Applying machine learning, text mining, and spatial analysis techniques to develop a highway-railroad grade crossing consolidation model.

The consolidation of Highway-Railroad Grade Crossing (HRGC) is one of the effective approaches to decrease the number of crashes between trains and vehicles. From 2015-2019, there were 57 HRGC crashes at crossings in East Baton Rouge Parish (EBRP), resulting in thirteen injuries with $346,875 cost of vehicle damages. Consolidation programs help to close redundant crossings and thereby decrease the crash risks; however, it is difficult to find the best crossing in a neighborhood for closure. In our previous research working on HRGC consolidation models in 2019, from among four Machine Learning algorithms, eXtreme Gradient Boosting (XGboost) performed better in HRGC prediction models. In continuation of our previous studies on developing a HRGC prediction model, this research employed Text Mining Techniques, and Geospatial Analysis in addition to the XGboost Machine Learning algorithm. The aim was to develop a consolidation model that is customized for local implementation. The results indicated an overall accuracy of 88 % for the proposed model. The relative importance of the variables input to the model was also reported and offers an in-depth understanding of the model's behavior. Considering the different correlation threshold, a sensitivity analysis was also performed on different aggregation gain values. Subsequently, it resulted in the development of a simplified model utilizing 14 variables, with aggregated gain values of 95 % and a correlation threshold of 0.5. Based on this model, 15 % of current highway-rail grade crossings should be closed.

A Comprehensive Railroad-Highway Grade Crossing Consolidation Model: A Machine Learning Approach.

In the United States, there are approximately 212,000 highway-rail grade crossings, some of which experience vehicle-train incidents that often cause a massive financial burden, loss of life, and injury. In 2017, there were 2,108 highway-rail incidents resulting in 827 injuries and 307 fatalities nationwide. To eliminate collision risks, crossing grade separation and active alarm improvement are commonly used. Moreover, crossing closures are considered to be the most effective safety improvement program. While it may be very difficult, and in some cases impossible to close crossings, there are some incentive programs that facilitate the closure process. One of these programs is a working consolidation model that aims to determine crossing closure suitability. Using details of highway-rail grade crossings in the United States and applying an eXtreme Gradient Boosting (XGboost) algorithm, this paper proposes a data-driven consolidation model that takes into consideration a number of engineering variables. The results indicated an overall accuracy of 0.991 for the proposed model. In addition, the developed XGboost consolidation model reported the relative importance of the variables input to the model, offering an in-depth understanding of the model's behavior. Finally, for the practical implementation of the model, a simplified version containing fewer variables was developed. A sensitivity analysis was performed considering the aggregate gain and the different correlation threshold values between variables. This analysis developed a simplified model utilizing 14 variables, with aggregated gain values of 75% and a correlation threshold of 0.9 which would perform similarly to the full model. Based on this model, 62% of current highway-rail grade crossings should be closed.

Market basket analysis of safety at active highway-railroad grade crossings.

INTRODUCTION: Data from the Federal Railroad Administration show high numbers of incidents at the approximately 210,446 highway-railroad grade crossings across the United States. One cause for this unsettling trend is the problem of drivers stopping within the dynamic envelope zone (DEZ) of the train while in queue. This research seeks to study DEZ stopping behavior at highway-railroad grade crossings by evaluating regulatory signage and further analyze variables that may affect this behavior. METHOD: A comparative safety analysis is undertaken to evaluate the effectiveness of the standard "Do Not Stop on Tracks" (R8-8) sign by using percentage change calculations and chi-squared tests. The study then conducts a market basket analysis (MBA) to extrapolate on these results and to identify underlying factors associated with observed driver behavior using variables influenced by visibility, perception, and maneuverability. RESULTS: Rather low reductions in safety violations resulted from the R8-8 installation, including a 2.2% reduction in DEZ stopping behavior and only a slight 3.7% increase in compliance. The results of the MBA identified associations that affirmed assumptions about driver behavior, while other associations were not as direct but altogether helped broaden the understanding of interactions at grade crossings. CONCLUSIONS: This study concluded that the R8-8 had a positive but minimal effect on driver behavior at the grade crossings. The MBA successfully demonstrated its value by providing further insight on the safety analysis and by increasing the number of variables that can be analyzed simultaneously. The methodology offers the scientific community an innovative approach to analyzing driver behavior. Practical Applications: The results identified important variables for developing preventive measures, which will ultimately help reduce safety violations at grade crossings. The MBA can provide practical insight for railroad safety professionals and transportation engineers when determining problematic intersections and can be used to improve the education on grade crossing interactions.

Post and during event effect of cell phone talking and texting on driving performance--a driving simulator study.

OBJECTIVE: A number of studies have been done in the field of driver distraction, specifically on the use of cell phone for either conversation or texting while driving. Researchers have focused on the driving performance of drivers when they were actually engaged in the task; that is, during the texting or phone conversation event. However, it is still unknown whether the impact of cell phone usages ceases immediately after the end of task. The primary objective of this article is to analyze the post-event effect of cell phone usage (texting and conversation) in order to verify whether the distracting effect lingers after the actual event has ceased. METHODS: This study utilizes a driving simulator study of 36 participants to test whether a significant decrease in driver performance occurs during cell phone usage and after usage. Surrogate measures used to represent lateral and longitudinal control of the vehicle were standard deviation (SD) of lane position and mean velocity, respectively. RESULTS: RESULTS suggest that there was no significant decrease in driver performance (both lateral and longitudinal control) during and after the cell phone conversation. For the texting event, there were significant decreases in driver performance in both the longitudinal and lateral control of the vehicle during the actual texting task. The diminished longitudinal control ceased immediately after the texting event but the diminished lateral control lingered for an average of 3.38 s. The number of text messages exchanged did not affect the magnitude or duration of the diminished lateral control. CONCLUSION: The result indicates that the distraction and subsequent elevated crash risk of texting while driving linger even after the texting event has ceased. This finding has safety and policy implications in reducing distracted driving.

Experimental modeling of the effect of hurricane wind forces on driving behavior and vehicle performance.

While traffic planning is important for developing a hurricane evacuation plan, vehicle performance on the roads during extreme weather conditions is critical to the success of the planning process. This novel study investigates the effect of gusty hurricane wind forces on the driving behavior and vehicle performance. The study explores how the parameters of a driving simulator could be modified to reproduce wind loadings experienced by three vehicle types (passenger car, ambulance, and bus) during gusty hurricane winds, through manipulation of appropriate software. Thirty participants were then tested on the modified driving simulator under five wind conditions (ranging from normal to hurricane category 4). The driving performance measures used were heading error and lateral displacement. The results showed that higher wind forces resulted in more varied and greater heading error and lateral displacement. The ambulance had the greatest heading errors and lateral displacements, which were attributed to its large lateral surface area and light weight. Two mathematical models were developed to estimate the heading error and lateral displacements for each of the vehicle types for a given change in lateral wind force. Through a questionnaire, participants felt the different characteristics while driving each vehicle type. The findings of this study demonstrate the valuable use of a driving simulator to model the behavior of different vehicle types and to develop mathematical models to estimate and quantify driving behavior and vehicle performance under hurricane wind conditions.