Use of SHRP 2 NDS Data and RID in Highway Safety and Operations Analytics
The second Strategic Highway Research Program (SHRP 2) is at the forefront of transportation innovation. The safety focus area of SHRP 2 conducted the largest naturalistic driving study (NDS) in the world to better understand the interaction among various factors involved in highway crashes—drivers, vehicles, traffic, environmental conditions, and infrastructure. The SHRP 2 Roadway Information Dataset (RID) that is geospatial database is produced in conjunction with the NDS data to provide the context (e.g., roadway features and characteristics, traffic volumes, crash histories, and weather) for the NDS trips. The NDS data and RID are available for researchers to explore and uncover the clues of how the interacting factors led to the various roadway safety and operations events. Knowledge gained through this work will help develop better safety and operations countermeasures to save lives, reduce injury severity, and gain better efficiency.
The SHRP 2 safety study produced two major datasets:
(1) NDS Dataset, https://insight.shrp2nds.us/documents/shrp2_background.pdf, which recorded the daily driving activities of 3,400 plus participants in six study sites in the US over a 1 to 2-year period. This dataset consists of detailed vehicle, driver, and driving information on approximately 5.4 million trips taken by the participants during their monitoring period, including over 1,400 crashes and nearly 3000 near-crashes.
(2) Roadway Information Database (RID), http://www.ctre.iastate.edu/shrp2-rid/, a geospatial database that provides the context for the millions of NDS trips.
Besides the original NDS data and RID, there are many reduced datasets produced by the FHWA that streamline working with the SHRP 2 data. There are other datasets such as the National Performance Management Research Data Set (NPMRDS) and Highway Safety Information System (HSIS) that can be conflated with the SHRP 2 data to make these data even more useful.
The applicant will collaborate with FHWA staff members at their Safety Training and Analysis Center (STAC), (http://www.fhwa.dot.gov/research/resources/stac/ ), housed within the Turner-Fairbank Highway Research Center in McLean, VA. The goals of this research could be one or more of the following:
(1) Use different conventional and emerging analytics in the analysis of the SHRP 2 and other conflated datasets to better understand the role of human behavior in occurrence of safety and operations-related events. Examples of these events are crashes, near-crashes, car-following, lane changing, and gap acceptance.
(2) Investigate the contributing factors to crash severity (e.g., roadway characteristics and features, vehicle speed, seat belt use, etc.).
(3) Develop data mining techniques and predictive models for safety and operations outcomes such as crashes, near crashes, and speed.
(4) Explore the use of SHRP 2 data in studying the safety and operational effect of Connected and Autonomous Vehicles (CAVs) with different penetration rates.
(5) Explore the use of SHRP 2 data in studying the advance techniques such as Advanced Driving Assistance Systems (ADASs), Variable Speed Limit (VSL), and Dynamic Message signs (DMSs).
Experience with analyses of SHRP 2 safety data sets is highly desirable. Talented data analysts with innovative ideas and expertise in analyzing big datasets involving geospatial information, large volumes of text data, video data, Artificial Intelligence (AI) and Machine Learning (ML), are encouraged to apply. Candidates demonstrating skills with statistical programs such as R, SAS and geospatial programs such as ArcGIS and proficiency in using Python and Matlab are highly desirable.
The following paper contains a systemic review of the NDS studies, especially SHRP 2 NDS studies and hundreds of related references.
Ahmed, M. M. et al., “Global lessons learned from naturalistic driving studies to advance traffic safety and operation research: A systematic review,” Accident Analysis and Prevention 167 (2022) 106568.
Data, SHRP2, NDS, Safety, Operations, Emerging Analytics, Emerging Data, CAV, ADAS