A data mining approach to investigate the factors influencing the crash severity of motorcycle pillion passengers

Kashani, Ali Tavakoli; Rabieyan, Rahim; Besharati, Mohammad Mehdi

doi:10.1016/j.jsr.2014.09.004

Cited by 73 publications

(19 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, several studies employed machine learning techniques to analyze and predict the crash severity of motorcycle crashes [26,29,31,67]. Anvari et al .…”

Section: Literature Reviewmentioning

confidence: 99%

“…Tavakoli Kashani et al . [26] utilized a classification and regression tree model to study factors affecting the crash severity of motorcycle pillion passengers. The study revealed that the predictive accuracy of their model showed considerable improvement compared to previous studies.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Furthermore, these two model types have diverse properties: statistical models provide good theoretical interpretability with explicit mathematics construction; whereas machine learning techniques employ a “black box” tactic to forecast crash severity and often lack a reasonable explanation of the model. Compared to statistical models, machine learning methods are more accommodating with no or little presumptions for explanatory variables [25,26]. In addition, these machine learning based algorithms are more accomplished in dealing with outliers, noisy and missing data [27].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A comparative study on machine learning based algorithms for prediction of motorcycle crash severity

2019

View full text Add to dashboard Cite

Motorcycle crash severity is under-researched in Ghana. Thus, the probable risk factors and association between these factors and motorcycle crash severity outcomes is not known. Traditional statistical models have intrinsic assumptions and pre-defined correlations that, if flouted, can generate inaccurate results. In this study, machine learning based algorithms were employed to predict and classify motorcycle crash severity. Machine learning based techniques are non-parametric models without the presumption of relationships between endogenous and exogenous variables. The main aim of this research is to evaluate and compare different approaches to modeling motorcycle crash severity as well as investigating the effect of risk factors on the injury outcomes of motorcycle crashes. Motorcycle crash dataset between 2011 and 2015 was extracted from the National Road Traffic Crash Database at the Building and Road Research Institute (BRRI) in Ghana. The dataset was classified into four injury severity categories: fatal, hospitalized, injured, and damage-only. Three machine learning based models were developed: J48 Decision Tree Classifier, Random Forest (RF) and Instance-Based learning with parameter k (IBk) were employed to model the severity of injury in a motorcycle crash. These machine learning algorithms were validated using 10-fold cross-validation technique. The three machine learning based algorithms were compared with one another and the statistical model: multinomial logit model (MNLM). Also, the relative importance analysis of the attribute was conducted to determine the impact of these attributes on injury severity outcomes. The results of the study reveal that the predictions of machine learning algorithms are superior to the MNLM in accuracy and effectiveness, and the RF-based algorithms show the overall best agreement with the experimental data out of the three machine learning algorithms, for its global optimization and extrapolation ability. Location type, time of the crash, settlement type, collision partner, collision type, road separation, road surface type, the day of the week, and road shoulder condition were found as the critical determinants of motorcycle crash injury severity.

show abstract

“…Recently, several studies employed machine learning techniques to analyze and predict the crash severity of motorcycle crashes [26,29,31,67]. Anvari et al .…”

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A comparative study on machine learning based algorithms for prediction of motorcycle crash severity

2019

View full text Add to dashboard Cite

show abstract

“…Hence, this methodology can be very useful for obtaining reliable conclusions for the decision-making process with regard to safety issues. Bayesian networks have also been applied in many other scientific fields, including civil engineering [ 19 ], geological engineering [ 20 ], ecology [ 21 ], medicine [ 22 ], road traffic safety [ 23 , 24 , 25 ], environmental assessment impact [ 26 , 27 ], business risk and product life-cycle analysis [ 28 ], workplace tasks [ 29 ], workplace risk areas [ 30 ], and the interrelation between hygienic workplace conditions and occupational accidents [ 31 ] or construction and mining accidents [ 32 , 33 ].…”

Section: Introductionmentioning

confidence: 99%

Analysis of Occupational Accidents in Underground and Surface Mining in Spain Using Data-Mining Techniques

Pera

Bascompta

Rossell

et al. 2018

IJERPH

View full text Add to dashboard Cite

An analysis of occupational accidents in the mining sector was conducted using the data from the Spanish Ministry of Employment and Social Safety between 2005 and 2015, and data-mining techniques were applied. Data was processed with the software Weka. Two scenarios were chosen from the accidents database: surface and underground mining. The most important variables involved in occupational accidents and their association rules were determined. These rules are composed of several predictor variables that cause accidents, defining its characteristics and context. This study exposes the 20 most important association rules in the sector—either surface or underground mining—based on the statistical confidence levels of each rule as obtained by Weka. The outcomes display the most typical immediate causes, along with the percentage of accidents with a basis in each association rule. The most important immediate cause is body movement with physical effort or overexertion, and the type of accident is physical effort or overexertion. On the other hand, the second most important immediate cause and type of accident are different between the two scenarios. Data-mining techniques were chosen as a useful tool to find out the root cause of the accidents.

show abstract

“…For the method assessment, the accuracy associated with the confusion matrix was introduced to measure the percentage of cases correctly classified by the classifier. In the investigation from existing studies, the prediction accuracy was acceptable when it exceeded 50% [28,29,44]. Thus, the decision tree model proposed in the study achieved a relatively satisfying performance in the cause-consequence analysis.…”

Section: Discussionmentioning

confidence: 83%

Characteristics, Cause, and Severity Analysis for Hazmat Transportation Risk Management

Zhou

Guo

Cui

et al. 2020

IJERPH

View full text Add to dashboard Cite

The accidents caused by hazardous material during road transportation may result in catastrophic losses of lives and economics, as well as damages to the environment. Regarding the deficiencies in the information systems of hazmat transportation accidents, this study conducts a survey of 371 accidents with consequence Levels II to V involving road transportation in China from 2004–2018. The study proposes a comprehensive analysis framework for understanding the overall status associated with key factors of hazmat transportation in terms of characteristics, cause, and severity. By incorporating the adaptive data analysis techniques and tackling uncertainty, the preventative measures can be carried out for supporting safety management in hazmat transportation. Thus, this study firstly analyzed spatial–temporal trends to understand the major characteristics of hazmat transportation accidents. Secondly, it presented a quantitative description of the relation among the hazmat properties, accident characteristics, and the consequences of the accidents using the decision tree approach. Thirdly, an enhanced F-N curve-based analysis method that can describe the relationship between cumulative probability F and number of deaths N, was proposed under the power-law distribution and applied to several practical data sets for severity analysis. It can evaluate accident severity of hazmat material by road transportation while taking into account uncertainty in terms of data sources. Through the introduction of the as low as reasonably practicable (ALARP) principle for determining acceptable and tolerable levels, it is indicated that the F-N curves are above the tolerable line for most hazmat accident scenarios. The findings can provide an empirically supported theoretical basis for the decision-makers to take action to reduce accident frequencies and risks for effective hazmat transportation management.

show abstract

A data mining approach to investigate the factors influencing the crash severity of motorcycle pillion passengers

Cited by 73 publications

References 21 publications

A comparative study on machine learning based algorithms for prediction of motorcycle crash severity

A comparative study on machine learning based algorithms for prediction of motorcycle crash severity

Analysis of Occupational Accidents in Underground and Surface Mining in Spain Using Data-Mining Techniques

Characteristics, Cause, and Severity Analysis for Hazmat Transportation Risk Management

Contact Info

Product

Resources

About