• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.45 no.2
• /
• pp.20-29
• /
• 2022

This study intends to present a traffic node-based and link-based accident prediction models using XGBoost which is very excellent in performance among machine learning models, and to develop those models with sustainability and scalability. Also, we intend to present those models which predict the number of annual traffic accidents based on road types, weather conditions, and traffic information using XGBoost. To this end, data sets were constructed by collecting and preprocessing traffic accident information, road information, weather information, and traffic information. The SHAP method was used to identify the variables affecting the number of traffic accidents. The five main variables of the traffic node-based accident prediction model were snow cover, precipitation, the number of entering lanes and connected links, and slow speed. Otherwise, those of the traffic link-based accident prediction model were snow cover, precipitation, the number of lanes, road length, and slow speed. As the evaluation results of those models, the RMSE values of those models were each 0.2035 and 0.2107. In this study, only data from Sejong City were used to our models, but ours can be applied to all regions where traffic nodes and links are constructed. Therefore, our prediction models can be extended to a wider range.

• Nuclear Engineering and Technology
• /
• v.55 no.2
• /
• pp.493-505
• /
• 2023

Comprehensive condition monitoring of large industry systems such as nuclear power plants (NPPs) is essential for safety and maintenance. In this study, we developed novel system-scale diagnostic technology based on deep-learning and IR thermography that can efficiently and cost-effectively classify system conditions using compact Raspberry Pi and IR sensors. This diagnostic technology can identify the presence of an abnormality or accident in whole system, and when an accident occurs, the type of accident and the location of the abnormality can be identified in real-time. For technology development, the experiment for the thermal image measurement and performance validation of major components at each accident condition of NPPs was conducted using a thermal-hydraulic integral effect test facility with compact infrared sensor modules. These thermal images were used for training of deep-learning model, convolutional neural networks (CNN), which is effective for image processing. As a result, a proposed novel diagnostic was developed that can perform diagnosis of components, whole system and accident classification using thermal images. The optimal model was derived based on the modern CNN model and performed prompt and accurate condition monitoring of component and whole system diagnosis, and accident classification. This diagnostic technology is expected to be applied to comprehensive condition monitoring of nuclear power plants for safety.

• Journal of Radiation Protection and Research
• /
• v.45 no.1
• /
• pp.16-25
• /
• 2020

Background: To investigate radiological effects on biota, it is necessary to assess radiation dose for flora and fauna living in a terrestrial ecosystem. This paper presents a dynamic model to assess radioactivity concentration and radiation dose of terrestrial flora and fauna after a nuclear accident. Materials and Methods: Litter, organic soil, mineral soil, trees, wild crops, herbivores, omnivores, and carnivores are considered the major components of a terrestrial ecosystem. The model considers the physicochemical and biological processes of interception, weathering, decomposition of litter, percolation, root uptake, leaching, radioactive decay, and biological loss of animals. The predictive capability of the model was investigated by comparison of its predictions with field data for biota measured in the Fukushima forest area after the Fukushima nuclear accident. Results and Discussion: The predicted radioactive cesium inventories for trees agreed well with those for evergreens and deciduous trees sampled in the Fukushima area. The predicted temporal radioactivity concentrations for animals were within the range of the measured radioactivity concentrations of deer, wild boars, and black bears. The radiation dose for the animals were, for the whole simulation time, estimated to be much smaller than the lower limit (0.1 mGy·d^-1) of the derived consideration reference level given by the International Commission on Radiological Protection for terrestrial flora and fauna. This suggested that the radiation effect of the accident on the biota in the Fukushima forest would be insignificant. Conclusion: The present dynamic model can be used effectively to investigate the radiological risk to terrestrial ecosystems following a nuclear accident.

• Journal of the Korean Society of Safety
• /
• v.32 no.1
• /
• pp.128-133
• /
• 2017

This study deals with the roundabout traffic accidents by traffic impeding factor. The purpose of this study is to comparatively analyze the characteristics of accidents and to develop the accident models. In pursuing the above, this study used a statistical program SPSS 20.0 to analyze 2,342 accidents occurred within 79 roundabouts in Korea. The main results are as follows. First, 4 accident models which were all statistically significant were developed. Second, the traffic volume and width of right-turn-only lane were analyzed to be common variable in the bus stop related models. The variables such as right-turn-only lane, street light, turning radius of entry lane were selected as specific variables. Especially street light and turning radius of entry lane were evaluated to have negative effects to the accidents. It is, therefore, essential to install the street light and place a sufficient turning radius in order to reduce the roundabout accidents. Finally, the traffic volume and number of entry lane were analyzed to be common variable in the on-street parking related models. Also, the width of right-turn-only lane and bus stop were evaluate to be specific variables in the model with on-street parking. This can be expected to give some implications to making the accident reduction guidelines.

• International Journal of Highway Engineering
• /
• v.15 no.2
• /
• pp.131-138
• /
• 2013

PURPOSES : The intents of the study are to identify the accident factors and to demonstrate the potentials of tobit model as a tool to study the number of accidents on arterial roads segments. METHODS : This paper uses a tobit regression as a methodology to analyze the factors affecting the number of accidents. In pursuing the above goal, this study gives particular attentions to analyzing the data of 2,446 accidents (1,610 in major arterial roads and 836 in minor arterial roads) occurred on arterial roads in 2007 to 2010. RESULTS : First, 3 accident models which were classified by total arterial roads, major arterial roads and minor arterial roads, and were all statistically significant were developed. Second, the exclusive right-turn lane as common variable, and the number of accident, traffic volume, number of lanes, link length, rate of median, number of entrances, number of pedestrian crossings, number of curves, number of bus stops and exclusive left-turn as specific variables of the models were selected. Finally, the paired sample t-test could not be rejected the null hypotheses of three types of models. CONCLUSIONS : Using data from vehicle accidents on arterial roads, the estimation results show that many factors related to roadway geometrics and traffic characteristics significantly affect to the number of accidents.

• International Journal of Highway Engineering
• /
• v.11 no.4
• /
• pp.1-7
• /
• 2009

Traffic accident by aging drivers is expected to be on the rise rapidly as the number of aging drivers is rising along with the aging trend being progressed. In this study, traffic accident features depending on the classification of aging population and non aging one was evaluated. As a result of this evaluation, effect factors influencing over the aging population was found to be expressed differently from that of the non aging one. Odds ratio between the aging population and non aging one was evaluated through logit model and a model with potential accident probability of the aged drivers was developed. Accident risk of the aged drivers under the condition of curved road, cutting section and moistured road was revealed to be higher than that of the non aging population.

• Safety and Health at Work
• /
• v.12 no.4
• /
• pp.505-510
• /
• 2021

Background: Factors related to root causes can cause commonly occurring accidents such as falls, slips, and jammed injuries. An important means of reducing the frequency of occupational accidents in small- to medium-sized enterprises (SMSEs) of South Korea is to perform intensity analysis of the root cause factors for accident prevention in the cause and effect model like decision models, epidemiological models, system models, human factors models, LCU (life change unit) models, and the domino theory. Especially intensity analysis in a robot system and smart technology as Industry 4.0 is very important in order to minimize the occupational accidents and fatal accident because of the complexity of accident factors. Methods: We have developed the modern cause and effect model that includes factors of root cause through statistical testing to minimize commonly occurring accidents and fatal accidents in SMSEs of South Korea and systematically proposed educational policies for accident prevention. Results: As a result, the consciousness factors among factors of root cause such as unconsciousness, disregard, ignorance, recklessness, and misjudgment had strong relationships with occupational accidents in South Korean SMSEs. Conclusion: We conclude that the educational policies necessary for minimizing these consciousness factors include continuous training procedures followed by periodic hands-on experience, along with perceptual and cognitive education related to occupational health and safety.

• Nuclear Engineering and Technology
• /
• v.51 no.3
• /
• pp.723-730
• /
• 2019

Acquiring instrumentation signals generated from nuclear power plants (NPPs) is essential to maintain nuclear reactor integrity or to mitigate an abnormal state under normal operating conditions or severe accident circumstances. However, various safety-critical instrumentation signals from NPPs cannot be accurately measured on account of instrument degradation or failure under severe accident circumstances. Reactor vessel (RV) water level, which is an accident monitoring variable directly related to reactor cooling and prevention of core exposure, was predicted by applying a few signals to deep neural networks (DNNs) during severe accidents in NPPs. Signal data were obtained by simulating the postulated loss-of-coolant accidents at hot- and cold-legs, and steam generator tube rupture using modular accident analysis program code as actual NPP accidents rarely happen. To optimize the DNN model for RV water level prediction, a genetic algorithm was used to select the numbers of hidden layers and nodes. The proposed DNN model had a small root mean square error for RV water level prediction, and performed better than the cascaded fuzzy neural network model of the previous study. Consequently, the DNN model is considered to perform well enough to provide supporting information on the RV water level to operators.

• Nuclear Engineering and Technology
• /
• v.55 no.3
• /
• pp.814-826
• /
• 2023

Sensor faults in nuclear power plant instrumentation have the potential to spread negative effects from wrong signals that can cause an accident misdiagnosis by plant operators. To detect sensor faults and make accurate accident diagnoses, prior studies have developed a supervised learning-based sensor fault detection model and an accident diagnosis model with faulty sensor isolation. Even though the developed neural network models demonstrated satisfactory performance, their diagnosis performance should be reevaluated considering real-time connection. When operating in real-time, the diagnosis model is expected to indiscriminately accept fault data before receiving delayed fault information transferred from the previous fault detection model. The uncertainty of neural networks can also have a significant impact following the sensor fault features. In the present work, a pilot study was conducted to connect two models and observe actual outcomes from a real-time application with an integrated system. While the initial results showed an overall successful diagnosis, some issues were observed. To recover the diagnosis performance degradations, additive logics were applied to minimize the diagnosis failures that were not observed in the previous validations of the separate models. The results of a case study were then analyzed in terms of the real-time diagnosis outputs that plant operators would actually face in an emergency situation.

• International Journal of Highway Engineering
• /
• v.17 no.5
• /
• pp.39-46
• /
• 2015

PURPOSES: The operational characteristics of roundabouts are generally influenced by location as well as traffic volume. The goal of this study is to develop urban and rural roundabout accident models and to discuss safety improvement guidelines based on the model. METHODS : To analyze accidents, count data models are utilized in this study. This study used accident data from 2010 to 2013 for 56 roundabouts collected from the Traffic Accident Analysis System (TASS) of Road Traffic Authority. Poisson and negative binomial regression models were developed for this study using NLOGIT 4.0. RESULTS : The main results are as follows. First, the hypotheses that there are distributional differences in the number of accidents and injuries/fatalities among rural and urban roundabouts were accepted. Second, Poisson and negative binomial regression accident models, which were all statistically significant, were developed. Seven independent variables, which were statistically significant, were adopted. Third, the common variable of models was evaluated to be traffic volume. CONCLUSIONS : This study developed two negative binomial roundabout accident models and suggested some accident reduction strategies. The results are expected to give some implications to the safety improvement of roundabout.