• Journal of Korean Society of Transportation
• /
• v.16 no.4
• /
• pp.187-194
• /
• 1998

본 연구는 교통사고 심각도와 관련된 중요변수를 찾고 이들 변수를 바탕으로 신경망, Decision Tree, 로지스틱 회귀분석을 이용하여 사고 심각도 분류 예측모형을 추정하였다. 다수의 범주형 변수로 이루어진 교통사고 통계원표상의 설명변수 들로부터 사고 심각도 변화에 영향력 있는 변수 선택을 위하여 독립성 검정을 위한 $x^2$ test와 Decision Tree를 이용하였고, 선택된 변수들은 신경망과 로지스틱 회귀분석의 기초로 이용되었다. 분석결과 세가지기법간에 분류정확도에는 유의한 차이가 없는 것으로 나타났다. 그러나 Decision Tree가 설명변수 선택능력과 분석수행시간, 사고 심각도 결정요인 식별의 용이함 측면에서 범주형 종속변수인 사고 심각도의 분석에 적합한 것으로 보이며 사고 심각도에는 보호장구가 가장 큰 영향을 미치는 것으로 재입증되었다.

• Proceedings of the KOR-KST Conference
• /
• 1998.10a
• /
• pp.373-381
• /
• 1998

본 연구는 교통사고 심각도와 관련된 중요변수를 찾고 이들 변수를 바탕으로 신경망, Decision Tree, 로지스틱 회귀분석을 이용하여 사고 심각도 분류 예측모형을 추정하였다. 다수의 범주형 변수로 이루어진 교통사고 통계원표상의 설명변수 들로부터 사고 심각도변화에 영향력 있는 변수선택을 위하여 $X^2$ 독립성 검정과 Decision Tree를 이용하였고, 선택된 변수들은 신경망과 로지스틱 회귀분석의 기초로 이용되었다. 분석결과 세가지기법간에 분류정확도에는 유의한 차이가 없는 것으로 나타났다. 그러나 decision Tree가 설명변수 선택능력과 분석수행시간, 사고 심각도 결정요인 식별의 용이함 측면에서 범주형 종속변수인 사고 심각도의 분석에 적합합 것으로 보이며 사고 심각도에는 보호장구가 가장 큰 영향을 미치는 것으로 재입증되었다.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.14 no.6
• /
• pp.91-99
• /
• 2015

This study intends to build a traffic accident predictive model considering road geometrics, traffic and enviromental characteristics and identify the relationship of 4-legs intersection accidents in Seoul and Busan metropolitan area. The RPNB(Random Parameter Negative Binomial) model shows improvement over the fixed NB(Negative Binomial) and out of 53 variables, 10 variables (main road number of lane, main road vehicle traffic volume(left), minor road vehicle traffic volume(right), main road drive restriction, minor road sight distance, minor road median strip, minor road speed limit, minor road speed restriction) showed to have significant variables affecting traffic accident occurrences in 4-legs signilized intersections. Also, among 10 significant variables, 2 variables(minor road sight distance, minor road speed restriction) found to be random parameters.

• Journal of Korean Society of Transportation
• /
• v.24 no.2 s.88
• /
• pp.139-145
• /
• 2006

This paper develops pedestrian fatality models capable of producing the probability of pedestrian fatality in collision between vehicles and pedestrians. Probabilistic neural network (PNN) and binary logistic regression (BLR) ave employed in modeling pedestrian fatality pedestrian age, vehicle type, and collision speed obtained from reconstructing collected accidents are used as independent variables in fatality models. One of the nice features of this study is that an iterative sampling technique is used to construct various training and test datasets for the purpose of better performance comparison Statistical comparison considering the variation of model Performances is conducted. The results show that the PNN-based fatality model outperforms the BLR-based model. The models developed in this study that allow us to predict the pedestrian fatality would be useful tools for supporting the derivation of various safety Policies and technologies to enhance Pedestrian safety.

• Journal of Korean Society of Transportation
• /
• v.27 no.5
• /
• pp.135-144
• /
• 2009

This study estimates the expected number of accidents in Kyungbuk Province to capitalize on experience gained from four years of accident history using the Empirical Bayes (EB) Method. The number of accidents of each site in Kyungbuk Province is recalculated using the Equivalent Property Damage Only (EPDO) method to reflect the severities of the accidents. A cluster analysis is performed to determine similar sites and a unique Safety Performance Function (SPF) is established for each site. The overdispersion parameter is built to correct the difference between the actual number of accidents and the underlying probability distribution. To adjust for varying traffic characteristics of each site, a relative weight is applied and eventually estimates the expected number of accidents. The results show that the highest accident sites are Kimcheon, Youngcheon, and Chilgok, but on the other hand the lowest is Gunwi.

• Journal of Korean Society of Transportation
• /
• v.20 no.7
• /
• pp.87-94
• /
• 2002

2001년 경찰청 $\ulcorner$2001년판 교통사고통계$\lrcorner$ 자료에 의하면 사고피해의 정도를 나타내는 치사율에 있어서 중앙선침범사고가 과속사고와 함께 가장 높은 수치를 보이고 있으며, 대형 교통사고의 39.5%가 중앙선 침범에 의해 발생한 것으로 나타났다. 실제로 2000년 한 해 동안 총 사고건수 290,481건 중 중앙선 침범에 의해 18,931건 (6.52%)의 교통사고가 발생하여 사망자 1,472명, 부상자 35,046명으로 전체 교통사고 사망자의 14.38%, 부상자의 8.21%를 점하고 있는 것으로 나타났다. 이와 같이 전체 교통사고에서 중앙선침범사고가 차지하는 비중 이 높아지고 있어 중앙선침범사고를 줄이기 위한 대책의 일환으로 중앙분리대, 표지병, 시선 유도봉 등과 같은 중앙선침범 예방시설물을 설치하고 있으나, 해당 시설물에 대한 설치기준이 없는 실정이다. 이렇듯 중앙분리대 설치기준의 부재로 인해 중앙선침범사고 발생과 이에 따르는 경제적인 교통사고 비용손실, 그리고 중앙분리대의 과다설치로 인한 국가 재원의 낭비는 간과할 수 없는 문제점을 내포하고 있다. 이러한 점에 착안하여 기존에 지방부 2차로 도로의 직선부에 대한 중앙선침범사고 예측모델이 개발되었고 본 연구에서는 그 연장선상에서 지방부 2차로 도로의 곡선부에 대한 중앙선침범사고 예측모델에 대한 중앙선침범사고 예측모델의 개발을 통해 중앙선침범사고에 대한 사회적 비용을 산정하고, 현재 설치하여 운영중인 중앙선침범 예방시설물의 설치비용간의 비용-편익분석을 통해 얻어진 기준으로 중앙선침범 예방시설물 설치의 가부에 대한 보다 현실적이고 비용-효율적인 중앙선침범 예방시설물 설치기준을 제시함으로써 교통사고 비용손실을 최소화하고, 중앙분리대의 효율적 설치방안의 도입을 통한 경제적 이익창출을 도모하고자 한다.

• International Journal of Highway Engineering
• /
• v.12 no.4
• /
• pp.1-9
• /
• 2010

Predictive information on the freeway incident impacts can be a critical criterion in selecting travel options for users and in operating transportation system for operators. Provided properly, users can select time-effective route and operators can effectively run the system efficiently. In this study, a model is proposed to predict freeway incident impacts. The predictive model for incident impacts is based on short-term prediction. The proposed models are examined using MARE. The analysis results suggest that the models are accurate enough to be deployed in a real-world. The development of microscopic models to predict incident effects is expected to help minimize traffic delay and mitigate related social costs.

• The Korean Journal of Applied Statistics
• /
• v.30 no.1
• /
• pp.181-193
• /
• 2017

The paper introduces bottom-up and optimal combination methods that can analyze and forecast hierarchical time series. These methods allow forecasts at lower levels to be summed consistently to upper levels without any ad-hoc adjustment. They can also potentially improve forecast performance in comparison to independent forecasts. We forecast regional traffic accident counts as time series data in order to identify efficiency gains from hierarchical forecasting. We observe that bottom-up or optimal combination methods are superior to independent methods in terms of forecast accuracy.

• International Journal of Highway Engineering
• /
• v.14 no.2
• /
• pp.131-143
• /
• 2012

The past researches on roadway segment safety estimation focused on intersections, which are the primary traffic accident regions. The past researches on roadway segments, However, analyzed the effects of certain factors on the traffic accident occurrence rate by organizing the individual geometric structures of the roads, and there is still a dearth of researches on the development of a traffic accident estimation model for rural roadway segments. Therefore, this research focused on rural two-lane and multilane roadway segments and developed traffic accident estimation models through the application of statistical techniques. This is required to explain such high frequency of zero counts in the traffic accident data. In this research, it was found that the Hurdle model is more suitable than the Poisson or negative binomial-regression model for explaining the excess zeros case. In addition, main variables were chosen to estimate their effects on traffic accident occurrence at rural roadway segments, and the safety at such rural roadway segments was estimated. In this research, it was assumed that there are different factors that affect the safety at two-way lane and multilane roadway segments, and a traffic accident estimation model was developed by dividing the two-way lane and multilane roadway segments.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.31 no.3D
• /
• pp.343-353
• /
• 2011

The unsignalized intersections in a community road in the city of Seoul have 3,753 traffic accidents(9%) of total 41,702 cases in 2008, not high in the occurrence rate of traffic accidents, but seem to have a quite high potential of accidents due to the unreasonable and insufficient operation of systems and facilities in the part of traffic foundations. In particular, the un-signalized intersections in a community road have an insufficient measure for safety as compared to the crossroads with signals, and there are few analysis of traffic accidents and domestic researches on the model of affecting factors. Our country also has no concept of passing priority in operating a crossroad without signals, differently from foreign countries, so the researches and safety measures for improving the safety of a crossroad without signals in a community road are urgent. Therefore, This study set out to analyze the road conditions, traffic conditions, and traffic environment conditions on unsignalized intersection, to identify the elements that would impose obstructions in safety, and develop a traffic accident prediction model to evaluate the safety of an unsignalized intersection using the correlation between the elements and an accident. In addition, the focus was made on suggesting appropriate traffic safety policies by dealing with the danger elements in advance and on enhancing the safety on intersection in developing a traffic accident prediction model for an unsignalized intersection.