Chicago Vision Zero Study
PROJECT CONTEXT
The Chicago Vision Zero (VZ) project is an initiative dedicated to eliminating traffic fatalities and serious injuries in Chicago through strategic, data-driven approaches. This comprehensive effort leverages advanced data analytics, geographic information systems (GIS), and machine learning models to understand and address the root causes of traffic incidents. By examining detailed crash data over a five-year period, the project identifies high-risk areas, critical times, and vulnerable populations to implement targeted safety measures. Chicago Vision Zero underscores the city's commitment to creating a safer, more equitable transportation environment for all its residents and visitors, striving towards the ultimate goal of zero traffic-related fatalities.
538k*
Crashes recorded
10.6k*
Fatal & serious injuries recorded
22.2k*
Crashes involving bicycle and/or pedestrians
*data reported from September 2017 to August 2022
DASHBOARD
Crash data were gathered from the Chicago Open Data Portal, focusing on both the incidents and the individuals involved. The analysis incorporated numerical, spatial, and time-domain evaluations. A Power BI dashboard was developed to provide a visual representation of the data, enabling users to effectively slice and dice information. Key metrics displayed include the number of fatalities, serious injuries, and total crashes, with breakdowns by gender, type of crash, and temporal patterns.
Temporal Crash Insights
This analysis helps identify critical periods for implementing interventions to enhance road safety in Chicago.
-
Weekend Nights: Increased crash rates likely due to social activities, suggesting the need for heightened safety measures.
-
Thursday and Friday Nights: A rise in crashes post-work, indicating the importance of targeted safety campaigns during these times.
-
School and Evening Commutes: Notable peaks in crashes necessitate specific safety measures during these peak hours.
GIS INTEGRATION
Geospatial Mapping
Crash data is integrated within the dashboard using GIS mapping capabilities to intuitively visualize crash locations. By leveraging GIS tools, we can spatially analyze crash data and identify high-risk areas, aiding in targeted interventions to improve safety throughout the city.
GIS Space-Time Trend Analysis
The GIS Space-Time Trend analysis uses spatial and temporal data to identify significant trends over the year when crashes are grouped in census tracts for the city of Chicago. This analysis measures the trend in crash counts over time, providing a trend score and p-value to indicate the statistical significance of these trends.
This visualization aids in understanding where crash frequencies are increasing or decreasing, helping to focus safety measures and interventions in areas most impacted by significant changes.
GIS Hot Spot Analysis
The GIS Hot Spot analysis categorizes geographic areas based on the patterns of traffic crashes, such as new hot spots, consecutive hot spots, and intensifying hot spots. This approach pinpoints areas with a high concentration of crashes by examining the frequency and severity of incidents within a 1-mile radius, comparing adjacent census tracts.
The map below visualizes these patterns, effectively highlighting regions with persistent or emerging safety concerns. This spatial representation allows for targeted interventions and enhanced monitoring in areas identified as critical hot spots.
ANALYZING TRENDS
Data from Replica was used to produce visualizations in FSQ Studio, and analyze walking and biking patterns among residents from disinvested neighborhoods. We focused on severe injury and fatal crashes that occurred during a typical workweek in Fall 2019 to understand the spatial relationship between where people live and where they are involved in crashes, providing valuable insights for targeted safety improvements and resource allocation.
Walking and Biking Patterns
-
The dashboard leverages subscription-based data to examine network link volumes, providing insights into walking and biking traffic patterns.
-
The map's swipe function highlights how traditional equity analyses often miss capturing vulnerable groups who are exposed to traffic crashes outside their home geographies.
-
This enhanced visibility assists in understanding and addressing the mobility and safety needs of these at-risk populations more effectively.
Crash Locations Relative to Zipcodes
The dashboard distinguishes between the residence zip codes and actual crash locations to identify and highlight patterns of traffic incidents, such as the tendency for residents from specific zip codes to be involved in crashes at certain locations.
TIME-SERIES FORECASTS
A time-series forecast is a statistical technique used to predict future events by analyzing trends and patterns in historical data. The process involved a detailed analysis of each census tract over a designated time period. This procedure was methodically applied to 800 different census tracts to ensure comprehensive coverage and accuracy.
Forecast Insights
- The time series graph displays the daily recorded crashes, highlighting spikes and trends over the years.
- The ML forecast provides predictions with different confidence intervals, helping to understand the potential range of future crashes.
- The results were compiled into a comprehensive data table for further analysis and planning.
Data collected from all 800 tracts were then aggregated and analyzed in FSQ Studio.
Forecast Insights
-
For each census tract, the forecast predicts the daily number of crashes over a 14-day period.
-
A color gradient on the map visualizes the likelihood of crashes: darker shades indicate a higher probability of experiencing 7 or more crashes, while lighter shades suggest fewer crashes.
-
This spatial visualization is instrumental in pinpointing areas with elevated crash risks, thereby facilitating targeted interventions and strategic resource allocation to enhance road safety.