|This summary report is an archived publication and may contain dated technical, contact, and link information|
Publication Number: FHWA-HRT-14-077
Date: November 2013
PDF files can be viewed with the Acrobat® Reader®
On November 6–7, 2013, at the Turner–Fairbank Highway Research Center in McLean, VA, the Federal Highway Administration’s (FHWA) Office of Safety Research and Development, with support from the Exploratory Advanced Research (EAR) Program, convened the workshop, ”Utilizing Various Data Sources for Surface Transportation Human Factors Research.” The workshop addressed the increasing number of different datasets and multiple ways of collecting data—from naturalistic driving and simulator studies to eye trackers and surveys—that can be used to increase an understanding of human errors.
Human errors are still a major cause of injuries and fatalities; however, a number of different datasets have recently become available to analyze human errors. These datasets point in different directions within different areas of interaction. Experts in human factors research, transportation safety, and driver behavior and performance analysis, met to discuss and determine which datasets were best and how one might resolve the differences. The information provided by the different datasets is sometimes complementary, sometimes competing, and sometimes confirmatory. The workshop brought together a panel of experts to share their research experience of using multiple methods to gain insights about different aspects of driver and traveler behavior and performance.
During day one of this workshop, participants heard seven presentations on using various datasets from sources such as driving simulators, field studies and field operational tests, and naturalistic driving studies. The experts discussed various methods to study behaviors that lead to errors and shared strategies they have deployed to gain insightful information about what datasets to use to target one or more human factors or behavior issues. The workshop also presented the idea of using multiple data collection methods to “cross-reference” analysis results, validate conclusions, and enhance the understanding of behaviors.
On day two of the workshop, an expert panel discussed issues related to consolidating data from multiple types of collection methods. The experts discussed how datasets must be carefully examined when combined from different sources. For example, some data sources are contradictory, leaving researchers with the need to conduct additional research to resolve the controversies. Alternatively, other data sources can be complementary and provide information in the field and in the laboratory on driver behaviors that point in a similar direction. How best to create complementary datasets also needs to be carefully considered. In addition, very few data sources are comprehensive, and they do not provide information on both driver behavior and crashes. The ability to develop models that can link behavioral datasets with crash datasets, leading to comprehensive datasets, is still in its infancy. The expert panel went on to identify several potential research topics to address the challenges that must be overcome to integrate data from multiple sources.
At the end of day two, the workshop sponsor divided the participants into three groups so that detailed discussion could be held to identify research gaps related to the following interactions of drivers: (1) with other road users, (2) with changing elements of the roadway and infrastructure, and (3) with their own vehicle. All three groups presented summaries of their discussion and recommendations to conclude this workshop.
Workshop panelists and participants noted two different ways of seeing how best to deal with multiple contradictory datasets, as follows:
Panelists were unanimous in recommending that there should be an attempt to understand how to use the different types of data in a study that includes the following components:
As part of the final workshop recommendations, participants identified many areas of priority for human factors research that could make use of the expanding datasets now available and soon to be available. These included modeling, safety, roadway departure, urban intersections, vehicle, pedestrian and bicyclist interaction, and data analysis. Participants suggested a number of specific items for further research, as follows:
To further understanding and use of multiple data types, participants recommended a study, possibly focused at intersections, which includes multiple sites, multiple data types gathered at each site, multiple user types, and multiple methods of analysis. This study could provide critical information on how to resolve contradictions among datasets, how to put together complementary datasets that describe risky behaviors, and how to generate comprehensive datasets that link behaviors and crashes.
Topics: research, exploratory advanced research
Keywords: research, exploratory advanced research, Surface transportation, human factors research, data sources, human errors, datasets, data integration, driving simulators, field studies, field operational tests, naturalistic driving studies
TRT Terms: research, Information organization, Activities leading to information generation, Research, Research projects