A congressionally-mandated naturalistic study was conducted to evaluate the operational, safety, fatigue, and health impacts of the restart provisions in Sections 395.3(c) and 395.3(d) of Title 49, Code of Federal Regulations. A total of 235 commercial motor vehicle drivers representative of the industry contributed data while working their normal schedules, with 181 drivers completing all 5 months of the study. Drivers were monitored via electronic logging devices to track driving and working hours; onboard monitoring systems to detect safety-critical events; wrist actigraph devices for sleep-wake tracking; and smartphone apps for self-ratings of fatigue, sleepiness, stress, sleep quality, and caffeine intake, as well as Brief Psychomotor Vigilance Test (PVT-B) performance testing.

Drivers provided 26,964 days of data (17,628 duty days and 9,336 restart days). A total of 3,287 restart/duty cycle sampling units were analyzed. Statistical comparisons were performed using linear and non-linear mixed-effects modeling designed to ensure results were free of selection bias. An analysis of the safety-critical event data did not identify any differences in performance. Drivers’ fatigue ratings were higher, and sleep quality ratings were lower, during 1-night versus 2-night restarts [Section 395.3(c)]. Drivers averaged slower PVT-B response times and more PVT-B lapses during restarts after 168 hours than prior to 168 hours [Section 395.3(d)]. During restarts, drivers obtained significantly more sleep (on average, 2 hours more per day), and rated their sleep quality higher and their stress lower as compared to duty days, regardless of provision use. Results indicate that restarts serve to mitigate driver fatigue, stress, and sleep loss (i.e., the restart effectively provides the functional equivalent of a “week end” to recover from fatigue and sleep loss).

Commercial Motor Vehicle Restart Study Overview

Public use dataset files in SAS or CSV format.

Commercial Motor Vehicle Restart Study Public Use Dataset (SAS or CSV format)

Analysis result files in SAS or CSV format.

Commercial Motor Vehicle Restart Study Analysis Files (SAS or CSV format)

SAS codebooks used for the CMV Restart Study.

Commercial Motor Vehicle Restart Study Data Dictionaries

Commercial Motor Vehicle Restart Study Final Report

Commercial Motor Vehicle (CMV) Restart Study

The aim of the Motorcoach Driver Fatigue Study was to determine motorcoach driver duty hours, sleep time, fatigue, and performance while operating within the limits of the Federal Motor Carrier Safety Administration motorcoach hours-of-service regulations. Motorcoach drivers actively working as drivers, fit to drive by their company’s standards, and who volunteered to participate were enrolled.

The study collected data from 84 commercial motorcoach drivers for one month to look at duty start time, total duty time, and total sleep time per 24 hours, with sleepiness, fatigue, and performance measured as they were going on and off duty. Drivers worked their normal schedules of duty and rest for the average 31 days of participation. Drivers wore a wristwatch actigraph to measure continuous sleep/wake history and kept a duty/sleep diary. They took a 5-minute psychomotor vigilance test (PVT) and rated their fatigue and sleepiness going on and off duty.

Drivers self-identified as driving for Charter, Tour, Regular Route, or Commuter Express operations. The sample was, on average, middle-aged, overweight, and predominantly male. From the data: 1) duty start times clustered in the morning; 2) average total duty time for duty days was slightly more than 9 hours; 3) average total sleep time per 24 hours was in the range of 7 to 9 hours, with less sleep during on-duty days and more sleep during off-duty days. During on-duty days, longer total duty times were associated with shorter sleep. Drivers performed worse on the PVT and reported increased sleepiness and fatigue at the end of a duty period relative to the beginning. These findings were in the context of an estimated average of 43 hours on duty per week. Thus, drivers in the sample on average started work in the morning, worked approximately 9-hour days, and slightly more than a 40-hour week, and obtained satisfactory amounts of sleep. On average, drivers did not push the limits of the hours-of-service regulations.

Motorcoach Driver Fatigue Study Overview

Motorcoach Driver Fatigue Study Public Use Datasets (CSV format)

Data dictionary for the Restorative Rest for Motorcoach Drivers Study

Motorcoach Driver Fatigue Study Data Dictionary

Motorcoach Driver Fatigue Study Final Report

Motorcoach Driver Fatigue Study

The objective of this study was to prospectively examine a wide array of driver personal and situational factors to determine the prevalence of these factors as well as their relationship to being involved in a crash. This study involved the collection of driver medical, demographic, and other personal factors to examine the relationship between these factors with various safety outcomes. Interested drivers completed initial study materials during a driver orientation session with the participating carrier at one of eight different recruitment sites across the United States. Safety outcomes included crash data from the participating carrier and crashes and moving violation convictions from national datasets. Depending on the analysis, exposure included the driver’s tenure at the participating carrier during the study or the amount of time under observation during the study. Data from over 21,000 drivers were collected, and 20,753 of these drivers were included in the analyses.

Due to Personally Identifying Information (PII) we are only able to share some of the above described data online as a public use dataset. Due to IRB restrictions, the data have been split into unconsented drivers and consented drivers (approximately half of the drivers are in each group).

The following datasets are available:

1. Unconsented drivers (all data) – medical data, carrier crash data, MCMIS data, CDLIS data – this data may only be accessed in the VTTI Secure Data Enclave, please contact DataRepository@vtti.vt.edu for more information or to request a visit to the enclave.
2. Unconsented drivers (public use dataset) – medical data, CDLIS data – this de-identified data is available for public download at the link below.
3. Consented drivers (all data) – initial survey, medical data, carrier crash data, MCMIS data, CDLIS data, follow up survey – this data may only be accessed in the VTTI Secure Data Enclave, please contact DataRepository@vtti.vt.edu for more information or to request a visit to the enclave.
4. Consented drivers (DUL only) – initial survey, medical data, CDLIS data, follow up survey – due to IRB restrictions, this data is only available through a Data Use License; the use of this data will be tracked and users must agree to delete when requested. To request access to this data, please contact DataRepository@vtti.vt.edu.

Please note, regulatory requirements take precedence over the Health Insurance Portability and Accountability Act (HIPAA), therefor HIPAA does not prevent the release of the medical data included in this study. https://www.fmcsa.dot.gov/faq/are-dot-medical-examinations-covered-hipaa

Commercial Driver Safety Risk Factors Study Overview

Public use dataset file in Excel format (xlsm)

Commercial Driver Safety Risk Factors Study Public Use Dataset (Excel format)

Data dictionary for the Commercial Driver Safety - Risk Factors Study

Commercial Driver Safety Risk Factors Study Data Dictionary

Final report of the Commercial Driver Safety - Risk Factors Study

Commercial Driver Safety Risk Factors Study Final Report

Commercial Driver Safety Risk Factors

Highly Automated Commercial Vehicle Sensor Performance Guidelines and Naturalistic Driving Data Analysis

Automated driving system (ADS)–equipped commercial motor vehicles (CMVs) have the potential to reduce crashes by eliminating driver inattention and negative behavior errors. In order to assess the effectiveness of ADS-equipped CMVs, an understanding of baseline safety performance measures for non-equipped trucks is needed. This study analyzed naturalistic continuous field data including 3.12 million miles of Class 8 CMV operations to produce a leading performance baseline of typical CMV maneuvers. This baseline represents rates of vehicle maneuvers produced in real-time. The leading baseline was developed from six maneuvers including speed behavior, longitudinal deceleration, following distance, lateral acceleration, lane deviation, and lane stability. Maneuvers were matched across ten U.S. highway ODDs, nine speed limit categories, and seven number-of-lane categories.

Naturalistic Driving Data for ADS-Equipped CMVs Overview

Naturalistic Driving Data for ADS-Equipped CMVs Public Use Dataset (Excel format)

Data dictionary for the Naturalistic Driving Data for ADS-Equipped CMV Study

Naturalistic Driving Data for ADS-Equipped CMVs Data Dictionary

Final report of the Naturalistic Driving Data for ADS-Equipped CMV Study

Naturalistic Driving Data for ADS-Equipped CMV Final Report

Naturalistic Driving Data for ADS-Equipped CMVs

The objective of this study was to reduce and analyze previously collected heavy-vehicle naturalistic data to better understand crashes involving heavy-vehicle drivers and the efficiency of Commercial Motor Vehicle (CMV) operations. Two key elements of this study were to investigate driver distraction and the role it plays in CMV operations and to answer critical questions related to hours of service (HOS) regulations and fatigue.

More than 3.8 million miles of naturalistic data were collected from seven fleets and 10 locations under the original Onboard Monitoring System Field Operational Test (OBMS FOT) study. A total of 43 motorcoaches, 73 motorcoach drivers, 182 trucks and 172 truck drivers participated in this study. Key findings from the study showed an overall decrease in cell phone use compared to previous research. Hands-free cell phone use was found to be protective as it likely helps drivers alleviate boredom, while hand-held cell phone use was found to be risky as it takes the driver’s attention away from driving tasks. Additionally, the 8th driving hour showed the highest rate of safety critical event occurrence. This study provides needed insight into motorcoach and truck operations. 

The following datasets are available for public download:

1.	All Data – this file includes information on safety-critical events and baseline epochs including time of day, video reduction, hand-free cell phone use, shift data; PERCLOS 3 and ORD. Details of how data reduction was conducted can be found in the final report.
    
    a.	Time of day has been binned into the hour of occurrence, for example if an event took place on March 11 at 2:32 pm, the only information available in the public use dataset is that the event took place in the hour 2 bin. For the exact date and time of the event, please contact DataRepository@vtti.vt.edu to request a visit to the enclave.

2.	Raw Eyeglance – this file includes all eyeglance data that is available for this dataset. Details of how eyeglance was conducted can be found in the final report.
3.	Raw PERCLOS 3 – the All Data file includes the single-value calculation for PERCLOS 3, this file includes the raw data that was used to calculate the PERCLOS 3 single value.

For questions about these datasets, please contact DataRepository@vtti.vt.edu.


Analysis of Distraction and Drowsiness in CMV Drivers Overview

Analysis of Distraction and Drowsiness in CMV Drivers Public Use Datasets in various formats

Analysis of Distraction and Drowsiness in CMV Drivers Public Use Datasets

Analysis of Distraction and Drowsiness in CMV Drivers Data Dictionary

Analysis of Distraction and Drowsiness in CMV Drivers Final Report