Using Deep Learning in Severity Analysis of At-Fault Motorcycle Rider Crashes

Abstract

Motorcyclists are vulnerable highway users. Unlike passenger vehicle occupants, motorcycle riders do not have either protective structural surrounding or the advanced restraints that are mandatory safety features in cars and light trucks. Per vehicle mile traveled, motorcyclist fatalities occurred 27 times more frequently than passenger car occupant fatalities in traffic crashes. In addition, there were 4,976 motorcycle crash-related fatalities in the U.S. in 2014—more than twice the number of motorcycle rider fatalities that occurred in 1997. It shows that, in addition to current efforts, research needs to be conducted with additional resources and in newer directions. This paper investigated five years (2010–2014) of Louisiana at-fault motorcycle rider-involved crashes by using deep learning, which is a competent tool for mapping a high-multidimensional input into a smaller multidimensional output. The current study contributes to the existing injury severity modeling literature by developing a deep learning framework, named as DeepScooter, to predict motorcycle-involved crash severities. The final deep learning model can predict severity types with 100% accuracy with training data, and with 94% accuracy with test data, which is not attainable by using a statistical method or machine learning algorithm. The intensity of severities was found to be more likely associated with rider ejection, two-way roadways with no physical separation, curved aligned roadways, and weekends. It is anticipated that the DeepScooter framework and the findings will provide significant contributions to the area of motorcycle safety.

Application of various machine learning architectures for crash prediction, considering different depths and processing layers

“…Machine learning algorithm of DeepScooter was used to predict motorcycle at fault crashes. 3 The algorithm was able to reach an accuracy of 94%. Some of the contributory factors were found to be roadway geometric characteristics, and day of a week.…”

Application of deep learning techniques in predicting motorcycle crash severity

“…Various modeling approaches including parametric and nonparametric models have been employed throughout the literature to analyze crash severity by assessing injury-severity levels associated with the aforementioned factors. 2 Traffic, speed variation, 9 nationality, engine capacity, 10 motorcycle ownership, 11 motorist involved in approach turn collisions at signalized junction 12 are some of the factors being found to impact the severity of motorcycle crashes in the literature review.…”

“…Being a branch of artificial intelligence, it models sophisticated data through a series of processing layers. 2 Several studies have applied DL on other transportation engineering aspects such as traffic data imputation, short-term traffic flow prediction, vehicle classification, and sustainable guideline development. The following paragraphs provide a brief review of studies conducted on using DL techniques in transportation engineering.…”

“…This study also found that rider ejection, two-way road with no physical separation, single vehicle, curve aligned roadways, weekend, and young drivers are associated with higher likelihood of fatal crashes. 2 Sameen et al analyzed various DL techniques in predicting the injury severities occurring on Malaysian highways. 20 This study proposed three network architectures based on a simple feedforward neural networks (NN), RNN, and convolutional neural network (CNN).…”

“…Motorcyclists are vulnerable road users because of their high death rate relative to other vehicle users in traffic crashes . An absence of protective structural or the advance safety restraints, unlike other passenger vehicles, makes it more susceptible to severe crashes . Identifying a predictive model to evaluate the relationship between motor cycle injury severity outcomes and the factors associated with this type of crashes are essential in addressing the motorcycle safety issues.…”