• 한국안전학회지
• /
• 제24권1호
• /
• pp.69-77
• /
• 2009

This paper reports the aging driver traffic accident severity modeling results. For the modeling, Poisson regression approach is applied using the data set obtained from the Korea Transportation Safety Authority's simulator-based driver aptitude test results. The test items include the estimations of moving objects' speed and stopping distance, drivers' multi-task capability, and kinetic depth perception and so on. The resulting model with the response variable of equivalent property damage only(EPDO) indicated that EPDO is significantly influenced by moving objects' speed estimation and drivers' multi-task capabilities. More interestingly, a comparison with the younger driver model revealed that the degradation of such capabilities may result in severer crashes for older drivers as suggested by the higher estimated parameters for the older driver model.

• 대한토목학회논문집
• /
• 제34권2호
• /
• pp.561-571
• /
• 2014

전라북도내 14개 시 군의 교통사고 자료를 활용하여 고령운전자의 교통사고 피해 심각성을 분석하였다. 교통사고는 1차적으로 개인 및 운전환경 속성과 2차적으로 도시관련 속성에 의해 영향을 받는 2단계 위계적 특성을 갖는 것으로 가정하였다. 위계적 특성을 고려한 피해 심각성에 대한 영향요인을 분석하기 위해 다수준분석모형을 활용하였다. 분석결과로서 65세 이후의 고령운전자는 연령이 증가할수록 교통사고로 인한 피해상황이 심각해짐을 보여주며 안전한 운전방법의 교육과 교통사고를 미연에 방지하기 위한 대안이 필요하다. 음주운전은 고령운전자에게 사고발생시 피해 심각성을 크게 할 경향이 높은데, 사망사고에 있어서 비고령자에 비해 발생비율이 약 3.0배 이상 높았다. 고령운전자는 야간 교통사고 발생빈도가 높은 편이나, 낮 시간대의 교통사고일수록 피해 심각성은 높아졌다. 고령운전자는 비고령자보다 흐린 날씨에서 사고 발생빈도가 높으나, 심각성에서는 맑은 날에 높아짐을 보였다. 습윤상태의 노면이 피해 심각성에 큰 영향을 주고 있는데, 비고령자에 비해서 고령운전자가 중상 및 사망비율도 높은 것으로 분석되었다.

• 한국안전학회지
• /
• 제26권2호
• /
• pp.89-98
• /
• 2011

The traffic accidents at a construction site, which happen due to construction vehicles' frequent access to a construction site, its subsequent conflicts with ordinary vehicles and pedestrians, and inappropriate installation & management of traffic security facilities, have not many proportions in all traffic accidents, but obviously, the accident damage is quite serious when comparing the level of the fatal per one accident. This research conducted an analysis of traffic accident injury severity using Ordered Probit Model in relation to 241 traffic accident cases that occurred caused by construction sites among the traffic accidents that took place in Seoul and Gyeoggi-do region for two years from 2006 until 2007. As a result, the significant variables enough to explain traffic accident injury severity were analyzed to be the state of road surface, linear shape of an accident spot & whether the damaging car belongs to the vehicle for construction, and whether vehicles have access to a construction site at the time of an accident. Through this, this research found out some fact as follows: first, there need to be more aggressive management of the vehicles for construction and a year-round placement of the manpower who can control vehicular access to a construction site. Second, it is necessary to get drivers to recognize the fact that there exists a construction site on the construction section which is on the border of curved roads in advance to prevent a traffic accident, helping to reduce socioeconomic loss & costs incurred by a traffic accident.

• 대한안전경영과학회지
• /
• 제17권1호
• /
• pp.1-11
• /
• 2015

The efficient safety estimation for a business should analyze an accident data by considering every possible and potential factor. Thus, we consider several factors to build the safety estimation model to meet fairness and rationality. This paper present the yearly statistic data of accident from KOSHA analyze the data by industry, scale, year of service of a employee, age and other factors; build the safety estimation model for the business based on the accident report derived the analysis. The estimation model is established by the weights for accident type, degree, scale, industry, year of service, and age of the employee derived from ANP(Analytic Network Process).

• 한국안전학회지
• /
• 제27권3호
• /
• pp.161-166
• /
• 2012

This study deals with the accident of roundabout. The objective is to analyze the traffic accidents occurred in 3-legged and 4-legged roundabouts through the developed models. In developing the multiple linear regression models, this study uses the number of traffic accidents as a dependent variable and such the variables as geometric structures, traffic characters and others as the independent variables. The correlation and multicollinearity of variables were analyzed using SPSS17.0. The main results are as follows. First, R-square value of developed models were analyzed to be 0.851(3-leg) and 0.689(4-leg), respectively. Second, the independent variables in the 3-legged roundabout accident model were analyzed to be the traffic volume and number of crosswalk, and the variables in the 4-legged roundabouts were evaluated to be the traffic volume and signal. Finally, the paired t-test shows that the predicted values and observed values are not statistically different.

• 대한안전경영과학회지
• /
• 제12권2호
• /
• pp.1-7
• /
• 2010

The efficient safety estimation for a business should analyze an accident data by considering every possible and potential factor. Thus, we consider several factors to build the safety estimation model to meet fairness and rationality. This paper present the yearly statistic data of accident from KOSHA analyze the data by industry, scale, year of service of a employee, age and other factors; build the safety estimation model for the business based on the accident report derived the analysis. The estimation model is established by the weights for accident type, degree, scale, industry, year of service, and age of the employee derived from AHP(Analytic Hierarchy Process).

• 한국재난정보학회 논문집
• /
• 제9권1호
• /
• pp.30-49
• /
• 2013

본 연구에서는 토지이용에 기반을 두는 새로운 교통사고 예측모형을 개발하였다. 다양한 지역의 특성을 반영할 수 있는 변수에 대한 시장분할 및 추가변수 도입을 토대로 Data Mining 기법의 하나인 의사나무결정법(Classification and Regression Tree)을 활용하여 새로운 유형별 교통사고 예측모형을 개발하였다. 분석결과를 살펴보면 주민등록인구수, 통근 등 활동변수와 활동의 대상이 되는 도로규모, 유발시설 등이 교통사고를 설명하는 변수로 도출되었다.

• 한국안전학회지
• /
• 제26권4호
• /
• pp.120-126
• /
• 2011

Under the current law or system, the range of traffic accident on the crosswalk does not reflect the characteristics of traffic accident and the pedestrian's walking pattern. Thus, this study conducted a video recording survey on the 250 spots which are high to traffic accident rate of pedestrian-vehicle to reset the range of traffic accident on or near the crosswalk considering the characteristics of traffic accident and the pedestrian's walking pattern. Based on the collected data through a video recording survey, this study analyzed the pattern of pedestrians and extracted the variables influenced in the pedestrian's walking pattern. After conducting the regression analysis, this study made the model of measuring the range of traffic accident on the crosswalk. Through all processes these, this study reset the range of traffic accident on the crosswalk which could minimize the disadvantages of pedestrian when they have an accident on the crosswalk and ensure the right of way of pedestrian.

• 대한교통학회지
• /
• 제27권3호
• /
• pp.141-148
• /
• 2009

용량의 산정이 적합하지 못하다면 고속도로의 분석 및 예측에 잘못된 결론을 도출하게 된다. 도로의 용량을 일시적으로 크게 감소시키는 교통사고는 예측 불가능한 비반복정체를 발생시켜 혼잡관리가 어렵다. 따라서 본 연구는 사고발생시 속도에 따른 도로용량 파악을 목적으로 한다. 본 연구에서는 사고발생구간의 차량의 행태를 파악하여 교통류 속도에 따른 임계차두간격 산출모형과 최대통과교통량 산출모형을 구축하였다. 구축된 모형을 토대로 사고발생시의 고속도로용량감소율을 산정하였다. 그 결과 교통류의 속도가 40km/h일 때, 도로용량이 37%감소할 것으로 예측되었고, 다른 결과값은 본문에 수록하였다. 구축된 모형에 대한 검증은 제대로 수행할 수 없었지만 속도에 따른 도로용량감소율을 파악하고자 했다는데 본 연구의 의의를 두고 싶다.

• 한국도로학회논문집
• /
• 제20권2호
• /
• pp.67-74
• /
• 2018

PURPOSES : The goal of this study is the development of roundabout accident models for urban and non-urban areas. METHODS : This study performed a comparative analysis of the regional factors affecting accidents. Traffic accident data were collected for the period 2010~2014 from the TAAS data set of the Road Traffic Authority. To develop the roundabout accident models, the Poisson and negative binomial regression models were used. A total of 25 explanatory variables such as geometry, and traffic volume were used. RESULTS : The key findings are as follows: First, it was found that the null hypotheses that the number of accidents is the same should be rejected. Second, three Poisson regression accident models, which are statistically significant (${\rho}^2$ of 0.154 and 0.385) were developed. Third, it was noted that although the common variable of the three models (models I~III) is the number of entry lanes, the specific variables are entry lane width, roundabout sign, number of circulatory roadways, splitter island, number of exit lanes, exit lane width, number of approach roads, and truck apron. CONCLUSIONS : The results of this study can provide suggestive countermeasures for decreasing the number of roundabout accidents.