 39, 354–369. [Google Scholar]
  48. Chen, Y.; Yang, J.; Yan, T.; Zhang, C.; Zhu, D. An Approach on Calculation of Natural Quality Score of Agricultural Land Based on Positive and Negative Weights. J. Shenyang Agric. Univ. 2013, 44, 268–272. [Google Scholar]
  49. Long, D.; Liu, L.; Chen, J.; Xiao, L.; Song, G.; Xu, C. Impact of Prior Crime Experiences of Street Robbers on Subsequent Crime Location Choices. Prog. Geogr. 2020, 39, 815–828. [Google Scholar] [CrossRef]
  50. Liu, L.; Sun, Q.; Xiao, L.; Song, G.; Chen, J. The Temporal Influence Difference of Drug-Related Personnels’ Routine Activity on the Spatial Pattern of Theft. J. Geo-Inf. Sci. 2021, 23, 2187–2200. [Google Scholar]
  51. Xu, C.; Chen, X.; Chen, J.; Chen, D. Exploring the Impact of Floating Population with Different Household Registration on Theft. ISPRS Int. J. GEO-Inf. 2022, 11, 443. [Google Scholar] [CrossRef]
  52. Song, G.; Liu, L.; Bernasco, W.; Xiao, L.; Zhou, S.; Liao, W. Testing Indicators of Risk Populations for Theft from the Person across Space and Time: The Significance of Mobility and Outdoor Activity. Ann. Am. Assoc. Geogr. 2018, 108, 1370–1388. [Google Scholar] [CrossRef]
  53. Helbich, M.; Arsanjani, J.J. Spatial Eigenvector Filtering for Spatiotemporal Crime Mapping and Spatial Crime Analysis. Cartogr. Geogr. Inf. Sci. 2015, 42, 134–148. [Google Scholar] [CrossRef]
Ijgi 13 00415 g001
Figure 1. Theoretical framework of the integrated assessment of security risk.
Figure 1. Theoretical framework of the integrated assessment of security risk.
Ijgi 13 00415 g001
Ijgi 13 00415 g002
Figure 2. Study area. Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.
Figure 2. Study area. Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.
Ijgi 13 00415 g002
Ijgi 13 00415 g003
Figure 3. Research process of the integrated assessment of security risk.
Figure 3. Research process of the integrated assessment of security risk.
Ijgi 13 00415 g003
Ijgi 13 00415 g004
Figure 4. Distribution characteristics of public premises and their security risks in JY City: (a) political units; (b) commercial establishments; (c) transportation hubs; (d) community facilities; (e) public premises.
Figure 4. Distribution characteristics of public premises and their security risks in JY City: (a) political units; (b) commercial establishments; (c) transportation hubs; (d) community facilities; (e) public premises.
Ijgi 13 00415 g004
Ijgi 13 00415 g005
Figure 5. Distribution characteristics of the security situation of security risk in JY City (partial). Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.
Figure 5. Distribution characteristics of the security situation of security risk in JY City (partial). Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.
Ijgi 13 00415 g005
Ijgi 13 00415 g006
Figure 6. Distribution characteristics of the urban morphology and their security risk in JY City: (a) political units; (b) commercial establishments; (c) transportation hubs; (d) community facilities.
Figure 6. Distribution characteristics of the urban morphology and their security risk in JY City: (a) political units; (b) commercial establishments; (c) transportation hubs; (d) community facilities.
Ijgi 13 00415 g006
Ijgi 13 00415 g007
Figure 7. Distribution characteristics of the police resources of security risk in JY City (partial). Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.
Figure 7. Distribution characteristics of the police resources of security risk in JY City (partial). Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.
Ijgi 13 00415 g007
Ijgi 13 00415 g008
Figure 8. Hierarchical distribution map of integrated security risk points in JY City. Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.
Figure 8. Hierarchical distribution map of integrated security risk points in JY City. Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.
Ijgi 13 00415 g008
Ijgi 13 00415 g009
Figure 9. Hierarchical distribution map of the integrated security risk areas in JY City. Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.
Figure 9. Hierarchical distribution map of the integrated security risk areas in JY City. Due to the signing of a confidentiality agreement with the Public Security Bureau, all drawings related to the research area have been encrypted, and it is not appropriate to add a compass.
Ijgi 13 00415 g009
Table 1. Data descriptions and sources.
Table 1. Data descriptions and sources.
TypeNameExplanationSource
Micro-assessment dataPOIClassified as political units, commercial establishments, transportation hubs, community facilities, and public security organsJY City Public Security Bureau
Street crimeRobbery, theft, etc.
Floating populationTemporary resident population
Key personnelIndividuals struggling with substance abuse and those with psychopathic traits
Police facilitiesCCTV and traffic surveillance
Macro-assessment dataPopulation
distribution		Resolution of 1 kmhttps://landscan.ornl.gov/ (accessed on 14 November 2023)
Nighttime lightResolution of 500 mhttp://nnu.geodata.cn/ (accessed on 14 November 2023)
Land useResolution of 30 mhttps://zenodo.org/ (accessed on 14 November 2023)
Basic geographic dataAdministrative boundariesCity levelhttps://www.resdc.cn/ (accessed on 5 September 2023)
RoadMain roads, secondary roads, branch roads, etc.https://www.openstreetmap.org/ (accessed on 14 December 2023)
Table 2. Weights of the integrated index system of security risk.
Table 2. Weights of the integrated index system of security risk.
Indicator AttributeCriteria LayerIndicator LayerOverall Sorting Result
Positive (1.100)C1 Public premises (0.317)C11 Political units (0.313)0.1091
C12 Commercial establishments (0.163)0.0568
C13 Transportation hubs (0.264)0.0921
C14 Community facilities (0.160)0.0558
C15 Other categories (0.100)0.0349
C2 Security situation (0.504)C21 Public security issues (0.408)0.2262
C22 Floating population (0.281)0.1558
C23 Key personnel (0.311)0.1724
C3 Urban morphology (0.179)C31 Population distribution (0.413)0.0813
C32 Nighttime light (0.130)0.0256
C33 Impervious surface (0.457)0.0900
Negative (−0.100)C4 Police resources (1.000)C41 Public security organs (0.800)−0.0800
C42 Police facilities (0.200)−0.0200
Table 3. Basic attributions of the different levels of security risk points.
Table 3. Basic attributions of the different levels of security risk points.
Level of Security Risk PointsQuantityProportion (%)Integrated Risk Value Division
Level 11115.490.6658~0.4825
Level 21419.720.4739~0.4020
Level 32535.210.3961~0.3226
Level 42129.580.3179~0.2738
Table 4. Proportion of police resources in the different levels of security risk areas.
Table 4. Proportion of police resources in the different levels of security risk areas.
Level of Security Risk AreasPolice Station (%)Police Facilities (%)Police Resources (%)
Level 12.121.241.26
Level 212.714.985.08
Level 317.379.299.40
Level 424.5818.4118.49
Remaining area43.2266.0765.77
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Chen, J.; Li, W.; Li, Y.; Chen, Y. Integrated Assessment of Security Risk Considering Police Resources. ISPRS Int. J. Geo-Inf. 2024, 13, 415. https://doi.org/10.3390/ijgi13110415

AMA Style

Chen J, Li W, Li Y, Chen Y. Integrated Assessment of Security Risk Considering Police Resources. ISPRS International Journal of Geo-Information. 2024; 13(11):415. https://doi.org/10.3390/ijgi13110415

Chicago/Turabian Style

Chen, Jieying, Weihong Li, Yaxing Li, and Yebin Chen. 2024. "Integrated Assessment of Security Risk Considering Police Resources" ISPRS International Journal of Geo-Information 13, no. 11: 415. https://doi.org/10.3390/ijgi13110415

APA Style

Chen, J., Li, W., Li, Y., & Chen, Y. (2024). Integrated Assessment of Security Risk Considering Police Resources. ISPRS International Journal of Geo-Information, 13(11), 415. https://doi.org/10.3390/ijgi13110415

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop