Name: TrafficCruising-DSSG2017
Owner: UW eScience Institute
Description: Public Repository for Traffic Cruising DSSG 2017 Project
Created: 2017-08-18 17:25:56.0
Updated: 2017-08-18 21:00:47.0
Pushed: 2017-08-31 16:33:33.0
Homepage: null
Size: 39841
Language: Python
GitHub Committers
User | Most Recent Commit | # Commits |
---|
Other Committers
User | Most Recent Commit | # Commits |
---|
A substantial portion of urban traffic can result from drivers searching for parking spaces and waiting for (or traveling between) passengers, in the case of for-hire vehicles. These driving patterns are known collectively as traffic cruising, and are thought to be major contributors to congestion in downtown Seattle. However, the magnitude and location of traffic cruising are poorly understood. By analyzing traffic sensor data, we devised a system for visualizing temporally- and spatially-explicit variations in the intensity of traffic cruising across Seattle's Central Business District. We use a series of heuristics and algorithms to estimate vehicle routes from anonymous data. We then describe each route with calculated metadata, by which we label it as either demonstrating cruising behavior or not, using a semi-supervised machine learning approach. Finally, we create an interactive heat map of downtown Seattle that can be used to visualize magnitudes of each type of traffic cruising pattern.
This research has the potential to help transportation agencies, technology companies, and car companies predict the availability of parking and more accurately direct travelers with online, mobile, and connected tools, thereby reducing congestion impacts, emissions, and fuel costs.
DSSG Fellows
Data Science Leads
Project Lead