Geographic Area: New York City
Year Published: 2017
Goal: Analyze near-collision events to improve traffic interventions throughout the city.
Datasets: Mobileye sensor data, Census data
The practice of mapping traffic collisions to understand the most hazardous parts of the urban landscape has become a useful tool for cities to target traffic interventions. However, cities have more recently moved to the next step of traffic safety analysis: examining collisions that never came to be. By looking at near-collision events, municipalities can gain insights into the determinants of accidents without the cost of collisions.
By teaming up with visual analytics company Mobileye, GIS leader Esri has sought to increase understanding of near-collision events with pedestrians in order help cities prevent accidents. Mobileye’s technology works by equipping city vehicles with visual sensors that can detect the locations of pedestrians, and therefore identify situations in which a collision nearly occurred. On its Mobileye and Esri story map, Esri gathered Mobileye data on near misses in New York City in a number of visualizations that enhance understanding of hidden pedestrian risk areas.
The story map provides a visualization of the exact locations of collision warnings across the city, an aggregate of near-collisions by hex bin (100 hexagonal meter areas), as well as a visualization of data by streets. Esri then used demographic data to understand how near-collisions intersect with other indicators, providing insights that could inform cross-sector interventions.
Analysts then clustered data points to reveal hidden patterns across the city. For example, the data showed that most near-collisions happen in the morning during commutes from residential areas, and afternoon near-collisions tend to cluster around business districts.
Analysis also revealed the situations that traditionally lead to near-collisions with pedestrians. While in Manhattan, near misses usually happened with pedestrians walking with traffic, in Brooklyn, near-collisions were more common with pedestrians crossing the street—insights that can help the city direct pedestrian movements in different neighborhoods.
Using this information, Esri also identified hot spots for near-collision events throughout the city, highlighting those areas in greatest need of intervention.
To further understanding of overall trends, mapmakers created a space time cube map representing the number of collision warning each month and an interactive mesh view that shows where near-collisions happen throughout the day. And for users interested in doing their own analysis, Esri provided interactive maps of collision warnings by hex bin and clusters of collisions by time of day.
The Mobileye and Esri story map reveals the great potential offered when mapping meets innovative data collection and analytics technologies. By collecting information previously thought impossible to gather and displaying it in an accessible platform, GIS leaders can push into new areas of analysis and revolutionize our understanding of topics government has studied for decades.