• Journal of Korean Society of Transportation
• /
• v.16 no.2
• /
• pp.177-195
• /
• 1998

지능형교통체계(ITS:Intellegent Transport System)의 구현을 위한 가장 중요한 요소중의 하나는 교통정보의 생성이다. 교통정보의 생성은 루프 검지기, 폐쇄회로(CCTV), probe 차량, 경찰, 통신원 등을 수집된 제보자료들을 분석 및 가공함으로써 이루어진다. 그러나 이들 수집원은 주어진 시간에 있어 모든 네트웍을 통해서 자료가 완전히 수집되어지는 것은 아니다. 즉, 특정 지역에 수집원이 몰려 있는 경우가 있는 반면, 전혀 수집되어지지 않는 지역이 발생할 수도 있다. 이러한 공간적인 불균형적 특성은 동시에 발생한 다량의 자료를 처리하는 기술과 자료가 수집되지 않은 지역에 대한 처리기술을 요하게 된다. 본 논문은 전술한 바와 같은 사항에 대하여 ITS의 진행 단계별로 드러날 수 있는 문제점을 검토하고, 자료통합에 대한 일반적인 개념을 우선 설명한다. 다음에 특정시각에 주어진 자료의 통합을 위해 퍼지선형회귀모형(fuzzy linear regression model)과 데이터 퓨전(data fusion)기법의 내용을 소개하고, 신뢰성있는 단일 교통정보생성을 위한 테이터 퓨전 알고리즘을 제시한다. 또한 제시된 알고리즘을 토대로 가상의 자료를 이용하여 적용가능 봉? 타진해 보았다. 제시되어진 알고리즘은 향후 교통정보 수집환경이 어느 정도 형성된다고 볼 때, 예측치와 실측자료간의 자료검증을 통하여 신뢰도를 가질 경우 보다 광범위하게 사용되어질 수 있을 것으로 판단된다.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.16 no.6
• /
• pp.67-78
• /
• 2017

Traffic congestion cost is increasing annually. Specifically congestion caused by the CDB traffic contains more than a half of the total congestion cost. Recent advancement in the field of Big Data, AI paved the way to industry revolution 4.0. And, these new technologies creates tremendous changes in the traffic information dissemination. Eventually, accurate and timely traffic information will give a positive impact on decreasing traffic congestion cost. This study, therefore, focused on developing both recurrent and non-recurrent congestion prediction models on urban roads by adopting Recurrent Neural Network(RNN), a tribe in machine learning. Two hidden layers with scaled conjugate gradient backpropagation algorithm were selected, and tested. Result of the analysis driven the authors to 25 meaningful links out of 33 total links that have appropriate mean square errors. Authors concluded that RNN model is a feasible model to predict congestion.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.3-3
• /
• 2015

임진강 유역면적은 $8,138.9km^2$이나 이 중 62.9%인 $5,108km^2$가 군사분계선 이북에 놓여있어 유역특성에 따른 수문량을 분석하는 데 어려움이 있다. 1996년, 1998년, 1999년 및 2011년 임진강 유역의 이상 강우로 인해 약 1조 원의 재산피해와 136명의 인명피해가 발생하였다. 이처럼 국지성 호우의 발생 여부 및 강우의 지역적 편차 등 수문 정보를 예측하지 못하여 상황 대처가 어려운 실정이다. 따라서 미계측 유역이 많은 임진강 유역의 홍수피해 최소화를 위해 예측 강우와 같은 수문 정보의 필요성이 증대되고 있다. 본 연구에서는 임진강 유역 중 미계측 유역이 97%에 달하는 군남홍수조절지 유역에 예측 강우 자료의 적용을 위해, 임진강 유역의 한탄강 유역($2,436.4km^2$)에 대하여 예측 강우자료의 적용성을 평가하였다. 예측 강우 자료는 기상청의 Local Data Assimilation and Prediction System(LDAPS) 자료를 사용하여 선행 시간에 따른 예측 정확도로부터 적용성을 평가하였다.

• Proceedings of the KOR-KST Conference
• /
• 1999.10a
• /
• pp.102-107
• /
• 1999

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.6
• /
• pp.13-20
• /
• 2022

Road conditions and weather conditions are very important factors in the case of traffic accident fatalities in fog and ice sections that occur on roads in winter. In this paper, a simulation was performed to estimate the traffic accident risk rate assuming traffic accident prediction data. In addition, in this paper, in order to reduce traffic accidents and prevent traffic accidents, factor analysis and traffic accident fatality rates were predicted using the WEKA data mining technique and TENSOR FLOW open source data on traffic accident fatalities provided by the Korea Transportation Corporation.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.237-240
• /
• 2023

본 논문은 철도 건널목(교차로)에서 발생하는 차량, 보행자 및 야생 동물 사고 등의 상황에서 발생하는 위험 요소를 설정하고 철도 건널목(교차로)의 운행상황을 확인할 수 있도록 모형 철도 주변에 유형별 센서들을 설치하고 데이터를 인지하여 시스템에 저장하고, 유효한 데이터 분석을 통해 Orange3 머신러닝 기법을 적용한다. 철도 건널목에 관련된 이미지 중 위험인자로서 차량, 보행자 및 야생동물등의 객체를 감지하고 데이터를 수집하여 활용한다. 또한 이러한 데이터들은 이용자 상황에 맞는 철도 데이터 운영 시스템으로 적용할 수 있도록 위험 예측 시스템을 제안한다.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2019.11a
• /
• pp.24-25
• /
• 2019

최근 빅데이터 기술 개발로 방대한 데이터의 유의미한 분석 및 예측이 용이해졌다. 선박교통관제센터에서는 각종 센서와 다양한 정보를 기반으로 VHF 교신을 통해 선박교통관제를 수행한다. 관제사가 활용하는 레이더, AIS, Port-MIS. 센서 등의 데이터들이 디지털로 저장되고 있으며, 관제사의 VHF 교신내용은 디지털파일로 저장되어 선박교통관제센터의 서버 2개월간 보관된다. 본 논문에서는 관제 결과로 저장되고 있는 관제 로그 데이터를 활용하여 빅데이터를 구성하고 이를 기반으로 유의미한 정보를 생성할 수 있는 방안을 연구하였다.

• Journal of Korean Society of Transportation
• /
• v.35 no.4
• /
• pp.348-359
• /
• 2017

Recently, various public transportation activation policies are being implemented in order to mitigate traffic congestion in metropolitan areas. Especially in the metropolitan area, the bus information system has been introduced to provide information on the current location of the bus and the estimated arrival time. However, it is difficult to predict the travel time due to repetitive traffic congestion in buses passing through complex urban areas due to repetitive traffic congestion and bus bunching. The previous bus travel time study has difficulties in providing information on route travel time of bus users and information on long-term travel time due to short-term travel time prediction based on the data-driven method. In this study, the path based long-term bus travel time prediction methodology is studied. For this purpose, the training data is composed of 2015 bus travel information and the 2016 data are composed of verification data. We analyze bus travel information and factors affecting bus travel time were classified into departure time, day of week, and weather factors. These factors were used into clusters with similar patterns using self organizing map. Based on the derived clusters, the reference table for bus travel time by day and departure time for sunny and rainy days were constructed. The accuracy of bus travel time derived from this study was verified using the verification data. It is expected that the prediction algorithm of this paper could overcome the limitation of the existing intuitive and empirical approach, and it is possible to improve bus user satisfaction and to establish flexible public transportation policy by improving prediction accuracy.

• Journal of Korean Society of Transportation
• /
• v.20 no.3
• /
• pp.17-30
• /
• 2002

Incident duration prediction is one of the most important steps of the overall incident management process. An accurate and reliable estimate of the incident duration can be the main difference between an effective incident management operation and an unacceptable one since, without the knowledge of such time durations, traffic impact can not be estimated or calculated. This research presents several multiple linear regression models for incident duration prediction using data consisting of 384 incident cases. The main source of various incident cases was the Traffic Incident Reports filled out by the Motorist Assistant Units of the Korea Highway Corporation. The models were proposed separately according to the time of day(daytime vs. nighttime) and the fatality/injury incurred (fatality/injury vs. property damage only). Two models using an integrated dataset, one with an intercept and the other without it, were also calibrated and proposed for the generality of model application. Some findings are as follows ; ?Variables such as vehicle turnover, load spills, the number of heavy vehicles involved and the number of blocked lanes were found to significantly affect incident duration times. ?Models, however, tend to overestimate the duration times when a dummy variable, load spill, is used. It was simply because several of load spill incidents had excessively long clearance times. The precision was improved when load spills were further categorized into "small spills" and "large spills" based on the size of vehicles involved. ?Variables such as the number of vehicles involved and the number of blocked lanes found not significant when a regression model was calibrated with an intercept. whereas excluding the intercept from the model structure signifies those variables in a statistical sense.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.9
• /
• pp.89-96
• /
• 2016

According to the promoting policy for bicycle riding, the bicycle road infrastructure in Korea has been widely established. As the number of bicycle rider increases, bicycle traffic accidents also increase year after year. In this paper, we analyze bicycle traffic accident data from 2007 to 2014 which is provided by Road Traffic Authority and present statistical results of bicycle traffic accidents. And also regression analysis is applied to predict the number of daily traffic accidents in Seoul using ASOS(Automated Synoptic Observing System) climate data observed in the Seoul sector which are provided by Korea Meteorological Administration. In addition, decision tree analysis techniques are used to forecast the level of traffic accidents severity. In the analytic results of this research, we expect that it will be helpful to establish the collective policy of bicycle accident data and protective strategy in order to reduce the number of bicycle accidents.