Transportation Safety Board of Canada

Gatineau, Canada

Transportation Safety Board of Canada

Gatineau, Canada
SEARCH FILTERS
Time filter
Source Type

Rudin-Brown C.M.,Transportation Safety Board of Canada | Roosenboom D.,Transportation Safety Board of Canada | Finch-Field C.,Transportation Safety Board of Canada
Transportation Research Record | Year: 2017

The widespread availability and affordability of voice and video recording systems allows researchers in all modes of transportation to observe, document, and study operator behavior. For example, naturalistic driving studies use technology to record discreetly everyday vehicle operations and driver behavior to understand the risk factors that contribute to unsafe situations. It has been proposed that similar technology could be used in locomotives to observe and evaluate the crew's use of controls and human factor issues. The joint (government-industry) Locomotive Voice and Video Recording (LVVR) project was convened to assess LVVR technology, document safety benefits, and identify best practices. An evaluation of safety-relevant human factors and operational issues was conducted to determine the adequacy of three types of recording systems (voice only, video only, and voice and video) to record and document locomotive crew behavior. All the LVVR systems that were evaluated, regardless of mode, allowed the identification of crews' responses to train control signals. Although it was often possible to identify operators' roles in the cab and their use of locomotive controls, evaluation of human factor issues - such as crew resource management, stress, fatigue, workload, situation awareness, and distraction - was less reliable. Recording modality and system-specific issues that limit the use of LVVR were identified. Collectively, the results indicate that LVVR systems that collect voice and video data and that are of sufficient technical quality to provide clear, unambiguous recordings are the most conducive to the assessment of crew operational and human factor issues. © 2017, National Research Council. All rights reserved.


Rudin-Brown C.M.,Transportation Safety Board of Canada | George M.F.-S.,Transportation Safety Board of Canada | Stuart J.J.,Transportation Safety Board of Canada
Transportation Research Record | Year: 2014

Collisions between road vehicles and trains at passively controlled level crossings, where no indication is given to drivers of the presence (or absence) of an approaching train, raise important human factors safety concerns. A database search of Canadian level crossing crash data for the 10-year period between 2003 and 2012 and a targeted review of the scientific research literature were conducted to explore the human factors and related risks involved in crashes of this nature. Accidents at passively controlled, rural level crossings where the driver of the road vehicle did not stop constituted 15% of all level crossing crashes. Statistical analysis revealed that this type of accident was most likely to occur in prairie provinces, during daylight hours, and to involve a dis-proportionate number of heavy vehicles. Nine human factors Issues were identified. Four affected driver detection of an approaching train, including sightlines, train conspicuity, unchanging retinal image, and train horn audibility. Five issues related to driver decision making, including looked-hut-failed-to-see errors, faulty activation of schema-mental model, distraction, impairment, and information processing. These issues were overlaid across four previously proposed driver approach zones at passive level crossings. Analysis of countermeasures designed for application within the approach zone, which began at the decision sight distance point, revealed several potential countermeasures, including the observation that current minimum sightline guidelines for passive level crossings might not have been adequate to ensure that drivers have sufficient time to assess the threat posed by an approaching train from both directions and to respond appropriately.


Henderson S.,Transportation Safety Board of Canada | Woods-Fry H.,University of Ottawa | Collin C.A.,University of Ottawa | Gagnon S.,University of Ottawa | And 4 more authors.
Accident Analysis and Prevention | Year: 2015

Our research group has previously demonstrated that the peripheral motion contrast threshold (PMCT) test predicts older drivers' self-report accident risk, as well as simulated driving performance. However, the PMCT is too lengthy to be a part of a battery of tests to assess fitness to drive. Therefore, we have developed a new version of this test, which takes under two minutes to administer. We assessed the motion contrast thresholds of 24 younger drivers (19-32) and 25 older drivers (65-83) with both the PMCT-10 min and the PMCT-2 min test and investigated if thresholds were associated with measures of simulated driving performance. Younger participants had significantly lower motion contrast thresholds than older participants and there were no significant correlations between younger participants' thresholds and any measures of driving performance. The PMCT-10 min and the PMCT-2 min thresholds of older drivers' predicted simulated crash risk, as well as the minimum distance of approach to all hazards. This suggests that our tests of motion processing can help predict the risk of collision or near collision in older drivers. Thresholds were also correlated with the total lane deviation time, suggesting a deficiency in processing of peripheral flow and delayed detection of adjacent cars. The PMCT-2 min is an improved version of a previously validated test, and it has the potential to help assess older drivers' fitness to drive. © 2015 Elsevier Ltd. All rights reserved.


PubMed | University of Ottawa, Transportation Safety Board of Canada and Systems Technology Incorporated
Type: | Journal: Accident; analysis and prevention | Year: 2015

Our research group has previously demonstrated that the peripheral motion contrast threshold (PMCT) test predicts older drivers self-report accident risk, as well as simulated driving performance. However, the PMCT is too lengthy to be a part of a battery of tests to assess fitness to drive. Therefore, we have developed a new version of this test, which takes under two minutes to administer. We assessed the motion contrast thresholds of 24 younger drivers (19-32) and 25 older drivers (65-83) with both the PMCT-10min and the PMCT-2min test and investigated if thresholds were associated with measures of simulated driving performance. Younger participants had significantly lower motion contrast thresholds than older participants and there were no significant correlations between younger participants thresholds and any measures of driving performance. The PMCT-10min and the PMCT-2min thresholds of older drivers predicted simulated crash risk, as well as the minimum distance of approach to all hazards. This suggests that our tests of motion processing can help predict the risk of collision or near collision in older drivers. Thresholds were also correlated with the total lane deviation time, suggesting a deficiency in processing of peripheral flow and delayed detection of adjacent cars. The PMCT-2min is an improved version of a previously validated test, and it has the potential to help assess older drivers fitness to drive.

Loading Transportation Safety Board of Canada collaborators
Loading Transportation Safety Board of Canada collaborators