Date on Senior Honors Thesis

5-2020

Document Type

Senior Honors Thesis

Degree Name

B.S.

Department

Geography and Geosciences

Degree Program

College of Arts and Sciences

Author's Keywords

GIS; Spatial; Temporal; Transporation; Lousiville; Planning

Abstract

Walking and cycling are health-conscious, environmentally friendly modes of transportation, yet very few American trips are accomplished using these methods. A major factor behind this is the fear of being involved in a crash with an automobile. From 2009-2019 there were over 5,200 automobile crashes involving either pedestrians or cyclists in Louisville/ Jefferson County, Kentucky. Researchers have found that these kinds of crashes exhibit spatiotemporal patterns in different cities across the globe. The objective of this study was to determine if there exist any spatial and/or temporal patterns regarding these kinds of crashes. Data for this study came from the Kentucky State Police and encompassed all pedestrian and cyclist crashes from 2009-2019. GISsystems were used to perform a network-based kernel density estimation for the spatial analysis. For the temporal analysis, the scales of time, day and month were observed and plotted. Hot-spots were found to exist within the study area, with some locations being hot-spots for both pedestrian and cyclist crashes. These shared hot-spot locations were analyzed in detail, using the original Kentucky State Police data, as well as Google Earth and Street View imagery.

Lay Summary

Walking and cycling are health-conscious, environmentally friendly modes of transportation, yet very few American trips are accomplished using these methods. A major factor behind this is the fear of being involved in a crash with an automobile. From 2009-2019 there were over 5,200 automobile crashes involving either pedestrians or cyclists in Louisville/ Jefferson County, Kentucky. The objectives of this study were to determine if any spatiotemporal (having to do with space and time) patterns exist regarding traffic crashes involving pedestrians and cyclists in Jefferson County, Kentucky. If so, the secondary objective was to determine what, if any, characteristics were present at significant crash locations that may have influenced the presence of crashes. Based on the results of previously mentioned studies showing that spatiotemporal patterns exist regarding these kinds of crashes in different cities, I hypothesized that spatiotemporal patterns do exist regarding traffic crashes involving pedestrians and cyclists in the study area. In this thesis, GISystems were used to determine if these kinds of crashes exhibited patterns relating to space and time within Jefferson County. For the temporal analysis, the scales of time, day and month were observed and plotted. To determine spatial patterns a Network-Based Kernel Density Estimation model was used to determine “hot-spot” locations. The results showed that there are spatiotemporal patterns regarding traffic crashes involving pedestrians and cyclists in Jefferson County. Temporal Data was analyzed independent of location and in conjuncture with location. The top three “hot-spot” locations shared by both pedestrian and cyclist crashes in terms of average density were analyzed to determine potential causal factors that influenced their presence, this was accomplished using a combination of temporal data, google images and street view and local knowledge.

Download

Share

COinS