Fork me on GitHub

Project Team 10: Visualizing Boston Crash Statistics, DS 4200 F20

Zhaoze Zhang, Jiaming Hu, Fan Xu

Service-Learning Course Project as part of DS 4200 F20: Information Visualization, taught by Prof. Cody Dunne, Data Visualization @ Khoury, Northeastern University.

Abstract

On average 4–5 people died in a traffic accident every hour in 2019. This visualization aims to raise public awareness of the danger of operating motor vehicles. The visualization will highlight high-risk zones in Boston streets, and allows filtering by road characteristics like surface type and speed limits. It will also explore the correlations between factors contributing and car crashes. Along this project, we will work with Massachusetts government officials to provide in-depth visualization. The visualization will offer a minimalist and interactive interface through HTML and JavaScript for Boston drivers to navigate and explore.

Visualization

Demo Video

Visualization explanation

We used the line chart to show the trend of crashes in Boston by hour. When hovering on the dot, it will display the number of crashes during the corresponding time period. Users can easily see the trend of the crashes happening over the day.The bar chart shows the number of crashes by county, and the map chart displays the distribution of the crashes of selected counties. You can zoom in to see the detailed information. The size of the circle is proportional to the number of crashes that happened within an area.By clicking on the bar, the bar will change color to black, and the corresponding county will be unselected. Then, the crash distribution of that county will disappear. The bar will change back to the original color when you click on it again, and the distribution will emerge.

Upon finishing our visualization, we identified morning and evening rush hour as two peaks of crash risks. However, the evening peak is much higher than the morning peak. We can see that the peak in the graph for the city of Boston is one hour earlier than that of the Boston Urban Area, and begins to fall off sharply an hour after. This can be explained by evening rush hour commuters leaving the city.

These are also some of the heat zones we identified in Boston city proper. A tailgating heat zone in the O'Neill tunnel of I-93. This segment of the roadway is frequently congested and following too closely presents a significant risk of crash. A speeding heat zone at the embankment road near the longfellow bridge. This is another high volume roadway connecting the storrow drive and I-93. The corner in this specific segment is sharper than it may appear, and the bridge blocks the driver's line of sight entering the corner. This gave a lot of drivers false confidence and went in the corner at dangerous speed. The rightmost one is a traffic marking heat zone at Morton Street in Jamaica Plain. This is a high density residential area with a lot of T junctions, crosswalks, and traffic signals.

Our Slides

Acknowledgments

List here where any code, packages/libraries, text, images, designs, etc. that you leverage come from.