• Journal of Korean Society of Disaster and Security
• /
• v.14 no.3
• /
• pp.17-27
• /
• 2021

In this study, classification models were built using machine learning techniques that can classify the soil creep risk into three classes from A to C (A: risk, B: moderate, C: good). A total of six machine learning techniques were used: K-Nearest Neighbor, Support Vector Machine, Logistic Regression, Decision Tree, Random Forest, and Extreme Gradient Boosting and then their classification accuracy was analyzed using the nationwide soil creep field survey data in 2019 and 2020. As a result of classification accuracy analysis, all six methods showed excellent accuracy of 0.9 or more. The methods where numerical data were applied for data training showed better performance than the methods based on character data of field survey evaluation table. Moreover, the methods learned with the data group (R1~R4) reflecting the expert opinion had higher accuracy than the field survey evaluation score data group (C1~C4). The machine learning can be used as a tool for prediction of soil creep if high-quality data are continuously secured and updated in the future.

• Journal of Korean Society of Disaster and Security
• /
• v.17 no.2
• /
• pp.1-11
• /
• 2024

Most reservoirs in South Korea are earthen dams, mainly because they are cost-effective and easy to construct. However, earthen dams are highly vulnerable to seepage and overtopping, making them prone to sudden failure during excessive flooding. Such sudden failures can lead to a rapid increase in flood discharge, causing significant damage to downstream rivers and inhabited areas. This study investigates the effect of riprap placement on the slopes of earthen dams in delaying dam failure. Delaying the failure time is crucial as it allows more time for evacuation, significantly reducing potential casualties, which is essential from a disaster response perspective. Hydraulic experiments were conducted in a straight channel, using two different sizes of riprap for protection. Unlike previous studies, these experiments were performed under unsteady flow conditions to reflect the impact of rising water levels inside the dam. The target dam for the study was a cofferdam installed in a diversion tunnel. Experimental results indicated that the presence of riprap protection effectively prevented slope failure under the tested conditions. Without riprap protection, increasing the size of the riprap delayed the failure time. This delay can reduce peak discharge, mitigating damage downstream of the dam. Furthermore, these findings can serve as critical reference material for establishing emergency action plans (EAP) for reservoir failure.

• Fire Science and Engineering
• /
• v.27 no.2
• /
• pp.18-24
• /
• 2013

Goverment and enterprises have been devoting a lot of dffort in order to prevent the loss of human and property due to a large, diversified fire, explosion, global warming and disasters caused by natural phenomena. But Enterprises came forward and running in the economy priority policy for profit-seeking until now in order to meet the macro purpose of quality improvement of national life and welfare promotion in company with National growth, as a result of lax risk management have had to endure the loss of precious lives and property. According to the '2007 Survey of public safety for the activation of the safety culture' of National Emergency Management Agency [NEMA], a insensitivity of safety in our society was surveyed 'a serious level'. In this study, surveyed need of 'step-by-step fire protection safety management system builds' in order to prevent fire safety accident and improvement of safety awareness level through a systematic management and real condition estimation of not only large business that can control risk management, but also small business that has a big risk to occur accident easily according to small budget and organization.

• Journal of Korean Society of societal Security
• /
• v.2 no.1
• /
• pp.81-89
• /
• 2009

A web-based information management system to share engineering data of truss bridge is developed through construction of standardized database of truss bridge. 3D shape information is stored in database according to AP 203 of STEP, and 3D visualization on the web is implemented by using the web 3D technology that helps users to understand geometrical shape of structures, directly. AP209 is used to store structural analysis information such as finite elements, material properties, and analysis result into relational database. Based on the developed database, a prototype of integrated information management system for truss bridge is developed, and it provides additional information such as specifications and inspection information related with shape object to end users.

• Journal of Korean Society of societal Security
• /
• v.2 no.1
• /
• pp.65-74
• /
• 2009

In this study, the air pollution management system based GIS has been developed to estimate the emission rate and air pollution modeling of air pollutants, effectively. This system is able to estimate emission rate of air pollutant and to analyze the emission characteristics with high spatial and temporal resolution. air pollution modeling. The air pollution management system was applied to Gwangyang Bay including large industry complex with a large number of emission sources. The air pollution management system was constructed using the spatial database of emission sources in Gwangyang Bay. It was found that the estimated emission rates of air pollutants is similar to the emission characteristics in Gwangyang Bay. Also, the spatial distribution of pollutants was similar to the location of emission sources. The predicted results of air pollution model was showed a good correlation coefficient (0.75) for TSP. The air pollution management system is expected to be effective tool (database system (GIS)) for the management and the control of air pollution.

• Journal of Korea Multimedia Society
• /
• v.15 no.1
• /
• pp.93-100
• /
• 2012

Recently, the security monitoring and control systems based on spatial information in various field are operated and being developed according to evolve the spatial information technology. Especially, the CCTV monitoring and control system can be used in various field as a typical system. However, the security vulnerability problems have become an issue because the system connected by computer network and getting bigger than before. Therefore we studied security vulnerabilities of CCTV monitoring and control system which is being developed and operated. In addition, it is important to consider disaster and terrorism with unauthorized changes on location information. Therefore we analyzed the performance of observation when the cameras are break down as a result by hacking to CCTV monitoring and control system.

• Spatial Information Research
• /
• v.19 no.4
• /
• pp.1-10
• /
• 2011

This research examined relationships between Scientifically Estimated Environmental Risks (SEERs) of floods, hurricanes and hazardous material releases, and household characteristics. This research showed that there were no statistically significant relationships between most of the demographic characteristics (age, household size, tenure at the present home) and SEERs of the two natural hazards (a flood and a hurricane). These results support Drabek's findings (1986) that people tend to underestimate or ignore natural hazards in selecting their residence regardless of age, household size, and house tenure. Educational attainment and yearly household income were positively correlated with hurricane risk, but not with flood risk. By contrast, SEER of hazardous materials was correlated with all demographic characteristics mentioned above. This result may show that those who are relatively poorer and have lower educational level tend to be limited to living in communities more vulnerable to human-made risk.

• Journal of Korea Spatial Information System Society
• /
• v.10 no.4
• /
• pp.103-107
• /
• 2008

This study mainly focused on the development of real-time based damage investigation system using web-GIS. The study area was Gangseo-Gu in Busan where frequently has relatively higher magnitude of damages from natural disasters. GIS DB was built to provide geospatial data such as 1:1,000 and 1:5,000 topographic maps and IKONOS high resolution satellite data. The web-GIS system has demonstrated the higher contribution for the better response and recovery against any type of natural disasters through real-time communication and data dissemination among government agencies and personnel. However, further researches need to be made to improve system capabilities. In addition, for more effective system operation and maximizing the benefits of exploiting web-GIS system, linkage with national DB such as NDMS is essential.

• Proceedings of the Korea Society for Energy Engineering kosee Conference
• /
• 2008.04a
• /
• pp.253-258
• /
• 2008

우리나라에서 원자력 시설의 안전 확보를 위한 현장 규제는 원전 부지내에 설치되어 운용되고 있는 주재관실과 방사능 재난 발생에 대응하기 위한 현장방사능방재지휘센터 등으로 분산되어 있다. 원유 고유가, 에너지 안보, 및 기후 변화 협약 등에 대한 대응책으로, 신규 원자력 시설의 건설이 진행 중에 있으며, 이에 따라 원자력 규제 대상이 증가하고 있다. 지역주민이 신뢰하는 원자력 시설의 안전 규제의 필요성에 따라 현장 규제의 업무량도 증가하고 현장 규제 요건도 강화되고 있는 추세이다. 분산되어 있는 원자력 안전규제 시스템을 통합적인 체계로 전환하기 위한 방안들 가운데, 이 연구에서는 규제의 효율성/효과성을 향상시키기 위해 원자력 안전 지역사무소의 설치가 고려되었다. 이 연구의 주요 목적은 시스템 다이내믹스 기법을 활용한 지역사무소 설치의 타당성 모델을 제시하는 것이다. 이 SD 모델을 통해 지역사무소 설치가 규제 효과성을 포함한 다른 변수들에 미치는 영향을 전체론적 (holistic) 관점에서 평가할 수 있다. 이 SD 모델에서 이해관계자 (예: 중앙행정기관, 규제기관, 지방자지단체, NGO, 국민 등) 각각의 우선적인 목표는 갈등적 다중목표가 된다. 이러한 다중목표는 원자력 안전 규제의 의사결정 단계에서 거버넌스(governance) 체제 (예: 규제 기관, 사업자, 지역주민 등이 참여하는 체제)의 구축, 원자력 관련 민원 해결을 통해 성취될 수 있다. 시스템 다이내믹스 접근법의 주요 변수 정의 단계에서는 다양한 관점의 상호영향을 고려하면서 지역사무소 설치에 필요한 기본 개념이 도출되었다. 시스템 다이내믹스 접근법의 모델링 단계에서는 이들 기본 개념에서부터 지역사무소 설치 타당성 정책의 인과 지도(causal map)가 작성되었다. 이러한 인과 지도들로부터 지역사무소의 설치 타당성으로 이끄는 다양한 자기-강화 (self-reinforcing) 루프들이 발견되었다.

• Journal of Korean Society of societal Security
• /
• v.1 no.2
• /
• pp.61-66
• /
• 2008

More than 70% of Korea consists of mountainous area and during the construction of roads and railroads many cut-slopes are inevitably formed. A number of environmental factors, such as the rainy season and frost heave during winter/thaw during spring, can result in rock falls and landslides. The failure of slopes is increasing every year and can cause damage to vehicles, personal injury and even fatality. In order to help protect people and property, there is a need for real-time monitoring systems to detect the early stages of slope failures. In this respect, the GMG has been using Translation Rotation Settlement (TRS) sensor units installed on slopes to monitor movement in real-time. However, the data lines of this system are vulnerable and the whole system can be damaged by a single lightning strike. In order to overcome this, GMG have proposed the use of Ubiquitous Sensor Networks (USN). The adoption of a USN system in lieu of data cables can help to minimize the risk of lightning damage and improve the reliability of slope monitoring systems.