Accident Prediction Models Research Articles

Development of Model for Road Crashes and Identification of Accident Spots

The aim of this study is to assess the safety of multi-lane rural highway in India. This paper shows the application of a generalized linear modeling technique for the analysis of road accidents on the Indian National Highway. Speed, traffic flow and road characteristics data on four-lane dived rural highway in Dahod are analyzed. The study proposes a novel approach to include average daily traffic (ADT) and average spot speed (AS) in the accident prediction model for a rural highway. The model has been developed for accidents per km as a dependent variable and significant variables such as Junction density, village settlement nearby, ADT, AS as independent variables. The findings from the model offer a better estimate of accidents for a multilane divided rural highway. Statistical Models cannot fully reflect the characteristics of each section due to the heterogeneous nature of road accidents, so the association rule mining technique has been used to identify accident spots as it can deal with the heterogeneous nature of accidents. Accident spots have been assessed by correlating various attributes to the severity of the accident (fatal, non-fatal). This research will help to improve road safety on rural highways.

Traffic Accident Prediction Based on LSTM-GBRT Model

Road traffic accidents are a concrete manifestation of road traffic safety levels. The current traffic accident prediction has a problem of low accuracy. In order to provide traffic management departments with more accurate forecast data, it can be applied in the traffic management system to help make scientific decisions. This paper establishes a traffic accident prediction model based on LSTM-GBRT (long short-term memory, gradient boosted regression trees) and predicts traffic accident safety level indicators by training traffic accident-related data. Compared with various regression models and neural network models, the experimental results show that the LSTM-GBRT model has a good fitting effect and robustness. The LSTM-GBRT model can accurately predict the safety level of traffic accidents, so that the traffic management department can better grasp the situation of traffic safety levels.

A New Accident Prediction Model for Highway-Rail Grade Crossings Using the USDOT Formula Variables

A New Accident Prediction Model for Highway-Rail Grade Crossings Using the USDOT Formula Variables

An approach driven critical review on the use of accident prediction models for sustainable transport system

Over the past decade, the use of statistical analysis for the prediction of road accidents in sustainable road infrastructure and road safety has become popular. However, due to lack of adequate knowledge, for practical applications, a gap has been found between state-of-art and state-of-practice. The primary objective of this paper is to introduce a new innovative statistical technique called Accident Prediction Model (APM) along with its types, and possible challenges with their solutions while dealing with these models. In nutshell, why and how APM can contribute to improving road safety has been discussed in depth. The study recommends the use of APM for precise, accurate and effective prediction of road accidents for sustainable road infrastructure.

A statistical model for prediction of road accidents in the State of Punjab

Accident Prediction Models (APMs) are extremely powerful tools that are getting used for road safety management as they have the potential to determine both degree and frequency of crashes. This paper focuses on one of the first attempts to develop APM in the Indian state of Punjab. The main objective of the study was to provide a tool that simplifies a proactive methodology to community planning which addresses the safety of roads before a problem arises. The models are developed using accident and socio-economic data which were collected from various departments of the Punjab state. The model was validated to check accuracy. The results showed that the proposed model is highly accurate in predicting road accidents in the state and it is hoped that the output of this work can be used for facilitating improved decisions by road safety teams, engineers, and planners for road users.

A Study on the Effect Evaluation of Traffic Enforcement and Street Safety Activity Using a Traffic Accident Prediction Model

A Study on the Effect Evaluation of Traffic Enforcement and Street Safety Activity Using a Traffic Accident Prediction Model

Accident data analysis and on-field inspections: do they lead to similar conclusions?

Network Safety Ranking and Road Assessment Programme (RAP) are two procedures used in road infrastructure safety management to rank the road sections of a network according to estimated road safety performance. The former uses indicators based on road accidents and their consequences while the latter method is based primarily on inspections of the road infrastructure.The aim of this study was to verify whether there is a relationship between the indicators used within these methods, in order to evaluate if and under which conditions RAP scores contribute to the prediction of accident numbers. The adopted approach is based on the calibration and analysis of a wide array of models in order to reproduce the observed events – injury road accidents, fatalities and injuries - with reference to different aggregations of the original data and in the two reference situations: with and without EuroRAP (European Road Assessment Programme) indicators.. From the obtained results, the introduction of the EuroRAP indicators ends up making the models that explain the frequency of accidents more convincing from a statistical point of view. These indicators, which summarise the road safety performance of a road section, could prove useful in replacing the covariates (e.g. presence of intersections, geometrical and functional features of roads) that are used in the calibration of accident prediction models. However, it should be stressed that not all types of accident are addressed by the EuroRAP approach, and this may represent a limitation for those roadways where such kind of accidents are frequently observed, like rear-end collisions on motorways. Probably, by including these accident types, the explanatory power of the EuroRAP variables would increase, at least on motorways.

Comparative Assessment Between Statistical and Soft Computing Methods for Accident Severity Classification

Globally traffic scenario has experienced a vast change over the years as volumes of traffic on urban roads have increased by leaps and bounds. This increase has also marked an increase in the number of road accidents, both fatal and non-fatal. Accident prediction has a significant role in the improvement in traffic safety, and urban traffic management. Different statistical methods and various soft computing techniques are being used for developing accident prediction models. In this study, ensemble ranking method, comprising three filter-based ranking methods such as Information Gain, Gain Ratio, and Symmetric Uncertainty, has been employed to select significant features for prediction of accident severity. Through analysis, eight significant features have been selected for accident severity prediction. These are Annual Average Daily Traffic, Road Width, Percentage of Left-Turn Vehicles, Driver’s Age, Percentage of Right-Turn Vehicles, Types of Vehicles Involved, Spot Speed, and Cause of Accidents. Multiple logistic regression (MLGR) and artificial neural network (ANN) have been applied for accident severity prediction in urban roads. This analysis concludes that ANN model gives more accuracy and minimum errors compared with the MLGR method at the time of severity classification of a vehicular accident.

Accident Study Based on Smeed’s Model: A Case Study of Yelhanka

Accident analyses are accomplished in order to identify the source of an accident. So has to forbid further accidents of homogenous kind. From the last twenty years the inhabitants of different nations in the globe is wide growing. Due to population explosion, automobiles and economic tasks roadway traffic disasters and their resulting mortality may be considered as a bigger communal and economic trouble. The salient purpose of this study to be developing road accident prediction model, identification of black spots and a micro level analysis of road accident is performed. On basis of this analysis effect of accident is identified. After analysis road traffic accident prediction models is developed based on different parameter.

TRAFFIC COLLISION TIME SERIES ANALYSIS (A CASE STUDY OF KARAJ–QAZVIN FREEWAY)

Abstract. The application of Auto-correlation Function (ACF) and Partial Auto-correlation Function (PACF) in recent years has been improved in analyzing big traffic data, modelling traffic collisions and decreasing processing time in finding collision patterns. Accident prediction models for short and long time can help in designing and programming traffic plans and decreasing road accidents. Based on the above details, in this paper, the Karaj-Qazvin highway accident data (1097 samples) and its patterns from 2009 to 2013 have been analyzed using time series methods.In the first step, using auto correlation function (ACF) and partial auto correlation function (PACF), the rank of time series model supposed to be autoregressive (AR) model and in the second stage, its coefficients were found. In order to extract the accident data, ArcGIS software was run. Furthermore, MATLAB software was used to find the model rank and its coefficients. In addition, Stata SE software was used for statistical analysis. The simulation results showed that on the weekly scale, based on the trend and periodic pattern of data, the model type and rank, ACF and PACF values, an accurate weekly hybrid model (time series and PACF) of an accident can be created. Based on simulation results, the investigated model predicts the number of accident using two prior week data with the Root Mean Square Error (RMSE) equal to three.

Safety performance functions in Dedicated Bus Lane of BRT on Caracas Avenue Corridor at Bogotá city

Introduction- Road safety is a global concern due to the fact that traffic accidents represent serious temporary and / or permanent damage to the health of those involved. On the other hand, the Bus Rapid Transit (BRT) systems carries a large volume of passengers and during their operation; they are involved in this problem. Objective- Accident prediction model implemented in the Highway Safety Manual 2010 or HSM is an alternative to evaluate the strategies that allow to reduce accidents in this type of systems. However, there is not specified safety performance functions (SPFs) developed for BRT systems. In the present work, the accident model of HSM is adapted by calibration of general SPFs expressions of the manual and also, SPFs were developed for BRTs installed on the central-line of main roads and use an exclusive lane of all other transport systems (both public or private) and mobility (e.g. bike paths). Method / Results - Crashes reports and traffic volumes data supplied by the Department of Transportation of Bogotá were used. The model was calibrated using the safety performance functions (SPFs) of the HSM and a specific developed functions for the BRT conditions. These SPFs were developed using a negative binomial model in roadway segments and intersections. Conclusions- Through the validation, it was found that the functions developed have a better fit than the established SPF of the HSM. The developed SPFs can be used as a tool to define safety performance guidelines of Bogotá's BRT corridors in the coming years.

운전자 근로여건에 따른 버스 교통사고 예측모형 및 안전등급 개발 연구

교통사고 발생과 관련하여 도로환경요인이나 차량요인에 비해서 인적요인이 가장 큰 비중을 차지하고 있다. 따라서 교통사고 예방을 위해서는 운전자에 대한 안전관리가 무엇보다 중요하다. 특히 노선버스와 같은 대중교통 수단은 많은 승객이 이용하고, 하루 종일 운행하기 때문에 운전자의 안전 운행 책임과 교통사고 발생 위험성을 내포하고 있다. 이 가운데 운전자의 근로여건은 지역별로 큰 편차를 보이고 있어, 버스 안전도 분명히 이에 영향을 받을 수 있을 것으로 판단했다. 이를 검증하기 위해서 음이항 회귀분석과 포아송 회귀분석을 이용한 안전성능함수를 개발하고 근로여건 변화에 따른 교통사고 발생건수를 예측하였다. 전국의 노선버스를 대상으로 버스공제조합의 지난 3년간 버스 교통사고 자료와 2017년의 버스회사별 경영 및 노무자료 등을 활용하였다. 분석 결과, 우도비 검정 통계량이 유의하게 나타나 과분산으로 판정되었고, 음이항 회귀분석 결과 총이탈도가 1에 가깝고, AIC 및 BIC값도 낮게 나타나 음이항 회귀모형이 적합하였다. 최종 안전성능함수는 보유대수를 노출변수로 선정하였고, 대당운전자수와 월급여가 유의한 변수로 선정되었다. 설명력은 부족했지만 근무형태와 월근로시간도 대당운전자수와 높은 상관관계를 가지고 있으므로 교통사고에 영향을 미칠 것으로 판단된다. 또한 교통사고 예측모형으로 구축된 안전성능함수를 이용하여 버스회사별 안전등급을 4단계(A-D등급)로 설정하였다. 안전등급이 D로 나빠질수록 평균 사고건수와 표준편차가 증가하였다. 또한 준공영제와 민영제 간 안전등급의 차이가 발생하였다. 준공영제에서는 A-B등급이, 민영제에서는 D등급 버스회사의 비중이 높았다. 본 연구에서의 버스회사 간 안전성의 차이가 해당 버스회사의 서비스평가 점수와 운행기록분석시스템 내 ‘100km당 위험행동건수’ 차이와 유사한 의미를 가지는 것으로 나타났다. 근로기준법이 개정되어 2019년 7월 1일부터 업체 규모별로 노선버스 운전자의 근로시간 단축이 시행되고 있기 때문에, 근로여건 개선에 따른 교통사고 감소효과를 미리 예측해볼 수 있는 좋은 기회라고 판단된다.

A Mechanistic Model for Predicting Accident Potential of Vehicles Transiting in Nigerian Roads

An accident prediction model was developed for determining the accident potential of a vehicle while on transit. The model identifies the various factors responsible for vehicle crashes. With the help of accident data obtained from the database of the Nigerian Federal Road Safety Corps (FRSC), the percentage contribution of each factor is calculated. These accident-cause factors were further grouped into three distinct classes: Human factors (HF), Mechanical factors (MF) and Environmental factors (EF). Analysis of the accident data showed that HF is the chief cause of most road accidents recorded, followed by MF and EF with probabilities of 0.846, 0.138 and 0.016 respectively. Also, driver age, travel distance and maintenance frequency of the vehicle were considered in the development of the model. The model gives an output ranging from 0-1. Values close to 0 mean low accident probability while values close to 1 signify high accident probability. Application and adherence to this model will significantly reduce the frequency of road accidents. Finally, transport companies and fleet operators are therefore encouraged to embrace and use this innovation for safer operations.

Traffic Accident Severity Prediction Using Naive Bayes Algorithm - A Case Study of Semarang Toll Road

A traffic accident was one of the leading cause of death in Indonesia. Toll Road is one of the places where traffic accidents occur. In 2007-2017 there were 501 accidents at Semarang Toll Road. Accident in Semarang Toll Road has a variety of severity. The most severe case is death. A traffic accident can lead to death. One of the ways to decrease the number of the accident was decreased the severity of the accident. This achieved by making a prediction model. The prediction model can predict the severity of the accident based on the attribute affecting the severity of the accident. In this research, Days, Type of Road, Weather, Condition of Road, Time of the accident, Sex of Driver, and Type of Vehicle were chosen as attributes to make prediction model of accident severity. Naive Bayes algorithm was used to make the model which can predict accident severity. The result was an accident prediction model with an accuracy of 39.49% to predict accident severity and the probability of an accident.

Providing an Appropriate Prediction Model for Traffic Accidents: A Case Study on Accidents in Golestan, Mazandaran, Guilan, and Ardebil Provinces

Providing an Appropriate Prediction Model for Traffic Accidents: A Case Study on Accidents in Golestan, Mazandaran, Guilan, and Ardebil Provinces

Mechanistic Model for Predicting Accident Potential of Vehicle Transiting in Nigeria Road

An accident prediction model was developed for determining the accident potential of a vehicle while on transit. The model identifies the various factors responsible for vehicle crashes with the help of accident data obtained from the database of the Nigerian Federal Road Safety Corps (FRSC), the percentage contribution of each factor is calculated. These accident-cause factors were further grouped into three distinct classes: Human factors (HF), Mechanical factors (MF) and Environmental factors (EF). Analysis of the accident data showed that HF is the chief cause of most road accidents recorded, followed by MF and EF with probabilities of 0.846, 0.138 and 0.016 respectively. Also, driver age, travel distance and maintenance frequency of the vehicle were considered in the development of the model. The model gives an output ranging from 0-1. Values close to 0 mean low accident probability while values close to 1 signify high accident probability. Application and adherence to this model will significantly reduce the frequency of road accidents. Finally, transport companies and fleet operators are therefore encouraged to embrace and use this innovation for safer operations.

COMPARISON OF ROAD TRAFFIC ACCIDENT PREDICTION MODELS FOR TWO-LANE HIGHWAY INTEGRATING TRAFFIC AND PAVEMENT CONDITION PARAMETERS

In Nigeria, literature on the integration of traffic of pavement condition and traffic characteristics in predicting road traffic accident frequency on 2-lane highways are scanty, hence this article to fill the gap. A comparison of road traffic accident frequency prediction models on IIesha-Akure-Owo road based on the data observed between 2012 and 2014 is presented. Negative Binomial (NB), Ordered Logistic (OL) and Zero Inflated Negative Binomial (ZINB) models were used to model the frequency of road traffic accident occurrence using road traffic accident data from the Federal Road Safety Commission (FRSC) and pavement conditions parameters from pavement evaluation unit of the Federal Ministry of Works, Kaduna. The explanatory variables were: annual average daily traffic (aadt), shoulder factor (sf), rut depth (rd), pavement condition index (pci), and international roughness index (iri). The explanatory variables that were statistically significant for the three models are aadt, sf and iri with the estimated coefficients having the expected signs. The number of road traffic accident on the road increases with the traffic volume and the international roughness index while it decreases with shoulder factor. The systematic variation explained by the models amounts to 87.7, 78.1 and 74.4% for NB, ZINB and OL respectively. The research findings suggest the accident prediction models that should be integrated into pavement rehabilitation.
 
 Keywords: 

Accident Prediction Model Using Environmental Sensors for Industrial Internet of Things

Accident Prediction Model Using Environmental Sensors for Industrial Internet of Things

Safety assessment of Czech motorways and national roads

PurposeCzech motorways and national roads form the primary road network, which is critical in terms of safety. To be able to rationally manage network safety in both planning and operation stages, quality network-wide data and tools are needed. While such tools already exist in some countries, their transferability is limited. Authors therefore collected data and used it to develop tools, which allowed conducting state-of-the-art road safety impact assessment and network safety ranking in the Czech conditions. In addition to primary road network, focus was widened to include also secondary roads, in order to enable assessment of impacts on adjacent road network.MethodsAccident, road and traffic data was collected, using not only existing databases, but also including own collection of traffic volumes on motorway interchanges. Data was used to develop the tools, based on accident prediction models and accident modification factors.ResultsThe final accident prediction models and accident modification factors enabled conducting road safety impact assessment, for which simple on-line tool was also developed. For network safety ranking, accident prediction models were applied according to the Empirical Bayes method, in order to determine potential for safety improvement of the studied road network elements, with the final priority list visualized in an on-line map. Both outputs are shortly presented in the paper.ConclusionsData and sample size limitations lead to some compromises in modelling, such as using fixed proportions of observed accident severities or omitted variables. Nevertheless, the study established the practical framework for both road safety impact assessment and network safety ranking. It may serve as an example for other member countries, which also lack their local tools. Follow-up studies may focus on future model updating and improvements, as well as development of local accident modification factors.

Development of an accident prediction model for Italian freeways

The roadway safety management process plays an important role in the national efforts for improving road safety along the Italian freeway network. In 2016, 8.3% of the overall Italian road deaths and 6.3% of the overall road injuries occurred along the 6700 km-long freeway network. Accident Prediction Models (APMs) represent one of the best tools to perform a road safety quantitative assessment. With the aim of providing the Italian freeway agencies with a tool that allows to deal with potential safety issues, this paper defines two APMs for single- and multiple-vehicle fatal-and-injury crashes to be applied on Italian rural freeway segments, based on jurisdictional specific Safety Performance Functions (SPFs) developed in the PRACT project.The proposed procedure is based on the Highway Safety Manual (HSM) approach, and it introduces a new methodology to transfer the HSM to European motorways.In order to improve the prediction accuracy, the proposed APMs consist in a jurisdictional specific base SPF, developed for the base data set as a function of Annual Average Daily Traffic (AADT) and segment length, combined with Crash Modification Factors (CMFs), in order to account for differences between each site and the base conditions. The full models are then calibrated based on the total number of accidents observed in the wide data set. For both full models (one for single-vehicle and one for multiple-vehicle crashes), the goodness of fit is evaluated in terms of chi square test, root mean square error, observed Vs predicted diagram and predicted Vs residual diagram.The results show a good aptitude of both models to describe the analysis data set. The proposed models represent a solid and reliable tool for practitioners to perform accident predictions along the Italian freeway network.