GIS-Based Spatial Analysis and Strategic Placement of Fine Dust Alert Systems for Vulnerable Populations in Gangseo District

Abstract

Air pollution, particularly fine particulate matter (PM), poses significant health risks to vulnerable populations such as children, older adults, and individuals with chronic illnesses. Understanding the spatial distribution of these populations and their access to air quality information is crucial for effective interventions. In urban areas like Gangseo District, the distribution of essential facilities and accessibility varies greatly. While studies have highlighted the health impacts of PM, research on optimizing air quality monitoring for at-risk groups remains limited. This study aims to identify optimal locations for air quality monitoring by analyzing the spatial distribution of vulnerable populations and facility accessibility. Using Geographic Information Systems (GIS) and isochrone maps, we identified areas with high concentrations of vulnerable groups and poor access to healthcare facilities. Our findings revealed significant disparities in access to air quality information, with some high-risk areas underserved by current monitoring systems. This study integrated demographic data and spatial analysis to propose strategic monitoring placements. The methodology can be applied to other urban settings and offers a framework for improving air quality management. This study underscores the importance of targeted air quality monitoring to protect vulnerable populations and suggests practical steps for policymakers to enhance public health.

1. Introduction

Fine particulate matter (PM) refers to airborne or suspended particles that can have serious health impacts depending on their size [1,2]. PM is particularly harmful as it can cause respiratory and cardiovascular diseases, posing a significant threat to vulnerable populations such as children, older adults, and individuals with chronic illnesses [3,4]. This study combines environmental health, which examines the health impacts of air pollution, with urban planning and spatial analysis using geographic information systems (GIS), and social welfare, which aims to improve the quality of life for vulnerable populations [5,6]. Given these health risks, fine PM has gained global attention as a critical social challenge [7]. In major South Korean cities, including Seoul and its surrounding areas, frequent PM alerts significantly affect residents’ health and daily lives [8]. Although various technological systems measure and provide real-time PM concentrations, research is limited on how well these systems meet the needs of vulnerable populations. Additionally, disparities persist in access to PM information between areas where vulnerable populations reside and the facilities they frequently use [9].

Recent studies have highlighted that PM in Seoul and nearby areas originates from various sources, including vehicular emissions, industrial activities, and secondary aerosol formation from chemical reactions in the atmosphere [10,11,12]. These sources contribute significantly to the regional PM concentration and affect both visibility and health outcomes. For example, high levels of sulfate and nitrate have been associated with decreased visibility in urban areas, as well as with adverse health effects, including increased cardiovascular and respiratory mortality [10,12]. In the Seoul Metropolitan Area, PM levels have shown fluctuations influenced by both seasonal meteorological factors and human activities, with a notable increase in PM episodes related to stagnant weather conditions and reduced ventilation [11]. Understanding these sources and their impact on health is essential for developing effective PM monitoring and mitigation strategies.

Despite the availability of real-time PM monitoring systems, there is a gap in research on how effectively these systems address the specific needs of vulnerable populations in terms of accessibility and timeliness. Furthermore, there are still significant disparities in access to air quality information in different regions, particularly in areas where vulnerable groups live. This study seeks to fill this gap by analyzing the spatial distribution of vulnerable populations and proposing optimal locations for PM alert systems. The key research questions this paper addresses are (1) how can PM alert systems be strategically placed to better serve vulnerable populations? And (2) what factors should be considered to ensure equitable access to air quality information across different areas?

Gangseo District, located in western Seoul, faces significant environmental challenges due to its proximity to Incheon International Airport and industrial zones. Air pollution from vehicular emissions, industrial activities, and construction contributes to high levels of PM2.5 and PM10, posing health risks to local residents, especially the elderly and children. The urban heat island effect worsens air quality by trapping pollutants, making it critical to address these issues to protect public health. Gangseo District experiences a humid continental climate with four distinct seasons. Summers are hot and humid, while winters are cold and dry. In particular, fine dust levels rise during the spring due to yellow dust storms. These seasonal variations in air quality highlight the need for an effective fine dust monitoring system to protect vulnerable populations.

This study aimed to identify the residential distribution and related facilities of vulnerable populations in Gangseo District to propose optimal locations for PM alert systems. To achieve this, first, we analyzed the locations and accessibility of key facilities, such as hospitals and welfare centers. Second, we identified and visualized densely populated areas of vulnerable groups. Finally, we synthesized these data to propose optimal locations for installing PM alert systems. The core premise of this research is that improving the living environment of vulnerable populations requires providing PM information centered on their primary activity spaces and residences. This study aimed to establish practical and concrete measures to protect vulnerable populations from the health hazards of PM. Unlike previous studies that focused on measuring and forecasting PM concentrations, this study proposes a customized information provision plan tailored to the actual living environments of vulnerable groups. Through this study, we obtained detailed information on the distribution of residences and facilities frequented by vulnerable populations in Gangseo District. The optimal locations for PM alert systems identified can make a practical contribution to protecting the health of these vulnerable groups. Additionally, these findings serve as valuable data for local policymakers in formulating environmental policies to protect vulnerable populations. In conclusion, this study presents practical alternatives for protecting vulnerable populations from PM, thereby contributing to a reduction in social inequality.

The remainder of this paper is structured as follows: Section 2 reviews the relevant literature on air pollution and its health impacts. Section 3 details the methodology used for spatial analysis and data collection. In Section 4, we present the results of the analysis, identifying the optimal locations for PM alert systems in Gangseo District. Finally, Section 5 discusses the findings and their implications, and provides the conclusion, including recommendations for future research and policy implementation.

2. Literature Review

2.1. Particulate Matter and Health

PM refers to tiny particles suspended in the air, categorized by size into PM10 (particles with a diameter of 10 μm or less) and PM2.5 (particles with a diameter of 2.5 μm or less) [13]. These particles originate from both natural and anthropogenic sources. Natural sources include volcanic eruptions, wildfires, and dust storms, while anthropogenic sources include vehicular emissions, industrial activities, and construction sites. Smaller particles pose greater health risks as they can penetrate deeper into the respiratory system.

PM adversely affects human health through various pathways. It is primarily associated with respiratory diseases, as PM can reach the alveoli in the lungs, impairing lung function and exacerbating conditions such as asthma and chronic obstructive pulmonary disease (COPD) [14]. Additionally, PM exposure is associated with cardiovascular diseases, increasing the risk of hypertension, heart attack, and stroke. Beyond these conditions, PM can cause skin and eye irritation and weaken the immune system, and long-term exposure is associated with an increased risk of cancer.

Certain populations are particularly vulnerable to the health effects of PM. Older adults, with weaker immune systems and pre-existing health conditions, face heightened risks of PM exposure [15]. Children are more susceptible due to their faster breathing rates, leading to higher PM intake, which can negatively affect their growth and development [16]. Individuals with chronic diseases also face significant risks as PM can worsen their symptoms. Protecting these vulnerable groups is crucial for mitigating health risks and improving overall well-being.

The effective management of PM is essential for reducing its health impacts. Critical measures include providing real-time information on PM levels, installing air purifiers, and recommending limited outdoor activities during periods of high PM levels [17]. Installing PM monitors in areas with high concentrations of vulnerable populations, such as near hospitals, schools, and senior centers, can provide timely alerts and enable swift action to protect public health [18]. These measures are vital for minimizing the health effects of PM and improving the quality of life for vulnerable populations. Globally, PM is recognized as a significant environmental and public health issue, necessitating a comprehensive understanding and management [19]. Systematic research and the implementation of effective strategies, particularly for protecting vulnerable populations, can mitigate the adverse health effects of PM and enhance quality of life.

2.2. Spatial Analysis Techniques

Spatial analysis is a crucial technique for understanding and interpreting geographic data to identify patterns, relationships, and trends within a spatial context [6,20]. It plays a vital role in various fields, including urban planning, environmental studies, and transportation management. The core methods of spatial analysis include spatial statistical analysis, spatial regression analysis, and cluster analysis. Spatial statistical analysis applies statistical techniques to analyze the spatial distribution and patterns of data [21]. This method is essential for identifying hotspots, trends, and anomalies within geographic data, providing insights that are not immediately apparent through simple visual inspection. Spatial regression analysis, on the other hand, examines the relationship between spatial and non-spatial variables [22]. This technique allows researchers to understand how geographic factors influence outcomes such as population density, pollution levels, and disease spread. By integrating geographic information into regression models, researchers can make more accurate predictions and understand underlying spatial dynamics [23]. Cluster analysis groups data points based on similarities or distances [24]. It identifies areas with similar characteristics, such as regions with high pollution levels, high population density, or specific socioeconomic conditions, and is particularly useful for detecting spatial patterns and making informed decisions on resource allocation and policy implementation.

Isochrone analysis is a specialized type of spatial analysis used to represent areas that can be reached within a specific timeframe from a given location [25]. It considers factors such as transportation networks, traffic conditions, and travel speeds to accurately calculate travel times. The principles of isochrone analysis involve creating lines, known as isochrones, on a map that connect points with equal travel times to a central location [26]. Isochrone analysis provides a detailed and realistic representation of accessibility within a given area by incorporating data on road networks, public transportation routes, and traffic patterns. It is widely used in transportation planning to assess the accessibility of public transportation stops and optimize routes and stops to ensure efficient and equitable service coverage [27]. It is also critical in emergency service planning, helping to determine the optimal locations for facilities such as fire stations and hospitals to minimize response times. Additionally, businesses use isochrone analysis for site selection to maximize accessibility to potential customers within specific travel times. By visualizing accessibility and travel times, isochrone analysis aids in making informed decisions regarding the placement and optimization of various services and facilities, enhancing urban planning and public service delivery, ultimately contributing to an improved quality of life for communities.

2.3. The Necessity and Effectiveness of Installing Particulate Matter Monitors

PM monitors play a crucial role in public health by providing real-time information on air quality, particularly on the levels of PM such as PM10 and PM2.5 [28]. These monitors raise awareness and enable prompt responses to high PM levels, allowing individuals and communities to make informed decisions about outdoor activities and protective measures. PM monitors also serve as important resources for researchers and policymakers, providing data to study pollution patterns, assess the effectiveness of air quality regulations, and develop strategies to reduce exposure to harmful particulates [18,29].

Installing PM monitors is particularly necessary in areas with high concentrations of vulnerable populations, such as older adults, children, and those with pre-existing health conditions [28]. These groups are more susceptible to health issues related to PM, including respiratory and cardiovascular diseases. Strategically placing PM monitors near hospitals, schools, and senior centers, communities ensures that at-risk individuals receive timely information regarding air quality. Moreover, certain regions may have varying levels of air pollution due to local sources of PM such as traffic, industrial activities, or construction sites. Installing monitors in these areas allows the identification of pollution hotspots and the assessment of localized air quality issues [30]. These data are critical for implementing targeted interventions to improve air quality and protect public health.

The primary benefit of PM monitors is their ability to provide real-time air quality information, which is essential for protecting public health [31]. When PM levels rise, timely alerts can prompt individuals to take protective measures such as staying indoors, using air purifiers, or wearing masks. Immediate access to air quality data empowers people to make informed decisions and reduce their exposure to harmful particulates. For policymakers and urban planners, the data collected by PM monitors are invaluable for shaping air quality management strategies [32]. By analyzing trends and patterns in PM concentrations, authorities can identify sources of pollution and implement regulations or interventions to address them. This data-driven approach ensures that resources are allocated effectively and that air quality improvement efforts are based on solid evidence. Furthermore, PM monitors can enhance community engagement and awareness of air pollution issues [33]. By making air quality data accessible to the public through websites, mobile apps, or public displays, communities can foster a greater understanding of environmental health risks and encourage proactive behaviors to mitigate exposure. Increased awareness can drive collective action and support policies aimed at reducing air pollution. In addition to health benefits, PM monitors can contribute to broader environmental and economic gains. Improved air quality reduces healthcare costs associated with treating pollution-related illnesses. It also enhances the overall quality of life, making cities more attractive places to live, work, and visit. Consequently, investments in air quality monitoring can yield significant long-term returns for both public health and economic development.

3. Methods

This study employed a comprehensive methodology to identify optimal locations for air quality monitoring aimed at protecting vulnerable populations in the Gangseo District, as shown in Figure 1. Our approach integrated spatial data analysis, accessibility assessment, and multicriteria decision analysis to provide actionable insights. The overall process is depicted in Figure 2, which outlines the sequential steps of the proposed methodology. For the purposes of research and analysis, we implemented a systematic approach to gather and prepare the data. The process began with defining and identifying all necessary data sources, followed by collection from each identified source. Data were then processed and refined using appropriate tools tailored to each data type, as detailed in Appendix A. This study handled various types of data, including structured data (e.g., databases and spreadsheets), semi-structured data (e.g., JSON and XML), and unstructured data (e.g., text and images).

3.1. Analysis of Locations of Facilities for Vulnerable Populations

The first step in our methodology involves identifying and mapping the locations of key facilities frequented by vulnerable populations in the Gangseo District, such as hospitals, nursing homes, and other care centers. To achieve this, we collected spatial data on these facilities from government databases, healthcare directories, and local administrative offices. The data included geographic coordinates, facility types, and services provided. Using geographic information systems (GIS), we mapped the spatial distribution of these facilities. This mapping visualized the geographic spread and clustering of facilities, providing a clear picture of their accessibility to vulnerable populations.

3.2. Accessibility Analysis for Vulnerable Populations

Next, we analyzed the accessibility of these facilities to vulnerable populations using isochrone maps, which represent areas accessible within a specific time frame from a given point. We generated isochrone maps using GIS software, depicting areas that could be reached within various time intervals (e.g., 5, 10, and 15 min) on foot or by public transport from each facility. By comparing these isochrones with population density maps, we identified regions with poor accessibility to essential services. These gaps were flagged as priority areas for intervention.

3.3. Analysis of Residential Distribution of Vulnerable Populations

Understanding the locations of vulnerable populations is critical for targeted interventions. We obtained demographic data, including age, sex, and household composition, from national census data and local government records. These data were analyzed using GIS to create density maps of vulnerable populations. These maps illustrate areas with high concentrations of vulnerable groups, aiding in the identification of neighborhoods that may require more focused attention.

3.4. Determining Optimal Locations for Air Quality Monitors

By combining insights from the facility location and residential distribution analyses, we determined the optimal locations for installing air quality monitors. We developed criteria for placing monitors by considering factors such as population density and proximity to key facilities, and identified accessibility gaps. Using multi-criteria decision analysis (MCDA), we integrated these criteria to rank potential locations for air quality monitors. This method enables a systematic evaluation of multiple factors to identify the most effective and efficient placement sites.

3.5. Visualization and Dissemination of Data

The final step involved creating comprehensive visualizations and making the data accessible to the public. We began by creating detailed maps illustrating the distribution of vulnerable populations, locations of key facilities, accessibility isochrones, and proposed sites for air quality monitoring. These maps were generated using GIS software (R 4.2.2) and designed to be user-friendly, ensuring that the information is easily interpretable by a broad audience. To further enhance accessibility, we developed an interactive web platform where users could view and interact with data in real time. This platform provided tools for users to explore the data and gain a deeper understanding of the spatial dynamics of air quality and vulnerability in Gangseo District.

By following this structured methodology, we aimed to provide a robust framework for improving the quality of life for vulnerable populations through the strategic placement of air quality monitoring systems.

4. Results

4.1. Spatial Distribution of Key Facilities for Vulnerable Populations

To ensure that the most current information was used in the analysis, we established a process for automatically collecting, cleaning, and analyzing raw data. This process is crucial for continuously updating our dataset and ensuring that our analysis reflects the latest conditions. Specifically, we focused on identifying areas with a high concentration of facilities serving vulnerable populations and visualizing regions where medical facilities are lacking.

The analysis targeted several key components to understand the distribution and accessibility of facilities used by vulnerable populations. First, we constructed a comprehensive dataset of medical facilities, including hospitals and clinics, with a focus on their geographical locations and overall accessibility. To assess accessibility, we generated isochrone maps depicting areas within specific walking distances of these medical facilities. Additionally, we expanded the analysis to include spatial data on other critical facilities, such as primary schools, daycare centers, kindergartens, and senior centers. These facilities play a vital role in supporting vulnerable populations, and their inclusion in the spatial data analysis was essential for developing a complete picture of the overall infrastructure available in the district. By integrating these spatial data, the analysis provided valuable insights into both the concentration of these facilities and the ease of access for the populations who rely on them.

The data sources for the analysis included two primary datasets. The first was the Medical Facilities Licensing Data, which provided comprehensive information on the number and location of hospitals and clinics throughout Gangseo District. This dataset was instrumental in mapping the distribution of medical facilities. The second dataset comprised the Standard Data for Facilities for the Elderly and Children, which contained detailed information on the locations of senior centers, daycare centers, kindergartens, and primary schools. These data were essential for understanding the spatial distribution of key support facilities for vulnerable populations within the district.

To analyze the data, we employed several analytical techniques to create visual representations that clearly illustrate the spatial distribution of key facilities. Specifically, we processed the spatial data for medical and senior centers to produce detailed visualizations depicting their distribution across Gangseo District. These visualizations were crucial for identifying areas of high facility concentration and regions where these essential services are less accessible to vulnerable populations.

A comprehensive map was generated categorizing hospitals and clinics into four types: internal medicine, obstetrics and gynecology, pediatrics, and otorhinolaryngology (ear, nose, and throat; Figure 3). This classification helped to understand the distribution of these key medical services. As of July 2024, there were 322 hospitals and clinics across the Gangseo District, excluding duplicates among the specified medical departments. The highest concentrations were found in Banghwa 1-dong, Balsan 1-dong, and Hwagok 1-dong. In contrast, areas such as Deungchon 2-dong, Banghwa 3-dong, and Gayang 3-dong were found to have fewer medical facilities. Notably, obstetrics and gynecology services were absent in Hwagok 2-dong, Deungchon 2-dong, and Gayang 3-dong, indicating potential vulnerabilities in these regions.

Another map was created to illustrate the distribution of facilities for the elderly and children, including senior centers, daycare centers, kindergartens, and primary schools, as shown in Figure 4. As of July 2024, the Gangseo District had 577 facilities for older adults and their children, including 312 daycare centers, 192 primary schools, 35 kindergartens, and 38 senior centers. The areas with the highest concentrations of these facilities were Banghwa 1-dong, Yeomchang-dong, and Balsan 1-dong. However, the distribution was less dense in Hwagok-dong, Gayang 2-dong, and Banghwa 2-dong, suggesting that these areas may have fewer support services for vulnerable populations.

4.2. Evaluation of Accessibility to Essential Services for Vulnerable Populations

In this section, we present a detailed accessibility analysis of vulnerable populations by constructing and operating isochrone data based on pedestrian and vehicle movements. Traditional methods of urban environment analysis often set usage areas without considering the actual road networks or terrain. This approach tends to result in low-resolution outputs and a high error rate owing to the oversimplification of the spatial environment.

To address these limitations, our study employed an advanced methodology that accounts for the three-dimensional spatial environment, including terrain and road network complexities. Additionally, we focused on the primary modes of transportation used by vulnerable populations and their typical travel speeds. This allowed us to generate a more accurate and realistic representation of accessible areas, by incorporating time as a factor rather than solely relying on distance.

For hospitals and clinics, we established a standard walking distance of 15 min as the threshold for accessibility. We visualized the areas (parcels) within Gangseo-gu that could be reached within a 15 min walking distance from each hospital or clinic. As depicted in Figure 5, of the 42,850 lots in Gangseo-gu, 10,096 lots were identified as inaccessible within a 15 min walking distance to any hospital or clinic. This analysis highlights significant gaps in healthcare accessibility, revealing regions where residents may face challenges in promptly accessing medical services. By identifying these gaps, this study provides crucial insights that can inform urban planning and healthcare service improvements for vulnerable populations.

4.3. Spatial Concentration and Distribution of Vulnerable Populations

In this section, we analyze the residential distribution of vulnerable populations within Gangseo-gu, focusing on two key demographic groups: older adults (aged 65 and over) and children under 14 years old. The analysis was based on population statistics by district and administrative division, categorized by sex and age.

4.3.1. Resident Population Status

The analysis began by examining the residential locations of the older adult population, revealing that their residences are predominantly concentrated around Gayang Station, Songjeong Station, and Hwagok-Kkachisan Station. As depicted in Figure 6, these areas show significant clustering of the older adult population, indicating potential zones where healthcare and social services should be prioritized.

For children under 14 years old, the analysis showed that their residences are primarily distributed around the Sinbanghwa-Magok Station and Yeomchang Station-Jungmi Station areas. This pattern, represented in Figure 7, suggests that these neighborhoods are particularly important when considering the placement of educational and recreational facilities for younger population.

The intensity of residential concentration is visually represented in Figure 6 and Figure 7, with areas of higher population densities indicated by darker, more saturated shading. This color gradient effectively highlights densely populated areas in each demographic group.

4.3.2. Analysis of Daytime Population Data

In addition to the residential population, we analyzed daytime population data to identify areas where vulnerable groups are concentrated throughout Gangseo-gu. This analysis included population density evaluations by time zone, allowing us to identify areas with significant population shifts during the day. By analyzing the density of specific age groups, particularly older adults and children under 14 years old, we identified the times and areas where these populations are most concentrated. For instance, Figure 8 (left) shows the concentration of the older adult population, while Figure 8 (right) illustrates the distribution of children under 14 years old.

Interestingly, our analysis of the daytime population revealed a different pattern from that of the residential population. During the day, the population tended to concentrate more on the outskirts of Gangseo-gu, contrasting with the more centralized residential patterns. This discrepancy indicates the necessity of considering both the daytime population and the residential population when assessing the vulnerability of these groups to environmental hazards, such as fine dust. This comprehensive analysis of both resident and daytime population data underscores the importance of incorporating dynamic population shifts into planning and intervention strategies. By understanding where and when vulnerable populations are most concentrated, more effective and targeted measures can be developed to protect them from exposure to fine dust and other environmental risks.

4.4. Optimal Placement Strategy for Air Quality Monitors

In this section, we propose optimal locations for installing air quality monitors by developing a comprehensive index that integrates multiple critical factors. The goal was to identify areas where air quality monitors are most urgently needed based on a systematic analysis of the concentration of vulnerable populations, accessibility to relevant facilities, and environmental conditions.

4.4.1. Data Definition and Preprocessing

To accurately assess the need for air quality monitoring, we defined and preprocessed three key datasets.

-

D1: Density of Vulnerable Populations

This dataset represents the density of populations vulnerable to PM across different regions. The density values range from 0 to 1, with values closer to 1 indicating higher concentrations of vulnerable groups, such as older adults and children under 14 years old.

-

D2: Accessibility to PM-Related Facilities

This dataset measures the accessibility of relevant facilities such as hospitals and clinics within each region. The accessibility score ranges from 0 to 1, with values closer to 1 indicating poorer accessibility, reflecting areas where vulnerable populations have limited access to essential services.

-

D3: Environmental Indicator (PM Concentration)

This dataset provides the average PM concentration for each region, with values normalized between 0 and 1. Higher values correspond to poorer air quality, indicating a greater need for air quality monitoring in these areas.

4.4.2. Calculation of the Comprehensive Index

We calculated a comprehensive index (S) for each region to determine the overall priority for air quality monitoring. This index is a weighted combination of the three datasets described above. The weight of each factor was assigned based on its relative importance.

-

W1: Density of Vulnerable Populations (Weight: 0.5)

Given the critical importance of protecting vulnerable groups, this factor was assigned the highest weight.

-

W2: Accessibility to PM-Related Facilities (Weight: 0.3)

Accessibility is a significant factor, as regions with poor access to health care and other facilities are at higher risk.

-

W3: Environmental Indicator (PM Concentration) (Weight: 0.2)

While important, air quality data alone were given a lower weight relative to the other factors.

The comprehensive index for each region was calculated using the following formula:

S = (W1 × D1) + (W2 × D2) + (W3 × D3)

(1)

This formula produces a value for S that ranges from 0 to 1, with higher values indicating regions where the installation of air quality monitors is urgently required. For example, a region with a comprehensive index S of 0.72. is considered a high priority area for monitor installation.

4.4.3. Ranking of Regions Based on the Comprehensive Index

After calculating the comprehensive index (S) for all the regions within Gangseo-gu, we ranked the regions based on their scores. This ranking helps prioritize the installation of air quality monitors, starting with regions that exhibit the highest values of S. The top 100 regions, based on their comprehensive index, were identified as the highest priority areas for air quality monitoring.

By following this systematic approach, we ensured that the placement of air quality monitors was data-driven and targeted areas where they could have the greatest impact. Regions with the highest comprehensive index scores were proposed as optimal locations for the installation of air quality monitors, ensuring that vulnerable populations are better protected from the adverse effects of PM.

4.5. Visualization Techniques and Data Presentation

4.5.1. Comprehensive Index Results

The visualization of the comprehensive index results, as shown in Figure 9, highlights the most vulnerable areas within Gangseo-gu, based on a combination of population density, accessibility to healthcare facilities, and environmental conditions. Regions with a higher index score represent areas with a high concentration of vulnerable populations coupled with poor accessibility to hospitals and clinics. Our analysis by administrative district identified Deungchon 1-dong, Hwagok 2-dong, Hwagok 1-dong, and Gayang 1-dong as the most vulnerable areas, in that order. These areas are particularly critical for interventions due to their high vulnerability scores.

4.5.2. Top 20 Proposed Locations for Air Quality Monitors

Based on the comprehensive index scores, we selected the top 20 regions with the highest values as the most urgent locations for the installation of air quality monitors, as shown in Figure 10. The area with the highest score recorded 99.3 points, indicating an extreme need for monitoring, whereas the 20th ranked area scored 39.4 points. These top 20 locations were proposed as primary sites for the installation of air quality monitors, ensuring that the most vulnerable populations received the necessary protection from PM exposure. This prioritization helps to strategically place monitors to maximize their impact and effectiveness across Gangseo-gu.

4.5.3. Uncertainty Assessment of the Comprehensive Index

To evaluate the robustness of the comprehensive index, an uncertainty assessment was conducted by varying the weights of the components in the index formula. This assessment aimed to determine the sensitivity of the index to changes in weight values and to examine the impact of these variations on the prioritization of air quality monitor locations, as shown in Figure 11.

The analysis revealed that the comprehensive index results remained relatively stable across the different weight configurations, indicating consistency in the identification of priority areas. Specifically, we found the following:

-

Increasing the weight of W1 (Density of Vulnerable Populations) led to proportional increases in the comprehensive index values for Yeomchang-dong, Gayang 2-dong, Hwagok 8-dong, and Deungchon 2-dong, highlighting the impact of population density in these areas.

-

Adjusting the weight of W2 (Accessibility to PM-Related Facilities) resulted in increased index values for Deungchon 3-dong, Yeomchang-dong, Hwagok 3-dong, and Hwagok 6-dong, emphasizing the importance of accessibility in these regions.

-

When the weight of W3 (Environmental Indicator—PM Concentration) was increased, there was a corresponding rise in the index values for Gonghang-dong, Banghwa 1-dong, Hwagok 2-dong, and Deungchon 3-dong, reflecting the influence of environmental conditions in these areas.

This uncertainty assessment demonstrates that while minor variations in index values were observed with changes in the weights, the prioritization of areas for air quality monitoring remained largely consistent. The stability of the comprehensive index under different weighting scenarios supports the robustness of this approach in identifying optimal locations for air quality monitors in Gangseo-gu.

5. Discussion

This study aimed to enhance the quality of life for vulnerable populations in the Gangseo District by identifying optimal locations for air quality monitoring. Through comprehensive spatial analysis and accessibility assessments, we proposed strategic placements for these monitors to ensure the effective protection of at-risk groups. The key conclusions derived from our detailed methodology are as follows.

First, our analysis revealed significant disparities in the accessibility of essential facilities among vulnerable populations. By mapping the locations of hospitals, nursing homes, and other critical care centers, we identified areas with poor accessibility and high population densities of vulnerable groups. These findings are crucial for determining optimal sites for air quality monitoring.

Second, using isochrone maps allowed us to visualize the accessibility of these facilities within specific timeframes, highlighting regions where vulnerable populations face significant barriers in accessing healthcare and other services. This step was instrumental in identifying areas requiring urgent intervention.

Finally, integrating demographic data with GIS enabled us to create detailed density maps of vulnerable populations. These maps provide a clear picture of where high-risk groups reside, further supporting our recommendations for placement monitoring.

Despite these significant findings, our study has several limitations. The accuracy of our results depended on the quality and completeness of the spatial and demographic data. Additionally, our study focused solely on the Gangseo District, limiting the generalizability of our findings to other regions. Future studies should consider the dynamic nature of urban environments. Continuous data updates and real-time monitoring are essential for adapting to changing conditions. Expanding this research to other districts and cities will provide a broader understanding of the air quality management needs in different urban settings. Another limitation is the lack of detailed air pollution data in this study. The primary objective was to enhance the ease and speed with which vulnerable populations could receive PM alerts. Given that air pollution trends, particularly PM concentrations, show similar patterns across local regions, the study did not focus on the distribution of PM levels. However, we acknowledge that the absence of detailed air pollution data could limit the precision of our proposed system. As a result, this limitation will be addressed in the conclusion to highlight the potential for future research to incorporate detailed PM concentration data for a more refined system design. In addition, the linear combination of variables used in determining optimal locations for PM alert systems may oversimplify the complexity of multi-criteria decision-making. We recognize that the method does not fully account for the varying importance of criteria under different stakeholder scenarios. A more sophisticated multi-criteria analysis could incorporate weight variations to assess how the top 20 proposed locations change under different assumptions. This would provide a more robust framework for evaluating the stability and sensitivity in the proposed locations. Future research should consider incorporating a sensitivity analysis or scenario-based approaches to better reflect the diverse priorities of stakeholders. Another important direction for future research is the application of this methodology to regions where air quality monitors are already installed. A comparative analysis between regions with existing monitors and proposed placements could provide valuable insights into the accuracy and effectiveness of our model. This would further enhance the robustness of the proposed placements and validate the method under different conditions.

These findings have practical applications in urban planning, public health, and environmental management. Policymakers can use the strategic placement of air quality monitors to develop targeted interventions, improve public health outcomes, and mitigate the adverse effects of air pollution on vulnerable populations.

In future research, it is crucial to maintain a multidisciplinary approach that incorporates advances in technology, environmental science, and social policy. Researchers should also prioritize community engagement to ensure that interventions meet the actual needs of the populations they aim to protect. This study provides a robust framework for optimizing air quality monitoring placement, emphasizing the importance of accessibility and demographic considerations. By addressing the specific needs of vulnerable populations, our approach can contribute to more equitable and effective air quality management strategies.