Study on the Effectiveness of Multi-Dimensional Approaches to Urban Flood Risk Assessment

Abstract

Increasing frequency and severity of urban flooding, driven by climate change and urban population growth, present major challenges. Traditional flood control infrastructure alone cannot fully prevent flood damage, highlighting the need for a comprehensive and multi-dimensional disaster management approach. This study proposes the Flood Risk Index for Building (FRIB)—a building-level assessment framework that integrates vulnerability, hazard, and exposure. FRIB assigns customized risk levels to individual buildings and evaluates the effectiveness of a multi-dimensional method. Compared to traditional indicators like flood depth, FRIB more accurately identifies high-risk areas by incorporating diverse risk factors. It also enables efficient resource allocation by excluding low-risk buildings, focusing efforts on high-risk zones. For example, in a case where 5124 buildings were targeted based on 1 m flood depth, applying FRIB excluded 24 buildings with “low” risk and up to 530 with “high” risk, reducing unnecessary interventions. Moreover, quantitative metrics like entropy and variance showed that as FRIB levels rise, flood depth distributions become more balanced—demonstrating that depth alone does not determine risk. In conclusion, while qualitative labels such as “very low” to “very high” aid intuitive understanding, FRIB’s quantitative, multi-dimensional approach enhances precision in urban flood management. Future research may expand FRIB’s application to varied regions, supporting tailored flood response strategies.

1. Introduction

As the intensity of heavy rainfall increases due to climate change, the damages caused by urban flood in densely populated areas are progressively escalating. Urban flood poses significant risks to human lives, infrastructure, and economic stability, with its impacts intensifying as extreme weather events become more frequent. The rising intensity of rainfall indicates that the effectiveness of existing structural measures—designed using historical rainfall data—is steadily diminishing. While these issues can be mitigated by adopting higher design frequencies or increasing the height of levees, such approaches often require substantial financial investment and time, as they involve extensive modification of existing infrastructure (Hallegatte et al. [1]).

In response, non-structural approaches have gained prominence as alternatives, aiming to maximize the utility of current flood control facilities while proactively identifying hazardous areas and applying localized flood mitigation measures. Among these, GIS, remote sensing, and Multi-Criteria Decision Analysis (MCDA) are representative methods that can establish more localized and effective flood management strategies, especially in urban environments with limited data availability (Li, C., et al. [2]). This method avoids large-scale infrastructure projects, offering cost-effectiveness and reduced implementation timelines as measures are targeted at specific risk-prone areas. However, within this framework, there are two distinct approaches to risk assessment: one relies solely on simple physical indicators, such as flood depth, while the other integrates a wide variety of indicators for a comprehensive risk assessment.

The physical-indicator-based approach is intuitive and straightforward, making it widely used for identifying risk areas. Various methods, such as approach through numerical analysis, machine-learning-based approaches, and fuzzy-logic-based approaches, have utilized physical indicators like flood depth. It has been adopted by numerous studies as a traditional method for flood risk assessment (EI et al. [3]; Ueda et al. [4]; Parvaze et al. [5]; Rinta et al. [6]). However, this approach primarily considers the magnitude of flood depth, often failing to account for critical factors like population density, proximity to essential infrastructure, or socio-economic conditions. This limitation has been shown to result in an inability to capture novel risk profiles that emerge due to network vulnerabilities in urban systems (Yin, K., et al. [7]). As a result, its ability to capture the multi-dimensional nature of flood risk is inherently limited.

Conversely, comprehensive risk assessments incorporate multiple indicators, such as vulnerability, hazard, and exposure, enabling a more precise evaluation of overall flood risk. This approach leverages diverse datasets to identify high-risk areas while considering socio-environmental factors that influence flood impacts. Many studies have demonstrated the effectiveness of this methodology in producing balanced risk profiles and guiding targeted disaster management strategies (Gigović et al. [8]; Song, [9]; Nam, [10]; Park, [11]; Kubal et al. [12]; Wang et al. [13]). It has also been emphasized that a multidimensional assessment integrating human resilience and environmental exposure performs significantly better than single-variable approaches in urban flood forecasting (Zhang et al. [14]). However, the applicability of such methods is often constrained by the availability and accessibility of region-specific datasets, limiting their use in areas like Korea where public data is often generalized or fragmented. Additionally, comparative analyses between the physical-indicator-based approach and comprehensive methods remain limited, leaving a gap in understanding the differences and relative advantages of these two approaches.

To overcome these limitations, this study proposes a comprehensive flood risk assessment methodology utilizing publicly available data in Korea. The proposed methodology integrates various risk indicators to evaluate flood risk from a multi-dimensional perspective, highlighting the advantages of this approach compared to traditional flood depth-based methods. By doing so, this study aims to enable effective flood risk management without relying on specialized datasets, providing a new direction for urban flood response and risk mitigation strategies.

2. Methodology

Previous integrated indicator-based risk assessments have largely been conducted at the regional scale, making it difficult to perform detailed building-level evaluations. Therefore, this study proposes a building-level risk assessment method using publicly available data and introduces the Flood Risk Index for Building (FRIB) for this purpose. In this study, ‘publicly available data’ refers to information accessible to the general public through government agencies, public data portals, or open-source map platforms. This approach aims to overcome the limitations of region-specific datasets and enable effective flood risk management without reliance on specialized datasets. FRIB is evaluated by considering the three attributes that constitute a disaster—vulnerability, hazard, and exposure. Subsequently, a comparative analysis was conducted with a single physical hazard approach (based on flood depth).

2.1. Flood Risk Index for Building (FRIB)

In the field of disaster management, risk is composed of three elements: vulnerability, hazard, and exposure (UNDDR, [15]). Vulnerability refers to the degree to which damage can be exacerbated, hazard indicates the extent of potential damage caused by flooding, and exposure represents the degree to which an area or entity is subjected to flood risk. In this study, FRIB (Flood Risk Index for Building) is defined—by considering these three properties—as “the extent of loss that a particular building may incur due to urban flooding and its overall likelihood of being exposed to such risk”. Figure 1 illustrates this definition, representing the relationships and definitions for each component.

Various indicators for evaluating the three components that constitute FRIB have been identified in previous research (Figure 2). From among these indicators, this study selected those that are publicly available government data within Korea (excluding paid or private company data) and can be applied at the building level. The data below were used as they were, or additional indicators were generated using those indicators.

2.2. Vulnerability

Vulnerability refers to the likelihood that a particular region or population will be significantly affected by urban flood, and it is commonly assessed through social and economic indicators. In this study, this operational definition concretizes the broader understanding of vulnerability as the degree to which damage can be exacerbated, as established in disaster management contexts. By examining these factors, one can determine how swiftly and efficiently a community can respond in the face of a crisis. This assessment also highlights which groups are most likely to suffer greater harm and clarifies the resources and procedures required during the recovery phase. Furthermore, the indicators listed in the table below can be used to evaluate the degree to which a community is exposed to potential urban flood (

Table 1. Vulnerability evaluation components and methods.

Category	Component	Evaluation Method
Vulnerability	Vulnerable population	Based on statistics, GIS data
	Building age	Building-data-based
	Distance to main facilities	Route analysis

).

2.2.1. Vulnerable Population

Groups classified as vulnerable in urban flood contexts tend to experience more severe impacts than the general population. Statistical data and GIS (Geographic Information System) analyses are particularly useful for identifying the locations where these groups are concentrated. GIS data analysis was conducted using QGIS 3.22 (Quantum Geographic Information System), an open-source GIS software. When high densities of vulnerable individuals are present in specific regions, additional support and resources may be required both during and after the event. This information plays a pivotal role in assessing a community’s overall susceptibility and in allocating essential resources ahead of time. In this study, the focus was on older adults and infants as representative vulnerable groups, and relevant GIS data were obtained from the National Land Information Platform [24].

2.2.2. Building Age

The building age serves as a fundamental measure of structural resilience in urban flood conditions. By examining local building records, one can determine how old these structures are and whether they align with current safety standards. Typically, older buildings pose a higher risk, indicating that reinforcement or renovations may be needed to bolster their durability. This assessment aids in ensuring that the built environment meets modern urban flood preparedness requirements, ultimately contributing to the community’s safety.

2.2.3. Distance to Main Facilities

Another important factor in evaluating urban flood readiness is the proximity of critical facilities. Through route analysis, it becomes possible to estimate how quickly emergency operations, such as rescues, can be initiated. Longer distances can lead to delays in response times, which may substantially affect both disaster management and the speed of recovery. In line with the Ministry of the Interior and Safety’s Emergency Response Command Regulations, this identified fire stations, police stations, and hospitals as the primary facilities necessary for an effective emergency response.

2.3. Hazard

Assessing hazard involves evaluating both the potential intensity of natural urban flood in a specific region and the extent of possible damage. Physical indicators such as the straight-line distance to the closest river or road, along with structural damage estimates based on flood depth (

Table 2. Hazard evaluation components and methods.

Category	Component	Evaluation Method
Hazard	Straight-line distance to the river	GIS analysis
	Straight-line distance to the road	GIS analysis
	Structural damage based on flood depth	Damage function-based analysis

), play a central role in this analysis. To accomplish these measurements, GIS tools were used for calculating direct distances, and damage functions were applied to derive structural damage rates. These parameters are critical for understanding the physical attributes of the area and for projecting how losses might be distributed in the event of an urban flood.

2.3.1. Straight-Line Distance to the River

The direct distance from a building to the nearest river serves as a key metric for assessing flood risk. Structures located near rivers face a higher likelihood of external flooding during a flood event, thereby elevating the risk of physical damage. To quantify this distance, a GIS analysis was performed to compute the shortest straight-line path from each building to the closest river. Employing this approach enables a more accurate evaluation of the potential for external flooding based on the building’s geographic position.

2.3.2. Straight-Line Distance to the Road

The distance to the nearest road, measured in a straight line, is also a pivotal factor during urban flood events, particularly for evacuation and emergency response. In crises, areas with well-developed road networks enable faster evacuation and facilitate rescue operations, thereby streamlining recovery efforts. By using GIS analysis, one can calculate how far each structure is from accessible roads to gauge its overall accessibility and disaster response capabilities. In this study, data from Open Street Map (OSM) was employed to conduct this evaluation.

2.3.3. Structural Damage Based on Flood Depth

In urban flood scenarios, the level of damage sustained by a building largely depends on its structural characteristics and the depth of inundation. A common method for evaluating flood-depth-related structural damage is to utilize depth damage functions, which compute damage rates according to flood depth and building type. Drawing on experimental data, these functions estimate how much structural harm a building might incur at varying flood depths. This study employed damage functions developed by the Korea Institute of Civil Engineering and Building Technology (KICT [25]) for different types of facilities. The findings not only detail the extent of structural damage but also assist in estimating financial losses, offering crucial insights into total damage across the affected region.

2.4. Exposure

Exposure assessment involves determining the extent to which populations and infrastructure may be influenced by an urban flood. Crucial metrics employed in this process encompass flood occurrence frequency, weight adjustments based on flood depth for underground structures, and the number of people at risk (

Table 3. Exposure evaluation components and methods.

Category	Component	Evaluation Method
Exposure	Flood occurrence frequency	Based on statistical and GIS data
	Underground risk by flood depth	Based on empirical formulas
	Population	Based on statistical and GIS data

).

2.4.1. Flood Occurrence Frequency

Flood occurrence frequency assesses how often floods have taken place in a particular area based on past records. By integrating statistical data and GIS analyses, one can determine the frequency of flood events in each region, thereby obtaining a more precise prediction of flood risk. In general, a higher frequency of flooding indicates an increased likelihood of severe damage during an urban flood. In this study, GIS analysis was performed using inundation trace data from Seoul between 2011 and 2022, alongside building data, to evaluate the flood occurrence frequency for each building.

2.4.2. Underground Risk by Flood Depth

The severity of flood damage can differ depending on whether underground structures are present. To evaluate the extent of damage based on flood depth, an underground weight factor is employed, which is calculated using a flow rate formula that estimates floodwater inflow at various depths. Specifically, this factor is determined by comparing the volume of floodwater flowing into underground facilities—derived from the above-ground flood depth—with the total volume of the underground space. The formula used to calculate floodwater inflow into underground facilities is shown below (Ministry of the Interior and Safety [26]).

$$Q_t=1.59L_w{h}_t^{1.65}$$

(1)

$${\gamma }_t^{{id}_b}={\displaystyle \frac{Q_t}{{\forall }_{0.3}^{{id}_b}}}$$

(2)

where ${\forall }_{0.3}$ represents the flood volume when the water depth is 0.3 m in the underground space, $\gamma$ refers to the underground risk rate, $Q$ is the inflow volume into the underground space, $L_w$ represents the inflow width (1.2 m), ${id}_b$ represents the unique identifier for each building ($b$), and $h_t$ denotes the above-ground water depth at time $t$. Under the Enforcement Rules of the Act on Guarantee of Convenience Promotion for Persons with Disabilities from the Ministry of Health and Welfare, the minimum effective doorway width is set at 0.9 m. However, in this study, $L_w$ has been adjusted to 1.2 m to provide additional clearance, a width commonly observed in typical doors. Furthermore, while door operation generally becomes impossible at a water depth of 35 cm, a more conservative threshold of 30 cm has been adopted to account for potential difficulties in operation. (Joo, J. [27])

2.4.3. Population

Population is a crucial metric used to gauge the number of individuals who may be impacted by an urban flood. Drawing on statistical data and GIS, researchers can analyze how populations are distributed throughout specific areas. In densely populated locations, the probability of large-scale human casualties increases during a disaster, leading to a higher demand for emergency response resources and personnel. In this study, population data sourced from the National Land Information Platform was employed to establish this indicator.

2.5. Level of FRIB

Sillim-dong in Gwanak-gu, Seoul, accounts for 58% of the district’s land area and half of its population, spread across 11 administrative districts. This region experienced severe flood damage in August 2022 due to exceptionally heavy rainfall, a situation intensified by its low-lying terrain and a concentration of underground residences. To address these challenges, this study established comprehensive criteria to assess the urban flood risk of Sillim-dong by integrating three key indicators—vulnerability, hazard, and exposure—into a single FRIB.

To combine these varied factors into one index, the Euclidean distance method was employed. This approach, serving as a Distance Measure Method (DMM), was chosen to quantify an index by comprehensively considering each factor (Song et al. [28]). Among the various types of DMMs, Euclidean distance has been widely utilized by many researchers to integrate multiple attributes into a single risk score (Song et al. [28]; NEMA [29]; Lee et al. [30]; Cho et al. [31]). By measuring how different indicators vary within a multi-dimensional space, this approach merges vulnerability, hazard, and exposure into a single FRIB score. The relevant formulas are shown below:

$$R_r=\sqrt{(x_r^2+y_r^2+z_r^2)}$$

(3)

$$R=\sqrt{(R_{Exposure}^2+R_{Vulnerability}^2+R_{Hazard}^2)}$$

(4)

where $R$ represents the overall risk; $r$ refers to the Exposure, Vulnerability, and Hazard components; and $x$, $y$, $z$ denote the first, second, and third indicators, respectively.

Through this calculation, the data for vulnerability, hazard, and exposure derived from inundation trace maps created after the 2022 disaster were integrated to quantify the FRIB for Sillim-dong. Finally, the resulting FRIB values for the 11 administrative districts were classified into four categories—very low, low, high, and very high—using Jenks Natural Breaks Classification (JNBC) (Figure 3). JNBC is designed to minimize variance within each category while maximizing variance among them, making it effective for categorizing data into fewer than ten classes (Jenks [32]). This method is widely used in flood research (Lee et al. [33]; Jeung et al. [34]) and serves as a reliable tool for delineating flood risk levels within the study area.

2.6. Validation of FRIB

In this study, a comparative analysis was carried out with existing disaster risk sectors to validate the proposed FRIB based risk assessment methodology. Disaster risk sectors are designated for the systematic management of areas prone to natural disasters, following a comprehensive review of historical disaster data, topographical features, expert evaluations, and other environmental factors. These sectors play a critical policy role by implementing targeted management and preventive measures that protect lives and property.

The 2022 inundation trace map was used to assess risk levels, specifically by identifying the number of “very high” areas in each region. The risk rankings were then determined by comparing these disaster-prone areas with the FRIB-based results from this study. Figure 4 presents (a) the 2022 inundation map and (b) the count of “very high” areas alongside existing disaster risk sectors.

Overall, the FRIB-based assessment showed a similar pattern to the current disaster risk sectors, with the highest-risk zones notably clustered in the left portion of the figure (Figure 4b). This indicates that the proposed method effectively captures high-risk conditions in regions already recognized as disaster-prone, suggesting that existing disaster risk sectors appropriately reflect vulnerable areas. However, some locations deemed high-risk through FRIB analysis were not included in the official disaster risk sectors. Because these areas are not under special management yet appear vulnerable, revisiting the designation criteria or conducting further analysis may be warranted. Incorporating these additional areas could lead to a more precise disaster response strategy.

3. Result

This section presents the quantitative comparison and analysis between the developed FRIB and risk assessments based solely on flood depth. First, the relationship between flood depth and FRIB was analyzed, followed by a detailed discussion of the differences between the two criteria. In particular, the second part not only highlights these differences but also evaluates the potential effectiveness of utilizing FRIB in practical urban flood response strategies.

3.1. Relationship Between Flood Depth and FRIB

By utilizing risk levels rather than solely relying on flood depth as a response criterion, more efficient disaster management strategies can be developed. In this study, the FRIB was derived using the urban flood map from the Ministry of the Interior and Safety′s Safety Map, and its relationship to flood depth was illustrated (Figure 5). Although FRIB has continuous values, for the practicality of disaster management and intuitive understanding, this study classified it into four discrete risk levels—‘very low’, ‘low’, ‘high’, and ‘very high’—using the Jenks Natural Breaks Classification explained in Section 2.5. The frequency of different flood depths for each risk level was then visualized through histograms and pie charts (Figure 6).

In order to examine how flood depths are distributed across different FRIB (Flood Risk Index) levels, this study employed two quantitative metrics—entropy and variance—to measure the uniformity and dispersion of the data, respectively. Entropy $H\left(P\right)$ is defined for a given probability distribution $P=\{p_1,p_2,\dots ,p_n\}$ by Equation (5).

$$H\left(P\right)=-\sum _{i=1}^n{p}_i{\log }_2\left(p_i\right)$$

(5)

where $p_i$ is the proportion of data points falling into category $i$. High entropy values indicate a more even spread of data across multiple categories, whereas low values suggest that the data is heavily concentrated in one or a few categories.

Variance, on the other hand, measures how far individual observations deviate from the mean, reflecting how widely the data is dispersed around its central tendency. It is calculated as Equation (6).

$$Variance={\displaystyle \frac{1}{n}}\sum _{i=1}^n{(x_i-\mu )}^2$$

(6)

where $x_i$ is an individual data point, $\mu$ is the mean of all data points, and $n$ is the total number of observations. High variance corresponds to a broad spread of values, while low variance indicates that the data is clustered tightly around the mean.

Table 4. Entropy and variance by FRIB level.

FRIB Level	Entropy	Variance
Very low	0.423	1777
Low	0.608	1492
High	1.839	347
Very high	2.419	98

summarizes how the distribution of flood depths differs among four FRIB levels—very low, low, high, and very high—by presenting the computed entropy and variance values. The results reveal that at lower FRIB levels (very low and low), an overwhelming majority of data points lie in the shallow flood depth range (<0.5 m), leading to low entropy (indicating an uneven distribution) and relatively high variance (reflecting that the small amount of deeper data points is far from the mean). As the FRIB level increases, entropy rises and variance declines, indicating a more balanced distribution of flood depths across multiple categories. In the very high FRIB level, the highest entropy (2.419) and lowest variance (98) signify that the data is nearly uniform, with no single category dominating the distribution.

These findings demonstrate that deeper flood depths alone do not fully determine a region’s risk level. Instead, a broad spectrum of flood depths—from shallow to deep—contributes to overall flood risk, underscoring the need for inclusive response strategies and infrastructure enhancements. By employing metrics such as entropy and variance, policymakers and urban planners can better understand how different flood depths collectively shape the risk profile at each FRIB level. In turn, this holistic perspective encourages more robust urban flood management strategies that go beyond simply targeting the maximum flood depth.

3.2. Comparison of FRIB-Based and Depth-Based Response Strategies

Using FRIB-based measures instead of traditional flood-depth-based response criteria can lead to more effective and efficient disaster management. To quantify and compare these two approaches, a 500-year frequency flood map was used to evaluate buildings under each criterion. Figure 7 provides a direct visual comparison between the traditional ‘flood-depth-based’ risk classification method and the ‘FRIB-based’ risk classification method proposed in this study. This comparison demonstrates that the FRIB approach can more accurately delineate high-risk areas and reflect nuances for various situations than relying solely on flood depth. Figure 7a shows the levels derived solely from flood depth, following the legend provided by the ‘Urban Flood Map’ within the Ministry of the Interior and Safety’s Safety Map. In contrast, Figure 7b presents risk categories determined by the FRIB methodology.

In a depth-based system, risk is linearly classified using a single indicator (flood depth), which makes it difficult to distinguish among various situations. Additionally, because the legend applies uniform depth thresholds to all areas, the resulting risk categories lack nuance. By contrast, the FRIB approach in this study incorporates multiple factors, enabling more refined differentiation of risk across different regions and facilitating the identification of truly at-risk buildings. For example, under the depth-based method, any building submerged beyond a certain flood depth is uniformly treated at the same risk level. However, with FRIB, even buildings designated as “very high” can be excluded from active response if their water depth is assessed to be relatively low, allowing resources to be allocated more selectively and reducing unnecessary interventions.

Figure 8 expands on a specific area (Section A) of Figure 7, providing a detailed comparison of the quantitative resource allocation efficiency of the two approaches under the assumption of an actual urban flooding scenario. This figure visually demonstrates how FRIB reduces unnecessary interventions, optimizes resource allocation, minimizes resource wastage, and improves disaster response efficiency by showing the number of buildings where such improvements can be achieved. Based on the depth-based measurement (1 m or more), 5124 buildings were identified as requiring response measures. In contrast, when applying FRIB, buildings classified as “low” within that same 1 m threshold were excluded, reducing the target list by 24. If the exclusion criteria are expanded to also omit “high” levels, 530 buildings could be excluded, as shown in

Table 5. The number of high-risk buildings reduced by applying FRIB.

Depth Based Measurement	Excluding FRIB Level Lower Than “Low”	Excluding FRIB Level Lower Than “High”
5124	5100 (−0.5%)	4594 (−10.3%)

. This illustrates how a multi-factor index like FRIB can decrease the number of buildings demanding immediate action and optimize resource allocation.

In a purely depth-based framework, all structures experiencing the specified flood level receive uniform emergency responses, potentially leading to an over-allocation of resources. The FRIB-based approach, however, allows buildings with lower composite risk to be excluded from full-scale responses, ensuring that resources and efforts remain concentrated on genuinely high-risk areas. Consequently, FRIB-based analysis not only enhances disaster response capacity but also provides a more tailored, efficient method for deploying personnel and equipment. Compared to the conventional depth-based model, this FRIB-based methodology offers finer distinctions in risk levels, ultimately playing a crucial role in strengthening urban disaster resilience and minimizing overall damage.

4. Conclusions

This study proposed the Flood Risk Index for Building (FRIB), enabling detailed risk assessment at the building level to improve urban flood risk management. FRIB integrates three key components—vulnerability, hazard, and exposure—to calculate tailored risk levels for individual buildings. Compared to conventional approaches based on simple physical indicators such as flood depth, FRIB offers a more comprehensive and precise analysis of flood risk.

First, the importance of building-level risk assessment was emphasized. Traditional region-based evaluations rely on administrative averages, failing to reflect the specific risk levels of individual buildings. In contrast, FRIB utilizes public data to calculate granular risk indicators at the building level, supporting the development of more precise countermeasures for flood damage.

Second, the study highlighted the necessity of a multi-dimensional approach. Conventional methods relying solely on single indicators such as flood depth often fail to capture interactions between various risk factors, leading to either excessive resource allocation or inadequate responses. FRIB, by integrating multiple factors, provides a balanced risk profile that addresses nuanced risk characteristics beyond the scope of single-indicator models.

Third, FRIB enables efficient resource allocation. By selectively excluding lower-risk buildings identified through FRIB-based analysis, disaster response resources can be concentrated in high-risk areas. This approach significantly reduces resource wastage during emergency response and recovery efforts while enhancing overall effectiveness.

Finally, this study demonstrates the feasibility of multi-dimensional approaches for future flood management. The FRIB effectively overcomes the economic limitations of traditional infrastructure. By identifying and addressing only the vulnerable areas during floods exceeding the design capacity, it provides targeted and localized solutions, which are more efficient and practical compared to large-scale infrastructure. Furthermore, by leveraging public data, FRIB enhances the transparency and accessibility of disaster management, offering actionable guidance to policymakers and urban planners. By integrating social factors such as population density, building age, and access to key facilities with physical indicators, FRIB enables comprehensive risk assessments that significantly contribute to strengthening urban resilience.

In conclusion, this study underscores the need for a multi-dimensional approach in urban flood management and introduces a new strategy for effectively utilizing limited resources. Future research could expand the application of FRIB to various regions and environments. However, since FRIB is developed based on publicly available data in South Korea, its application to other countries may require tailored adjustments depending on the availability and accessibility of local open data. Through such adaptations, it is possible to propose disaster response measures that reflect the unique characteristics of each region, ultimately contributing to the enhanced effectiveness of urban flood mitigation strategies.