Safe Road Design through a Human Factors Perspective
Sophia Vardaki (email@example.com), National Technical University of Athens (NTUA)Show Abstract
George Kanellaidis, National Technical University of Athens (NTUA)
The Safe System approach explicitly emphasizes the specific responsibilities of different road safety stakeholders who have a crucial role in realizing a sustainably safe road traffic system. With respect to safety measures implementation, the commitment of road authorities and other traffic professionals to the Safe System is expressed through continuous learning that ensures proper design decisions are made in order to implement a forgiving/safe road infrastructure. Safe human-centered road design needs to reflect Safe System and human factor principles. It also requires road engineers to behighlyaware of human limitations and motivations. Experienced road designers and engineers who have been educated in the context of traditional road design courses and road design guidelines need to assimilate the Safe System approach and become familiar with recent human factor developments. The present paper analyzes key Safe System principles and road design considerations in relation to human factors that can be applied to road infrastructure design, encouraging safer behavior. Safe road design guidelines should incorporate safe principles and concepts that reflect the new holistic and anthropocentric approach to road safety, giving a sound and adequate guide to road designers on the foundation and philosophy of the Safe System approach. A promising way of achieving this objective is the effective integration of relevant information from these fields into road design guidelines.
Influence of Pedestrian-to-Vehicle Technology on Driver Response and Safety Benefits Considering Pre-Crash Scenario Heterogeneity
Lishengsa Yue (2017lishengsa@Knights.ucf.edu), University of Central FloridaShow Abstract
Mohamed Abdel-Aty, University of Central Florida
Yina Wu, University of Central Florida
Jinghui Yuan, University of Central Florida
Morgan Morris, University of Central Florida
Pedestrian-to-vehicle (P2V) technology may offer a promising approach to reducing pedestrian crashes. However, its influences on both driver response and safety benefits have been little studied in previous research, particularly in regard to the variation of influences between different pre-crash scenarios. To investigate these influences, this study designed three pre-crash scenarios based on pedestrian crash contributing factors identified from crash reports, and collected 44 drivers’ driving simulator experiments’ data. The results clarified how using P2V technology to warn drivers of an impending collision improves safety by causing a series of changes for both brake operation and braking profile. These series of changes were further demonstrated to vary between scenarios. The study showed that P2V technology may be particularly useful in scenarios in which a pedestrian’s crossing intention is unclear; specifically, in this type of scenario, the P2V warning had changed the braking process from a panic brake of “slow reaction-hard brake” to a comfortable brake of “quick reaction-gentle brake.” In addition, the P2V warning may be less effective in “low-risk” level scenarios where a driver is confident that he/she can handle the situation through a more conservative evasive action and don’t need to react strongly to a warning. Moreover, depending on the pre-crash scenario, the P2V warning may be mostly beneficial for drivers who had a crash/citation in the past five years and working-aged drivers.
Exploring Longitudinal Driving Behavior on a Freeway Deceleration Lane Using a Field Operational Test Data
Nengchao Lyu (firstname.lastname@example.org), Wuhan UniversityShow Abstract
Yugang Wang, Wuhan University
Chaozhong Wu, Wuhan University
To investigate longitudinal driving behavior in the deceleration lane, 46 participants were recruited to carry out 89 driving tests on a freeway. Operation data was recorded through data synchronization technology. Drivers were grouped according to gender, occupation, and driving experience. Statistical analysis was carried out on driving behavior data, including speed, deceleration, and pedal operation, for different groups of drivers in the deceleration area. The results show that existing deceleration models and lane designs are not optimized for the actual behavior of drivers in the deceleration lane. Additionally, it was found that male drivers prefer to adopt higher driving speeds and more severe braking measures than female drivers, and skilled drivers operate the brake pedal earlier to decelerate and control the speed of the vehicle better than unskilled drivers. The results provide guidance for improving roadway design, traffic management, and driver training.
Understanding Gap Acceptance Behavior at Unsignalized Intersections Using Naturalistic Driving Study Data
Yingfeng Li, Virginia Polytechnic Institute and State University (Virginia Tech)Show Abstract
Haiyan Hao, Virginia Polytechnic Institute and State University (Virginia Tech)
Ronald Gibbons, Virginia Polytechnic Institute and State University (Virginia Tech)
Alejandra Medina, Virginia Polytechnic Institute and State University (Virginia Tech)
Even though drivers disregarding a stop sign has been widely considered a major contributing factor for crashes at unsignalized intersections, an equally important problem that leads to severe crashes at such locations is misjudgment of gaps. This paper presents results of an effort to fully understand gap acceptance behavior at unsignalized intersections using the SHPR2 Naturalistic Driving Study (NDS) data. The paper focuses on the findings of two research activities: the identification of critical gaps for common traffic/roadway scenarios at unsignalized intersections, and the investigation of significant factors affecting driver gap acceptance behaviors at such intersections. The study used multiple statistical and machine learning methods allowing a comprehensive understanding of the gap acceptance behavior while demonstrating the advantages of each method. Overall, the study showed an average critical gap of 5.25 sec for right turn movement and 6.19 sec. for left turn movement. While a variety of factors affected gap acceptance behaviors, gap size, wait time, major road traffic volume, and how frequent the driver drives each year were examples of the most significant.
Effectiveness of Non-Traffic Related Messages on Dynamic Message Signs
Irtiza Rafid Khan, University of Alabama, HuntsvilleShow Abstract
Vishal C. Kummetha (email@example.com), University of South Florida
Alexandra Kondyli, University of Kansas
Steven Schrock, University of Kansas
Dynamic message signs have been in use as early as the 1950s. However, the increase in the use of various messages on these signs by state departments of transportation and other agencies has created a research need to investigative their effectiveness. This research employed surveys and driving experiments in a fixed-base simulator to evaluate the effectiveness of non-traffic-related safety messages on dynamic message signs. Fourteen safety messages consisting of anti-speeding, anti-tailgating, observing the move-over law, and anti-texting were examined. Two participant pools consisting of 120 and 60 participants were recruited for the online surveys and a driving simulator study, respectively. Driving participants were classified into three age groups to determine any age-related performance differences. From the data collected, most of the participants reported seeing and reading the messages. Analysis of the driving data revealed some significant differences in temporary driving performance after perceiving the DMSs. Speed reduction was observed with the display of anti-speeding messages and following gaps were observed to increase. Drivers 50 years old and above experienced the largest increase in following gaps after perceiving the message when compared to the other age groups. Overall, the conclusions suggested that the extent of the performance changes might be of a more temporary nature and do not adversely impact safety.
A Rational Basis for LED Street Lighting Retrofit Luminaire Selection
Jennifer Brons, Rensselaer Polytechnic Institute (RPI)Show Abstract
John Bullough, Rensselaer Polytechnic Institute (RPI)
Daniel Frering, Rensselaer Polytechnic Institute (RPI)
Many municipalities are beginning to undertake efforts to retrofit their existing high pressure sodium (HPS) street lighting with light emitting diode (LED) luminaires. Unlike HPS lighting systems, which are available in a small number of standard wattages and configurations, LED street lighting systems vary widely in wattage and physical configuration. Moreover, LED technological performance continues to improve whereas HPS is a mature technology with substantial improvements unlikely in the future. In order to develop a sound basis for selecting LED lighting systems for retrofit street lighting, photometric simulation calculations under a range of pole spacing, road width and luminaire wattage were performed. The results indicated that LED luminaires can have substantially lower wattage than HPS luminaires in order to produce the same light levels on the road. Further, LED luminaires tend to direct more of their output onto the road compared to HPS luminaires. As a result, LED luminaires can be used that produce substantially fewer lumens overall than HPS systems. Because the white light from LED sources makes illuminated street scenes appear brighter than the yellowish light from HPS lamps, even further reductions in light output can be accomplished with LED street lighting systems to match the same visual effect under HPS.
QUANTIFY THE LINK PERFORMANCE AND CAPACITY WITH DEEP LEARNING AND BPR FUNCTION
Xinhua Wu, Southeast UniversityShow Abstract
Wenbo Zhang, Southeast University
Zhiyuan Liu (firstname.lastname@example.org), Southeast University
Meng Wang, Huawei Technologies Co. Ltd.
The link performance function relates link travel time to traffic volume, playing a vital role in traffic assignment and urban transportation planning. However, the existing studies have two common drawbacks for the determination of link performance functions, which are the inaccurate estimation of the link capacity and the weak generality of default parameters for link performance function. This study proposes a holistic deep learning framework with two major modules to accurately measure link performance as well as capacity. The first deep learning module investigates the capacity with various features of traffic flow, road geometric, and traffic control. The second deep learning module takes the capacity input from the first deep learning module and calibrates parameters to yield reliable link travel time. We also consider minor improvements on neural network components, including the activation function, the pretraining method and data normalization. We develop a case with 800 scenarios in the micro traffic simulation VISSIM, which differs in lane numbers, heavy vehicle proportion, driver behaviour characteristics and signal control strategy. The experiment results show the proposed model does well in quantifying link performance in terms of link travel time. More importantly, the link capacity, although as intermediate feature in the holistic deep learning framework, is close to the historical maximum traffic volume. This heavily supports the model reliability as well as interpretability. Such an accurate, data-driven, and efficient approach has important implications in road supply and link travel time estimation, in particular in dynamic and real-time transportation systems.
Impact of Time Pressure on Driving Behavior during Pre-crossing and Post-crossing Phases of Un-signalized Intersections
Nishant Pawar (email@example.com), Indian Institute of Technology, BombayShow Abstract
Monik Gupta, Indian Institute of Technology, Bombay
Nagendra Velaga, Indian Institute of Technology, Bombay
In recent years, time pressure has been identified as an adverse factor degrading driving behavior. However, very limited studies were conducted in the past to evaluate the effects of time pressure on driving behavior at un-signalized intersections. Therefore, the current study was designed to examine the impacts of time pressure on driving behavior at un-signalized intersections in terms of pre-crossing and post-crossing phases. All the participants were intentionally subjected to different stressful situations in the form of No Time Pressure (NTP), Low Time Pressure (LTP), and High Time Pressure (HTP). Mean deceleration and stop location were analyzed using generalized linear mixed models and half recovery time was modelled using parametric survival analysis. The model results showed that drivers decelerated aggressively because of which they were able to stop the vehicle at a farther distance from the intersection under time pressure conditions. Further, drivers under time pressure required less half recovery time compared to the baseline condition. It was found that HTP had acute effects on mean deceleration (45.53% increase), stop location (14.67% increase), and half recovery time (156.36% decrease) as compared to LTP with respect to baseline condition. Additionally, deceleration rate decreased whereas half recovery time increased with increment in driver’s age, and male drivers stopped closer to the intersection than female drivers. Overall, the study suggests that drivers behaved conservatively by strictly adopting scenario specific strategy under time pressure to compensate the complex driving situation at un-signalized intersections.
Quantification Metrics and Analysis of Human Factor, Workload, and Road Infrastructure
Samuel Tignor, Virginia Polytechnic Institute and State University (Virginia Tech)Show Abstract
The Federal Highway Administration, Office of Highway Safety, created a “Data Driven Safety Analysis (DDSA)” program to assess highway safety. Highway fatalities have hovered around 35,000 for over 40 years emphasizing the need of good DDSA. The paper illustrates a simplified approach for reviewing 2-lane road safety, the largest road system having the most annual fatalities. The paper incorporates the application of various tools highway agencies can use ‘in-house’ as an initial step for analyzing geometric road data. Subsequent, more formal analyses could be done when warranted. Focus is placed on users’ workload, static and dynamic temporal metrics, alternating curve-transit renewal processes to quantify the frequency of user decisions, and the Human Factors Guideline. The example illustrates how highway designers and traffic engineers gain insight into user workload demands. The analysis is a continuation of an earlier FHWA research study illustrating the application of road safety audits and the Interactive Highway Safety Design Model (IHSDM). The IHSDM resource was used because no funding was available for an on-road traffic analysis.
Impact of Connected Vehicle Warning Information on Tunnel Entrance Zone
Zhenlong Li (firstname.lastname@example.org), Beijing University of TechnologyShow Abstract
Guanyang Xing, Beijing University of Technology
Xiaohua Zhao, Beijing University of Technology
haijian li, Beijing University of Technology
Haolin Chen, Beijing University of Technology
The drastic changes of the space environment at the tunnel entrance have led to higher frequency and danger of traffic accidents. This paper designs a tunnel warning system based on connected vehicle, and uses a driving simulator to build a connected vehicle test platform to explore the impact of warning system on the safety of the tunnel entrance zone. Indicators are selected from three aspects including speed control, stability and urgency to analyze the impact of the warning system on the safety of different zones including the warning zone, the last 100m before the tunnel entrance and the first 300m after the tunnel entrance; in addition, the kernel density estimation and cumulative distribution graph are used to analyze the distribution of driver behavior characteristics. The results show that, for the warning zone, the tunnel warning system not only improves the driver's speed control, but also advances the driver's deceleration behavior while the stability is not significantly enhanced; for the last 100m before the tunnel entrance, the tunnel warning system not only improves the driver’s speed control and stability, but also reduces the driver’s urgency for accidents in the tunnel zone; for the first 300m after the tunnel entrance, the tunnel warning system improves the driver's speed control and stability, and also helps reduce speeding behavior. This study introduces the analysis method of the emergency degree of the tunnel entrance accident, and the construction of the connected vehicle tunnel warning system provides an active safety protection approach in the tunnel entrance zone.
Reliance Evaluation of Road Traffic Signs and Navigation Terminal Systems Based on Driver Behavior: A Driving Simulator-based Study
Yang Bian (email@example.com), Beijing University of TechnologyShow Abstract
Xiaolong Zhang, Beijing University of Technology
Xiaohua Zhao, Beijing University of Technology
Jianming Ma, Texas Department of Transportation
Liping Yang, Beijing University of Technology
Road traffic signs and navigation terminal systems are both a very useful tool that provides spatial orientation and direction information, and guides people to conduct activities in various public places. However, the coupling relationship between the two systems is still not entirely clear. This research proposes the concept of driver’s reliance, selects the urban ordinary intersection as the experimental road environment, and takes road guide signs and the audio navigation system as the research objects to analyze the coupling relationship between the two systems. Driving simulator technology is used to reproduce the experimental environment and obtain lane-changing behavior data. Four indicators including lane-changing frequency, lane-changing position, steering wheel angle, and lateral position were selected. The Analysis of Variance (ANOVA) with repeated measures and the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) were employed to analyze the driver’s reliance and construct the comprehensive reliance index. The results show that the audio navigation system, or the participation of the audio navigation system can effectively reduce the driver’s reliance on road guide signs to make decisions, improve the driver’s operational stability, and decrease the driver’s comprehensive reliance, but has no advantage on the vehicle’s operational stability. A major contribution of this study is that it makes clear the change of the driver’s reliance caused by the coupling relationship between road guide signs and the audio navigation system, and thus lays a foundation for improving the safety of driving decision-making and develops reasonable guidelines.
Analysis of Headway and Speed based on Driver Characteristics and Work Zone Configurations Using Naturalistic Driving Study Data
Dan Xu (firstname.lastname@example.org), Auburn UniversityShow Abstract
Chennan Xue, Auburn University
Huaguo Zhou, Auburn University
The objective of this paper is to analyze the headway and speed distribution based on driver characteristics and work zone (WZ) configurations by utilizing the naturalistic driving study (NDS) data. The NDS data provide a unique opportunity to study car-following models for different driver types in various WZ configurations, which cannot be achieved from traditional field data collection. The complete NDS WZ trip data of 200 traversals and 103 individuals, including time-series data, forward-view videos, radar data, and driver characteristics, was collected at four WZ configurations, which encompasses nearly 1,100 miles traveled, 19 hours driven, and over 675,000 data points. First, the time headway selections were analyzed with driver characteristics such as the driver’s gender, age group, and risk perceptions to develop the headway selection table. Further, the speed profiles for different WZ configurations were established to explore the speed distribution and speed change. The best-fitted curves of time headway and speed distributions were estimated by the generalized additive model (GAM). The change point detection method was used to identify where significant changes in mean and variance of speeds occur. The results concluded that NDS data can be used to develop car-following models at WZs, which would be helpful to enhance WZ planning and simulation models by considering different headway distributions based on driver characteristics and their speed profiles traversing the entire WZ.
Roadway Lighting and Non-Motorist Crashes: A Spatial Comparison of Cambridge, MA
Emily Hennessy (email@example.com), University of Massachusetts, AmherstShow Abstract
Chengbo Ai, University of Massachusetts, Amherst
Dark lighting conditions, including those occurring at dawn and dusk, are correlated with increased non-motorist crash frequency due to reduced visibility, but little research has been done which investigates the spatial relationship between roadway lights and non-motorist crashes on a community scale. This research used Kernel Density Estimation methods to calculate the commonalities between geolocated streetlight data and non-motorist-vehicle crashes from 2010 to 2018 in Cambridge, Massachusetts. It was observed that dawn, dusk, and darkness showed a significant correlation between non-motorist crashes and the absence of roadway lighting, all exceeding the control analysis done with crashes occurring in daylight. A Getis-Ord Gi* hot spot cluster analysis was also completed with this data, the results of which showed that the area of the city with the greatest density of streetlights was absent of non-motorist crash hot spots. Future research seeks to corroborate these findings with data from other cities and assess roadway lighting as a facet of pedestrian network connectivity.
Evaluation of LED Roadway Lighting Upgrades on a Major Arterial
John Bullough, Rensselaer Polytechnic Institute (RPI)Show Abstract
The use of light emitting diode roadway lighting is growing throughout the U.S., because of the benefits of longer effective life and reduced energy compared to high pressure sodium (HPS) lighting systems. The technological feasibility of LED roadway lighting systems has been established, but the impacts of LEDs on the economics of roadway lighting are situationally dependent. In many municipalities, roadway lighting is owned and maintained by the local electric utility. Numerous entities suggest that substantial cost savings can be achieved if municipalities own their roadway lighting systems. Utility-owned roadway lighting is still attractive, however, for many municipalities who are not prepared to purchase and maintain their lighting network. Following a pedestrian safety study indicating that improved lighting could result in reductions in pedestrian collisions along a particular major arterial, several LED luminaire types were installed along that roadway by the local utility. The LED lighting was selected to improve the visual conditions while using less energy. Over 30% energy savings were realized, although total utility costs were reduced by less than 5%. Subjective evaluations of the previous and LED lighting revealed that the LED lighting was judged to be a substantial improvement over the HPS lighting conditions.
Assessing the Effectiveness of In-vehicle Highway Back-of-queue Alerting System
Dan Shen, Indiana UniversityShow Abstract
Zhengming Zhang, Purdue University
Keyu Ruan, Indiana University
Renran Tian, Indiana University Purdue University Indianapolis
Lingxi Li, Indiana University
Feng Li, Indiana University
Yaobin Chen, Indiana University
Jim Sturdevant, Indiana Department of Transportation
Ed Cox, Indiana Department of Transportation
This paper proposes an in-vehicle back-of-queue alerting system that is able to issue alerting messages to drivers on highways approaching traffic queues. A prototype system was implemented to deliver the in-vehicle alerting messages to drivers via an Android-based smartphone app. To assess its effectiveness, a set of test scenarios were designed and implemented on a state-of-the-art driving simulator. Subjects were recruited and their testing data was collected under two driver states (normal and distracted) and three alert types (no alerts, roadside alerts, and in-vehicle auditory alerts). The effectiveness was evaluated using three parameters of interest: 1) the minimum Time-to-Collision (mTTC), 2) the maximum deceleration, and 3) the maximum lateral acceleration. Statistical models were utilized to examine the usefulness and benefits of each alerting type. The results show that the in-vehicle auditory alert is the most effective way for delivering alerting messages to drivers. More specifically, it significantly increases the mTTC (30% longer than that of 'no warning') and decreases the maximum lateral acceleration (60% less than that of 'no warning'), which provides drivers with more reaction time and improves driving stability of their vehicles. The effects of driver distraction significantly decrease the efficiency of roadside traffic sign alert. More specifically, when the driver is distracted, the roadside traffic sign alert performs significantly worse in terms of mTTC compared with that of normal driving. This highlights the importance of the in-vehicle auditory alert when the driver is distracted.
DISCLAIMER: All information shared in the TRB Annual Meeting Online Program is subject to change without notice. Changes, if necessary, will be updated in the Online Program and this page is the final authority on schedule information.