• Title/Summary/Keyword: Traffic accident prediction model

Search Result 84, Processing Time 0.02 seconds

Analysis of Bus Accidents Influential Factors on Bus Exclusive Lane in Seoul (Bus Median Lane and Bus Curb Lane Defined) (서울시 버스전용차로구간의 버스사고 영향요인 분석 연구 (중앙전용차로 및 가로변전용차로 구분))

  • Lim, Jun-Beom;Hong, Ji-Yeon;Chang, Il-Jun;Park, Jun-Tae
    • International Journal of Highway Engineering
    • /
    • v.14 no.2
    • /
    • pp.145-155
    • /
    • 2012
  • At present, Seoul City is putting the bus exclusive lane system into practice according to mass transit revitalization policy. Starting with the installation of roadside bus exclusive lane in the past, at present, even the road sections for central- lane bus exclusive lane system are on the increase. The purpose of this research is to analyze the factors giving impacts on bus accident on central bus exclusive lane and roadside bus exclusive lane. In case of the central bus exclusive lane, the 6 variables, such as the number of bus routes, number of access & entrance to central lanes patterns, whether the stop line of central lanes retreats or not, separated distance between the stop line of central lanes and crosswalks, traffic volume, and number of bus routes stopping at bus stops on reversible lanes, were found to have a significant influence on bus accidents. In case of roadside bus exclusive lane sections, the four variables such as the number of right-turn bus routes, whether to be chronic illegal parking & stopping, time for the walk signal, and forms of land use, etc. were found to have a significant influence on bus accident.

Analysis of the Mechanism of Automated Speed Enforcement Systems on Traffic Safety (자동과속단속시스템의 교통안전개선 메커니즘 분석)

  • 강정규;현철승;오세리
    • Journal of Korean Society of Transportation
    • /
    • v.17 no.1
    • /
    • pp.187-196
    • /
    • 1999
  • The increasing interest in the use of Automated Speed Enforcement (ASE) systems in Korea enables to enforce speed violation by National Police Agency. We have analyzed the mechanism of ASE systems on traffic safety throughout Korea. 1 The data collected on a 2km road-section of each 32 ASE stations during one rear period indicate significant safety improvement. The results were (a) a decrease in the total number of accidents of 28%, (b) a decrease in the number of fatalities of 60%. 2. The study also that ASE systems are effective to reduce average speed, speed variance, and short headway. 3. Based on the operational data collected at 15 locations, an aggregate safety prediction model is proposed as a multiple regressions form. The primary operational variables that appear to affect the frequencies of accident are : average speed, speed variance, and the number of vehicles exceeding 30km/h of posted speed limit.

  • PDF

A Study on the Analysis Effect Factors of Illegal Parking Using Data Mining Techniques (데이터마이닝 기법을 활용한 불법주차 영향요인 분석)

  • Lee, Chang-Hee;Kim, Myung-Soo;Seo, So-Min
    • The Journal of The Korea Institute of Intelligent Transport Systems
    • /
    • v.13 no.4
    • /
    • pp.63-72
    • /
    • 2014
  • With the rapid development in the economy and other fields as well, the standard of living in South Korea has been improved, and consequently, the demand of automobiles has quickly increased. It leads to various traffic issues such as traffic congestion, traffic accident, and parking problem. In particular, this illegal parking caused by the increase in the number of automobiles has been considered one of the main reasons to bring about traffic congestion as intensifying any dispute between neighbors in relation to a parking space, which has been also coming to the fore as a social issue. Therefore, this study looked into Daejeon Metropolitan City, the city that is understood to have the highest automobile sharing rate in South Korea but with relatively few cases of illegal parking crackdowns. In order to investigate the theoretical problems of the illegal parking, this study conducted a decision-making tree model-based Exhaustive CHAID analysis to figure out not only what makes drivers park illegally when they try to park vehicles but also those factors that would tempt the drivers into the illegal parking. The study, then, comes up with solutions to the problem. According to the analysis, in terms of the influential factors that encourage the drivers to park at some illegal areas, it was learned that these factors, the distance, a driver's experience of getting caught, the occupation and the use time in order, have an effect on the drivers' deciding to park illegally. After working on the prediction model, four nodes were finally extracted. Given the analysis result, as a solution to the illegal parking, it is necessary to establish public parking lots additionally and first secure the parking space for the vehicles used for living and working, and to activate the campaign for enhancing illegal parking crackdown and encouraging civic consciousness.

A Study on the Rollover Behavior of SUV and Collision Velocity Prediction using PC-Crash Program (PC-Crash를 이용한 SUV의 전복사고 거동 및 충돌속도 예측에 관한 연구)

  • Choi, Yong-Soon;Baek, Se-Ryong;Jung, Jong-Kil;Cho, Jeong-Kwon;Yoon, Jun-Kyu
    • The Journal of the Institute of Internet, Broadcasting and Communication
    • /
    • v.18 no.2
    • /
    • pp.227-235
    • /
    • 2018
  • Along with the recent increase in traffic volume of vehicles, accidents involving rollover of vehicles have been rapidly increased, resulting in an increase casualties. And to prevent this, various technologies such as vehicle crash test equipment and analysis program development have been advanced. In this study, the applied vehicle model is FORD EXPLORER model, and PC-Crash program for vehicle collision analysis is used to predict the rollover accident behavior of SUV and the collision velocity. Compared with the actual rollover behavior of SUV through the FMVSS No 208 regulations, the analysis results showed similar results, the characteristics of the collision velocity and roll angle showed a tendency that the error rate slightly increased after 1000 msec. Then, as a result of considering using the database of NHTSA, it is shown that the rollover accident occur most frequently in the range of the collision velocity of 15~77 km/h and the collision angle of $22{\sim}74^{\circ}$. And it is possible to estimate the vehicle speed and collision time when the vehicle roof is broken by reconstructing the vehicle starting position, the roof failure position and the stop position by applying the actual accident case.