InTrans / Nov 23, 2021

Data integration efforts aid in assessing crash factors

Skylar Knickerbocker

Crash data provide useful information about some of the factors at the time of the incident. However, the addition of other data sets—such as weather and traffic conditions—can begin to give a fuller picture of the incident and can ultimately lead to better decision-making.

Recently completed research from the Institute for Transportation’s (InTrans’) Center for Transportation Research and Education (CTRE) developed a proof-of-concept architecture to integrate mobility and weather data into available crash data to better understand the role these factors have in crashes.

“Despite access to unprecedented amounts of data, decision-makers are often restricted in their ability to explore multiple data due to factors that include the storage of data in silos and the size and availability of different data sources,” said Skylar Knickerbocker, the project principal investigator and CTRE research engineer.

The developed architecture is useful in itself, but as importantly the research created a foundation for integrating additional data sets in the future.

Additional data sets that can be used to enhance the crash data or used in future research based on input from the Iowa Department of Transportation (DOT) and other relevant stakeholders include the following: Advanced Traffic Management System, snowplow automatic vehicle location (AVL), winter road conditions, traffic and road weather snapshot and videos, pavement condition data, intersections, and work zones.

The architecture includes the following benefits over the existing system:

  • Data from multiple sources are saved in a format that leads to extended ability to query across these data sources. Thus, queries, such as how many crashes occurred during snow and congested conditions, should be easy to perform.
  • High-performance computing systems are used to store and manipulate the data, and hence the data processing time will be significantly reduced.
  • The crash data has been integrated with the roadway asset management system (RAMS) back to 2015 which will provide easier integrations with other data sets

In addition to developing the integration processes, the research team developed a prototype online safety and operations evaluation tool to explore the interactions between the crash, weather, and probe data. While the research team did explore methods of analyzing the data, it expects that future research will further explore and analyze the data to understand the relationships among crash factors.

“This research addresses some of the constraints on decision-makers but also opens up additional data sets for the Iowa DOT or other researchers to explore without the additional time and effort needed to integrate the data,” Knickerbocker said. “Users can spend additional time analyzing the data rather than interpreting and processing the raw data.”

Read more about the project here.