Rural Speed Safety Project for USDOT SDI
Statistical Analysis and Decision Support Tool
Subasish Das, Ph.D.
November 16, 2018
Outline
- Overview
- Data Preparation
- Exploratory Data Analysis
- Statistical Models
- Decision Support Tool
- Next Steps
Overview
- About this Project
- One of the Four USDOT Safety Data Initiative (SDI) Projects
- Research Problem
- A data-driven procedure to better understand the relationships between motor vehicle travel speed, roadway design elements, traffic volumes, and crash risks and severities has not yet established.
- Objective
- Examine the prevailing operating speeds on a large scale to see how speed and speed differentials interact with roadway characteristics to influence the likelihood of crashes.
Data Preparation
- Three States: North Carolina, Washington, and Ohio
- Three Databases
- Highway Safety Information System (HSIS)
- Highway Performance Monitoring System (HPMS)
- National Performance Management Research Data Set (NPMRDS)
- Year: 2015
- Task
- Conflate 3 databases for 3 states
Safety Performance Functions (SPFs)
![]()
SPFs at Segment level (Washington)
![]()
SPFs at Segment-temporal level (Washington)
![]()
Proximity of Crashes at Segment-temporal Level
![]()
Decision Support Tool
Live app link: https://ldwhite.shinyapps.io/RuralSpdSafety_Demo/
Work Progress
- Developed SPFs at both segment and segment-temporal levels for different facility types
- SPFs show that speed measure has effect on crash frequencies
- Developed time-series model to forecast speed
- Analysis on proximity of crashes at segment-temporal level aimed to classify crash and non-crash event
- Developed Decision Support Tool at segment level for Washington and Ohio
Next Steps
- Refine segment-temporal level SPFs
- Refine analysis on proximity of crashes at segment-temporal level
- Add North Carolina data in Decision Supprt Tool
Questions?
SPFs at Segment level (Washington)
![]()
SPFs at Segment-temporal level (Washington)
![]()
Proximity of Crashes at Segment-temporal Level
![]()
Decision Support Tool
Live app link: https://ldwhite.shinyapps.io/RuralSpdSafety_Demo/
Work Progress
- Developed SPFs at both segment and segment-temporal levels for different facility types
- SPFs show that speed measure has effect on crash frequencies
- Developed time-series model to forecast speed
- Analysis on proximity of crashes at segment-temporal level aimed to classify crash and non-crash event
- Developed Decision Support Tool at segment level for Washington and Ohio
Next Steps
- Refine segment-temporal level SPFs
- Refine analysis on proximity of crashes at segment-temporal level
- Add North Carolina data in Decision Supprt Tool
Questions?
SPFs at Segment-temporal level (Washington)
![]()
Proximity of Crashes at Segment-temporal Level
![]()
Decision Support Tool
Live app link: https://ldwhite.shinyapps.io/RuralSpdSafety_Demo/
Work Progress
- Developed SPFs at both segment and segment-temporal levels for different facility types
- SPFs show that speed measure has effect on crash frequencies
- Developed time-series model to forecast speed
- Analysis on proximity of crashes at segment-temporal level aimed to classify crash and non-crash event
- Developed Decision Support Tool at segment level for Washington and Ohio
Next Steps
- Refine segment-temporal level SPFs
- Refine analysis on proximity of crashes at segment-temporal level
- Add North Carolina data in Decision Supprt Tool
Questions?
Proximity of Crashes at Segment-temporal Level
![]()
Decision Support Tool
Live app link: https://ldwhite.shinyapps.io/RuralSpdSafety_Demo/
Work Progress
- Developed SPFs at both segment and segment-temporal levels for different facility types
- SPFs show that speed measure has effect on crash frequencies
- Developed time-series model to forecast speed
- Analysis on proximity of crashes at segment-temporal level aimed to classify crash and non-crash event
- Developed Decision Support Tool at segment level for Washington and Ohio
Next Steps
- Refine segment-temporal level SPFs
- Refine analysis on proximity of crashes at segment-temporal level
- Add North Carolina data in Decision Supprt Tool
Questions?
Decision Support Tool
Live app link: https://ldwhite.shinyapps.io/RuralSpdSafety_Demo/
Work Progress
- Developed SPFs at both segment and segment-temporal levels for different facility types
- SPFs show that speed measure has effect on crash frequencies
- Developed time-series model to forecast speed
- Analysis on proximity of crashes at segment-temporal level aimed to classify crash and non-crash event
- Developed Decision Support Tool at segment level for Washington and Ohio
Next Steps
- Refine segment-temporal level SPFs
- Refine analysis on proximity of crashes at segment-temporal level
- Add North Carolina data in Decision Supprt Tool