Correlation Analysis of Roadway Light Levels, Hard Braking, and Crashes
Loading...
Authors
MacGill, Eric
Issue Date
2024
Type
Thesis
Language
Keywords
Correlation , Hard Braking , Nighttime Crashes , Street Lighting
Alternative Title
Abstract
This study analyzed three primary datasets to investigate the correlation between street lighting illuminance, vehicle braking behavior, and crashes on urban and suburban roadways in Washoe County, Nevada. The first dataset, Street Lighting Illuminance Data, collected from March through December 2020, provides comprehensive illumination data for all urban and suburban roads within Washoe County which mainly consisted of the Reno and Sparks Truckee Meadows area. The second dataset, Washoe County, Nevada, urban and suburban Wejo Speed Connected Vehicle Dataset, collected from March 1 through March 16, 2021, provides detailed vehicle trajectory information, from which acceleration and deceleration data were extracted using a Python script. Positive values indicated acceleration, while negative values denoted deceleration. The third dataset, Washoe County, Nevada, urban and suburban traffic crash dataset, collected from the Nevada Crash Reporting System for years 2013-2017, provides detailed vehicle crash information for the same study area. A sequential methodology was employed to overlay the illumination levels with the deceleration and vehicle crash data. This involved defining a 100-foot, 300-foot, and 500-foot diameter buffer at the center of intersections and a 120-foot perpendicular buffer from the centerline of the roadway segment, along the 1/4-mile long roadway segments within the study area, where illumination, deceleration, and crash data were analyzed concurrently. The study further refined its analysis by employing Geographic Information System (GIS) techniques for visualizing the relationship between ambient illumination levels and hard braking events. This visualization facilitated an initial understanding of the potential correlation between the two variables. Subsequently, a coefficient to quantify the relationships. The study findings suggest a moderate to strong positive correlation between average illumination and nighttime crashes at both intersections and roadway segments. The findings also suggest a negligible to weak positive correlation between average illumination and hard braking events at intersections and roadway segments. This insight can be important for urban planning and traffic safety studies. One of the three main contributions of this study lies in its establishment of criteria for evaluating hard braking through the analysis of multi-second vehicle trajectory data. This new criteria does not contradict already established definitions for hard braking but instead complements these definitions by allowing multi-second data to more easily be evaluated and included in the research. Additionally, another main contribution of this study is the correlation analysis between the three datasets- lighting, hard braking events, and vehicle crashes. The last main contribution of this study is identifying that there is a negligible to weak positive correlation between average illumination levels and hard braking events. Such findings are significant for the formulation of traffic safety initiatives and the improvement of urban road lighting policies. By analyzing the relationships between roadway illumination, hard braking, and crashes, this research offers critical insights that can inform the development of targeted improvements aimed at enhancing traffic safety and urban driving conditions.