Grants and Contracts Details
Description
Abstract
OHS, Kentucky Traffic Incident Management Dashboard Update and
Secondary Crash Identification Automation TIM
The project aims to update the Kentucky Traffic Incident Management (TIM) Dashboard with
additional performance measures and visualizations. The project also aims to improve the existing
secondary crash measure by implementing an effective text mining model that can automate the
identification of secondary crashes using crash narratives. The project will improve the accuracy,
accessibility, and timeliness of crash data analysis.
The Kentucky Traffic Incident Management (TIM) Dashboard enables the KYTC and KSP to track
and monitor incident response performance by providing key metrics, including roadway clearance
time, incident clearance time, first responder vehicle crashes, commercial motor vehicle crashes, and
secondary crashes. The dashboard improves the accessibility and timeliness of crash data. However,
updating the secondary crash data is a challenging task due to the time-consuming process of
manually reviewing thousands of crash narratives. In this study, we aim to implement machine
learning models for automating the secondary crash review process. We will also work with KYTC
and KSP to identify additional needs they may have and expand the functionality of exiting
dashboard. The project will enable the KYTC to keep the TIM Dashboard up to date, facilitating
timely performance tracking and the identification of effective incident management strategies.
1. To automate the identification of secondary crashes and improve their accuracy from less
than 15% as reported in crash database to above 75%.
2. To improve the completeness of the Kentucky Traffic Incident Management (TIM)
Dashboard by adding new performance metrics and visual formats.
Status | Finished |
---|---|
Effective start/end date | 10/1/23 → 9/30/24 |
Funding
- KY Office of Highway Safety: $86,769.00
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.