• 한국ITS학회 논문지
• /
• 제21권2호
• /
• pp.167-182
• /
• 2022

교통사고보고서 상의 교통사고 관련 자료 속성들은 그 원인을 파악하고자 하는 전통적인 교통안전 관련 연구에서 뿐만 아니라 최근 자율주행자동차의 안전성 검증 시나리오 개발을 위한 연구에서도 활용될 수 있다. 다만, 교통사고보고서의 자료 속성들은 교통사고 상황 재현 및 시나리오 개발만을 위해 정의된 항목들이 아니므로, 본 연구에서는 확대된 활용성 측면의 교통사고보고서의 개선방향을 제시하고자 한다. 교통사고보고서의 개선방향은 각각 교통사고 발생 이전 상황(pre-crash), 교통사고 중 상황(on-crash), 교통사고 발생 이후 상황(post-crash)로 구분하여 제시하였으며, 각 구분에 따른 추가 자료 항목 또는 자료 처리 방안에 대하여 제시하였다. 또한, 정형화된 형태의 교통사고자료 외에 비정형화된 서술 형태의 교통사고 경위자료로부터 추출 가능한 정보항목들을 도출하여 제시하였다.

• 한국안전학회지
• /
• 제29권2호
• /
• pp.76-81
• /
• 2014

Railroad traffic accident consists of train accident, level-crossing accident, traffic death and injury accident caused by train or vehicle, and it is showing a continuous downward trend over a long period of time. As a result of the frequency comparison of train accidents and level-crossing accidents using the railway accident statistics data of Railway Industry Information Center, the share of train accident is over 90% in the 1990s and 80% in the 2000s more than the one of level-crossing accidents. In this study, we investigated time series characteristic and short-term prediction of railroad crossing, as well as seasonal characteristic. The analysis data has been accumulated over the past 20 years by using the frequency data of level-crossing accident, and was used as a frequency data per month and year. As a result of the analysis, the frequency of accident has the characteristics of the seasonal occurrence, and it doesn't show the significant decreasing trend in a short-term.

• 한국항행학회논문지
• /
• 제22권3호
• /
• pp.198-204
• /
• 2018

본 논문에서 국내 항공사고조사를 위한 통합 데이터 관리 시스템 프로토타입 개발을 제안한다. 최근 항공사고 조사 장비의 발전에 따라 사고조사 시스템은 다양한 형태의 jpg, avi, 및 wav 파일 자료들을 수집 및 관리해야한다. 하지만 ECCAIRS 시스템의 경우 항공사고 조사 시 발생하는 다양한 자료를 관리하기 위한 별도의 데이터베이스를 구축하고 있지 못한 상황이며, 국내 항공사고 관리시스템 역시 동일한 문제점을 가지고 있다. 따라서 본 논문에서는 해외 주요 국가의 항공사고보고서 시스템을 분석하고 국내 환경에 적용할 수 있는 방안을 마련한다. 통합 데이터 관리시스템 프로토타입을 통하여 기존 자료와 최근 조사한 자료를 입력을 통하여 성능을 확인하였다. 이 결과를 이용하여 최종 통합 데이터 관리시스템의 완성을 위한 기초 자료로 활용할 예정이다.

• 대한산업공학회지
• /
• 제26권4호
• /
• pp.354-362
• /
• 2000

Increasing amount of road tragic in 90's has drawn much attention in Korea due to its influence on safety problems. Various types of data analyses are done in order to analyze the relationship between the severity of road traffic accident and driving conditions based on traffic accident records. Accurate results of such accident data analysis can provide crucial information for road accident prevention policy. In this paper, we apply several data fusion, ensemble and clustering algorithms in an effort to increase the accuracy of individual classifiers for the accident severity. An empirical study results indicated that clustering works best for road traffic accident classification in Korea.

• 한국자동차공학회논문집
• /
• 제22권2호
• /
• pp.29-36
• /
• 2014

The accident statistics use the data from police accident reports and statistics. Although the official accident statistics are useful, they provide very limited information about how accidents occur, the cause of the accident and the injury mechanisms. This limitations could be overcome by carrying out the in-depth accident study and analysing investigations, collecting more detailed information. Meanwhile a net of in-depth investigation teams are operating worldwide, such as NASS (National Accident Sampling System) and CIREN (Crash Injury Research and Engineering Network) in US, OTS (On the spot investigation) in UK. In this study, the database structure and variables of Korea in-depth accidents investigation system would be proposed through considering the database structure of GIDAS (Germany In-Depth Accidents Study). GIDAS is one of the best system on the in-depth accident study system in the world. GIDAS was established in 1999 as a cooperation project between Federal Highway Research Institute of Germany (BASt) and research association on automotive engineering of German Car Industry(FAT). The iGLAD (Initiative for the Global Harmonization of Accident Data) was also considered to introduce into the database variables of Korea in-depth accident study. Current police reports were compared with GIDAS and iGLAD. To collaborate of the Worldwide in-depth accident data, this paper proposed that the database of Korea in-depth accident study would be introduced the structure of GIDAS and the core variables of iGLAD. Harmonization of the structures and core variables of Korea in-depth accident study will be better than the making unique ones. The database structure and core variables of KIDAS(Korea In-Depth Accident Study) introduced of GIDAS and iGLAD.

• 한국도로학회논문집
• /
• 제16권4호
• /
• pp.111-118
• /
• 2014

PURPOSES: The purposes are to analyze the pedestrian accident severity and to develop the accident models by arterial road function. METHODS: To analyze the accident, count data and ordered logit models are utilized in this study. In pursuing the above, this study uses pedestrian accident data from 2007 to 2011 in Cheongju. RESULTS : The main results are as follows. First, daytime, Tue.Wed.Thu., over-speeding, male pedestrian over 65 old are selected as the independent variables to increase pedestrian accident severity. Second, as the accident models of main and minor arterial roads, the negative binomial models are developed, which are analyzed to be statistically significant. Third, such the main variables related to pedestrian accidents as traffic and pedestrian volume, road width, number of exit/entry are adopted in the models. Finally, Such the policy guidelines as the installation of pedestrian fence, speed hump and crosswalks with pedestrian refuge area, designated pedestrian zone, and others are suggested for accident reduction. CONCLUSIONS: This study analyzed the pedestrian accident severity, and developed the negative binomial accident models. The results of this study expected to give some implications to the pedestrian safety improvement in Cheongju.

• 한국안전학회지
• /
• 제32권5호
• /
• pp.179-185
• /
• 2017

Accident analysis is an essential process to make basic data for accident prevention. Most researches depend on survey data and accident statistics to analyze accidents, but these kinds of data are not sufficient for systematic and detailed analysis. We, in this paper, propose an accident classification model that extracts task type, original cause materials, accident type, and the number of deaths from accident reports. The classification model is a support vector machine (SVM) with word occurrence features, and these features are selected based on mutual information. Experiment shows that the proposed model can extract task type, original cause materials, accident type, and the number of deaths with almost 100% accuracy. We also develop an accident ontology to express the information extracted by the classification model. Finally, we illustrate how the proposed classification model and ontology effectively works for the accident analysis. The classification model and ontology are expected to effectively analyze various accidents.

• 한국응급구조학회지
• /
• 제11권1호
• /
• pp.65-72
• /
• 2007

The purpose of this study was teacher's perception on young children's safety life, safety accident, and safety education and provided basic data of administrating teacher's education for young children's safety. Subjects of this study were teachers of young children attending for their education. 230 questionnaires were provided and 181 were collected and 170 were used for data analysis. Data collected were analyzed with SPSS WIN 2.0 program. The results of the study were as follows : 1. Regarding teacher's perception on types of young children's safety accident, play accident was high(70.0%). 2. With regard to teacher's perception on causes of accident, lacks of perception was high(64.1%). 3. Of transportation means in accident, 119 ambulance use was high(60.5%) 4. Regarding teacher' perception on accident prevention, direct attention of education by paramedics was high(48.2%).

• ETRI Journal
• /
• 제44권4호
• /
• pp.654-671
• /
• 2022

Accidental falls frequently occur during activities of daily living. Although many studies have proposed various accident detection methods, no high-performance accident detection system is available. In this study, we propose a method for integrating data and accident detection algorithms presented in existing studies, collect new data (from two stunt performers and 15 people over age 60) using a developed wearable device, demonstrate new features and related accident detection algorithms, and analyze the performance of the proposed method against existing methods. Comparative analysis results show that the newly defined features extracted reflect more important risk factors than those used in existing studies. Further, although the traditional algorithms applied to integrated data achieved an accuracy (AC) of 79.5% and a false positive rate (FPR) of 19.4%, the proposed accident detection algorithms achieved 97.8% AC and 2.9% FPR. The high AC and low FPR for accidental falls indicate that the proposed method exhibits a considerable advancement toward developing a commercial accident detection system.

• Nuclear Engineering and Technology
• /
• 제44권4호
• /
• pp.393-404
• /
• 2012

After the Fukushima nuclear accident in 2011, there has been increasing concern regarding severe accidents in nuclear facilities. Severe accident scenarios are difficult for operators to monitor and identify. Therefore, accurate prediction of a severe accident is important in order to manage it appropriately in the unfavorable conditions. In this study, artificial intelligence (AI) techniques, such as support vector classification (SVC), probabilistic neural network (PNN), group method of data handling (GMDH), and fuzzy neural network (FNN), were used to monitor the major transient scenarios of a severe accident caused by three different initiating events, the hot-leg loss of coolant accident (LOCA), the cold-leg LOCA, and the steam generator tube rupture in pressurized water reactors (PWRs). The SVC and PNN models were used for the event classification. The GMDH and FNN models were employed to accurately predict the important timing representing severe accident scenarios. In addition, in order to verify the proposed algorithm, data from a number of numerical simulations were required in order to train the AI techniques due to the shortage of real LOCA data. The data was acquired by performing simulations using the MAAP4 code. The prediction accuracy of the three types of initiating events was sufficiently high to predict severe accident scenarios. Therefore, the AI techniques can be applied successfully in the identification and monitoring of severe accident scenarios in real PWRs.