Geospatial Tools and Machine Learning to Map Predicted Crash Locations

Combining Geospatial Tools and Machine Learning to Map Predicted Crash Locations

Imagine you’re a state trooper and a vehicle collision has just occurred in your area. You need to get to the location as quickly as possible, attend to injured motorists, clear the road, and get traffic moving again to prevent additional crashes. Where would you position your patrol car to minimize response time?

Resultant partnered with the Indiana State Police (ISP), the Indiana Department of Transportation (INDOT), and the Management Performance Hub (MPH) to create an interactive web-based map that provides law enforcement and the general public relevant information about where crashes have occurred in the past and where they are more likely to occur in the future.

Download Whitepaper

What's Included

Consolidating Data Sources

Data is the engine behind every data science project. Multiple data sources were brought together by Resultant and its partners to power the Daily Crash Prediction Map, including Automated Reporting Information Exchange System (ARIES), U.S. Census and County Business Patterns (CBP), Average Annual Daily Traffic (AADT) from INDOT, and more.

Using Geospatial Tools

Learn how the Daily Crash Prediction Map uses data from the past to predict the future. After sub-dividing the state into one-kilometer squares, the map color codes each square based on the probability of a crash occurring during a three-hour period in the future, like a weather map.

Solving Problems while Preserving Privacy

For projects with data sensitivity, Resultant has developed a proprietary geospatial masking algorithm to preserve the privacy of individual data points while, at the same time, collectively using the data to draw key insights and provide actionable and useful information to system users and administrators.

We’re proud to help organizations thrive, and we’d love to tell you more.

Key Facts

  • Using our solution, officers can better position themselves in high-risk areas, cutting down response times to crash sites.
  • Resultant found that quick responses to a crash involving an injury are associated with a reduced likelihood of the injury being fatal. Seconds can make the difference between death and life; being close to the crash site when it happens matters.
  • The faster an officer can get there, the faster they can get traffic flowing again and prevent another collision. When traffic backs up because of the primary crash, there’s an increased risk for a secondary crash. Full-speed vehicles suddenly colliding with stopped cars can create even more severe injuries than the original accident.
  • The tool enables citizens to take a proactive role in traffic safety. They can adjust their routes or be more conscious of risk in areas with high probability of a crash, leading to fewer crashes.
    Resultant’s geospatial mapping techniques could be applied to many future applications. In addition to turning millions of points into actionable insight and displaying spatiotemporal results of machine learning models in a user-friendly map, real-time traffic, weather forecasts, and image recognition can be incorporated to accurately predict various outcomes.

Ready to challenge your thinking?

Have a question or request for Resultant? Fill out the form and we'll get back to you quickly.