Safety Evaluation of Variable Speed Limit System in British Columbia
Mohamed El Esawey, British Columbia Ministry of Transportation and InfrastructureShow Abstract
Joy Sengupta, British Columbia Ministry of Transportation and Infrastructure
John Babineau, British Columbia Ministry of Transportation and Infrastructure
Emmanuel Takyi, University of Saskatchewan
Mohamed Essa, University of British Columbia
Adverse weather conditions create a difficult environment for drivers to navigate safely. This study reports the safety impacts associated with the installation of Variable Speed Limit System (VSLS) on provincial rural highways in in British Columbia (BC), Canada. A VSLS is an advanced intelligent transportation system (ITS) scheme that can be employed to increase the safety level of highway facilities by varying the speed limit according to downstream operational condition and/or current weather conditions. The analysis made use of police-attended serious crashes (i.e. fatal + injury) that took place during winter seasons (October to March). Three winter seasons were available as a before-implementation period, and three winter seasons were available as an after-implementation period. The results of a simple-before-and-after were promising where overall reductions of 35.8% and 36.8% in winter serious collision (WSC) frequency and rate, respectively, were found for the evaluation corridors. An Empirical Bayes (EB) before-and-after safety evaluation was also carried out to ensure that the results are reliable. Safety Performance Functions (SPFs) were developed using data collected at similar sites. The EB analysis showed an overall statistically significant reduction of 34.4% in WSC. An economic assessment of the system was undertaken and the results showed that the benefits of implementing a VSLS exceeded the system cost with an overall benefit-cost (B/C) ratio of 4.3 and a NPV of C$34.41 million. The results of this study may motivate transportation agencies and stakeholders who are interested in pursuing similar systems for mitigating weather-related safety challenges.
Crash Modification Factors for Adaptive Traffic Signal Control: An Empirical Bayes Before-After Study
Houjun Tang, Pennsylvania State University, University ParkShow Abstract
Vikash Gayah (email@example.com), Pennsylvania State University
Eric Donnell, Pennsylvania State University, University Park
Adaptive traffic signal control (ATSC) is often deployed at high-volume intersections in order to mitigate traffic congestion and improve travel time reliability. While past studies have demonstrated its operational effectiveness, relatively few have focused on safety performance. Those that have tend to suffer from limitations including small sample sizes, insufficient study designs, or the lack of consideration of potential temporal and corridor effects after ATSC installation. Furthermore, results from previous studies are mixed: while many studies point to a safety improvement, more recent studies seem to indicate that ATSC systems might increase crash frequency. In this study, a comprehensive Empirical Bayes (EB) before-after observational study was conducted using ATSC data collected throughout Pennsylvania. Crash modification factors (CMFs) were estimated based on the following different case scenarios: crash severity levels and crash types (total, fatal and injury, rear-end, and angle crashes); intersection locations (all intersections and intersections along corridors only); and, intersection configurations (3-leg and 4-leg). Corridor-level CMFs were also developed to quantify changes in safety performance along corridors with ATSC installed. The results suggest that ATSC is associated with a nominal increase in total and angle crashes, and an expected decrease in fatal plus injury crashes and rear-end crashes. However, the results were not statistically significant. The safety effect estimates are similar when considering intersection locations and configurations. Finally, the magnitude of the corridor-level CMFs are slightly lower than the intersection-level CMFs, except for rear-end crashes.
Context-based Crash Modification Factors for Medians on Rural Four-Lane Roadways: A Bayesian Approach
Xiaobing Li, University of AlabamaShow Abstract
Jun Liu, University of Alabama
Chenxuan Yang, University of Alabama
Timothy Barnett, University of Alabama
Rural four-lane roadways provide important transportation accessibility and mobility to populations in rural areas. Practitioners are often challenged to determine cross-section types when both benefits and costs need to be considered. The Crash Modification Factors (CMFs) are often developed to evaluate safety effectiveness of alternative designs. However, safety effectiveness could vary significantly across contexts. Thus, the study aims to estimate CMFs for alternative cross-sections of rural four-lane roadways under different contexts characterized by traffic volume, truck percentage, and access point density. Using Georgia state-wide crash data, this study developed Safety Performance Functions (SPFs) to predict crash frequencies for different contexts. Considering linearity and independence assumptions of traditional negative binomial SPFs, this study adopts the Bayesian generalized negative binomial (BGNB) modeling approach to relax those assumptions and only follows the Bayes rule to form SPFs for CMF estimation. This study focuses on four typical cross-sections including 1) non-traversable medians; 2) two-way-left-turn lanes; 3) 4-ft flush medians; and 4) undivided roadways with double-yellow lines (the base cross-section). The results show that CMFs vary significantly across different contexts. Compared with base cross-section design, safety benefits of other three designs can be either positive or negative under different traffic or road conditions. For example, 4-ft flush medians are found to have positive safety benefits (CMF < 1) under lower average daily traffic volumes (e.g., <=6,000); but negative benefits (CMF >1) under greater average daily traffic volumes (e.g., >=15,000). The findings offer practitioners insights that cross-section designs may need to be varied for different contexts.
Safety Assessment of Existing Roads: A Preliminary Comprehensive Methodology for Restricted Environments for a Safety Improvement Program
Alaa Torkey, Cairo UniversityShow Abstract
Abdulrahman Alrafee, Cairo University
Dalia Said, Cairo University
Road safety has become a leading topic in recent scientific research. Researchers have studied the effect of different operational and geometric characteristics on road safety such as sight distance, operating speed profiles leading to design consistency evaluation and application of the Highway Safety Manual (HSM) for the development of local safety performance functions. Considering safety improvement projects, a comprehensive study is important to get an overall view of crashes’ causals, which would be beneficial in conducting cost-effective safety improvement measures especially in restricted environments where there is a lack of crash data. Surrogate methods are needed then for effective safety assessment. This paper presents a preliminary comprehensive operational analysis using an existing alignment in El Mansoura Governorate in Egypt to appraise the study objective. The analysis begins with sight distance analysis, then a design consistency evaluation and finally using HSM Safety Performance Function for rural two-lane two-way roadways. Two scenarios of remedial measures are suggested; one scenario is based on the comprehensive methodology suggested and the other one focuses on the road element characteristics which is a typical approach if HSM methodology is used. Benefit-cost ratio analysis is used for measuring the economic justification of each scenario. It is suggested to study the possibility of developing safety performance functions having sight distance and design consistency all together as building parameters which could be of great help especially in countries having poor crash data.
Sun Glare and Traffic Crashes: Identifying Patterns of Key Factors
Subasish Das, Texas A&M UniversityShow Abstract
Xiaoduan Sun, University of Louisiana, Lafayette
Bahar Dadashova, Texas A&M Transportation Institute
M. Rahman, University of Louisiana, Lafayette
Ming Sun, University of Louisiana, Lafayette
Sun glare is one of the major environmental obstructions that cause traffic crashes. Every year, many traffic crashes in the United States are attributed to sun glare. However, quantitative analysis of the influence of sun glare on traffic crashes has not been widely examined. This study used traffic crash narrative data for seven years (2010-2016) from Louisiana to identify crash reports that provided evidence of drivers indicating sun glare as the primary contributing factor of the crashes. Additional geometry and traffic information was collected to identify the list of key crash-contributing factors. This study used cluster correspondence analysis to perform the data analysis. After performing several iterations, six clusters were identified that provide additional insight regarding sun glare related crashes. The six clusters are associated with mixed (business and residential) localities, intersection related crashes on U.S. roadways, single vehicle crashes on residential two-lane undivided roadways, curve related crashes on Parish roadways in residential localities, interstate related crashes in open county localities, and curve related crashes in open county localities. The findings of the current study can add insights into the ongoing safety analysis on sun glare related crashes.
Effect of Motorcycle Composition to Motorcyclist and Other Motor Vehicle Accident Rate in Mixed Traffic Condition
Rizky Junirman (firstname.lastname@example.org), Shuto Daigaku TokyoShow Abstract
Hiroyuki Oneyama, Tokyo Metropolitan University
The disparity in road safety between low- and middle-income countries (LMIC) and high-income countries (HIC) is high. To confront this problem, local road characteristics must be understood. This study attempts to clarify how traffic composition contributes to accident rates in mixed traffic conditions for different road users in Indonesia. The study was conducted within an urban environment in Indonesia; hence, other contributing road elements to support this study are chosen to represent such conditions. Multivariate analysis using negative binomial regression revealed that motorcycle and motor vehicles have a slight difference in contributing factors to their respective accident rate. Furthermore, the motorcycle proportion contributes to accident risk for both motorcycle and motor vehicle. The two-fluid model and road type for undivided roads are not significant for both models. The difference lies in the fact that 1 km radii of through traffic are significant only for motorcycle accident rates. For other variables, they share the same significant variables. The significant variables are motorcycle percentage, network centrality for 10 km radii, access density, signalized intersection type, and road type for divided roads.
Safety Impact Evaluation of Narrow AV-Exclusive Lanes on Existing Freeways
Benjamin Melendez, San Diego State UniversityShow Abstract
Anagha Katthe, San Diego State University
Arash Jahangiri (AJahangiri@sdsu.edu), San Diego State University
Sahar Ghanipoor Machiani, San Diego State University
Alidad Ahmadi, Linscott, Law & Greenspan, Engineers
Walter Musial, Linscott, Law & Greenspan, Engineers
A full infrastructure adaptation to the emerging Automated Vehicle (AV) technology is not going to happen soon, especially given that transportation system will be serving both AVs and conventional vehicles for a while. On freeways, co-existence of AV-exclusive lanes and conventional vehicle lanes seems to be a viable solution. It has been suggested that AVs’ navigation capabilities could allow for infrastructure standard adjustments such as narrower lanes. Given the difference between the operation of AVs and human-driven vehicles and reliance of AVs on sensors as opposed to human capabilities, the questions are can we provide exclusive and narrower roadways for AVs while maintaining proper safety and what are the safety implications of a narrow AV-exclusive lane on a freeway? To answer these questions, this study conducted three tasks. First, a comprehensive review of existing AV technologies related to lateral control systems was compiled for fifteen different vehicle manufacturers. Then, consumers’ complaints from the NHTSA Vehicle Owners’ Questionnaires database were investigated to evaluate safety issues that consumers encountered related to lateral vehicle control technology in AVs. Finally, an expert interview was conducted to survey and explore relevant academic researchers’, transportation officials’, and industry leaders’ attitudes and opinions on the topic of AV-exclusive lanes. Conclusions were drawn and a series of recommendations were developed from the results of the study that are usable for practitioners and professional organizations pertaining to AV development.
Investigating Performance of a Novel Safety Measure for Assessing Potential Rear-End Collisions: An Insight Representing a Scenario in Developing Nation
Narayana Raju, Sardar Vallabhbhai National Institute of TechnologyShow Abstract
Shriniwas Arkatkar (email@example.com), Sardar Vallabhbhai National Institute of Technology
Said Easa, Ryerson University
Gaurang Joshi, Sardar Vallabhbhai National Institute of Technology
Road safety is one of the major concerns in the ever-growing traffic network. In addressing this, surrogate safety measures play a critical role in identifying collision instincts. Besides the added advantage of quantifying collision instincts in advance, surrogate safety measures have their limitations. For example, in some instances, those measures tend to show erroneous results. In this paper, a new surrogate safety measure Instant Heeding Time (IHT), is presented based on follower vehicle attention in the traffic streams. This new measure is integrated with a distance gap and the vehicles' speeds to assess probable rear-end collisions. Further, along with other safety measures, the developed safety framework is tested over a study section, with the help of trajectory datasets at three traffic flow conditions (free flow, capacity, and congested) under prevailing heterogeneous (mixed) traffic conditions. Based on the safety framework, it is observed that, in the case of free flow and capacity conditions, 23 and 55 probable rear-end collisions points are detected. At the congested conditions, no rear-end collision points are observed. Further, smaller vehicles in the traffic stream are associated with a higher number of rear-end collision instincts than other vehicle categories. The conceptualized safety framework can be applied on a real-time basis for monitoring the safety measures for vehicles in a mixed traffic stream.
Examining Driver Compliance with a Move-Over/Slow Down Law in Consideration of Vehicle Type and Messages Displayed on Upstream Dynamic Message Signs
Nusayba Megat-Johari, Michigan State UniversityShow Abstract
Megat-Usamah Megat-Johari, Michigan State University
Peter Savolainen, Michigan State University
Timothy Gates, Michigan State University
Eva Kassens-Noor, Michigan State University
Move-over laws are intended to enhance the safety of road agency and law enforcement personnel who are working on or near the roadway. This study examines driver behavior through a series of field studies where these types of vehicles are located on the outside shoulder of a freeway with their lights activated. The study also evaluates the use of upstream dynamic message sign (DMS) to discern whether targeted safety messages have any impact on behavior under this scenario. Upstream and downstream speed and lane position data are collected from vehicles originally traveling in the rightmost lane upstream of the DMS and emergency/service vehicle at two locations in Michigan. Logistic regression models are estimated to assess driver compliance with the law while considering important contextual factors, such as the type of vehicle on the shoulder and the message displayed on the DMS. The results indicate that drivers were more likely to move over or reduce their speeds when a police car was located on the shoulder as compared to a transportation agency pickup truck. In general, the type of message displayed had minimal impact on driver behavior. The one exception showed that drivers were likely to drive at or below the speed limit when targeted move over messages were shown as compared to standard travel time messages. For all message types, both speed and lane compliance were improved if the roadside vehicle was a police car.
Examining Trends in Traffic Crashes as They Relate to the Display of Safety Messages on Dynamic Message Signs
Megat-Usamah Megat-Johari, Michigan State UniversityShow Abstract
Nusayba Megat-Johari, Michigan State University
Peter Savolainen, Michigan State University
Timothy Gates, Michigan State University
Eva Kassens-Noor, Michigan State University
Transportation agencies have increasingly been using dynamic message signs (DMS) to communicate safety messages in an effort to both increase awareness of important safety issues and to impact driver behavior. Despite their widespread use, evaluations as to potential impacts on driver behavior, and the resultant impacts on traffic crashes, has been very limited. This study addresses this gap in the extant literature and assesses the relationship between traffic crashes and the frequency with which various types of safety messages are displayed. Safety message data were collected from a total of 202 DMS across the state of Michigan between 2014 and 2018. These data were integrated with traffic volume, roadway geometry, and crash data for segments that were located downstream, as well as at several locations upstream of each DMS. A series of negative binomial models were estimated to examine total, speeding-related, and nighttime crashes based upon historical messaging data while controlling for other site-specific factors. However, the results of this evaluation did not show any meaningful differences in safety performance based on message delivery. While crashes declined marginally when higher frequency messaging was utilized, none of these differences were statistically significant. These findings are in contrast to stated preference surveys, which suggest drivers would be more likely to adapt their behavior to such messaging strategies. Important issues are also highlighted with respect to methodological concerns that arise in the analysis of such data. Field research is warranted to investigate potential impacts on driving behavior at the level of individual drivers.
Light Delivery Vehicles Crashes: Identifying Insights using Joint Dimension Reduction and Clustering
Subasish Das, Texas A&M UniversityShow Abstract
Anandi Dutta, University of Texas, San Antonio
In the era of food delivery and grocery delivery startups, traffic crashes associated with light delivery vehicles have increased significantly. Because of the increasing number of these crashes, it is important to investigate light vehicle crashes to gain insights about potential contributing factors. This study collected seven years (2010-2016) of data from traffic crash narrative reports and structured traffic crash data from Louisiana. By using text search options and manual exploration, a database of 1623 light delivery related crashes have been examined by using a comparatively robust clustering method known as cluster correspondence analysis. The findings identified six clusters with specific traits. The key clusters are interstate related crashes due to inattention, fatigue, alcohol impairment, or a young driver on low to mode speed roadways. The findings of the current study can be used by the policy makers to perform data-driven policy development in a way to ensure safety for delivery related travels.
New Intersection Crash Prediction Models for the Second Edition of the Highway Safety Manual
Darren Torbic, Texas A&M Transportation InstituteShow Abstract
Daniel Cook, HDR
Joseph Grotheer, MRIGlobal
Richard Porter, VHB
Jeff Gooch, VHB
Kristin Kersavage, VHB
The objective of this research was to develop new intersection crash prediction models for consideration in the second edition of the Highway Safety Manual (HSM), consistent with existing methods in HSM Part C and comprehensive in their ability to address a wide range of intersection configurations and traffic control types in rural and urban areas. The focus of the research was on developing safety performance functions (SPFs) for intersection configurations and traffic control types not currently addressed in the HSM Part C. SPFs were developed for the following general intersection configurations and traffic control types:Rural and urban all-way stop-controlled intersections Rural three-leg intersections with signal control Intersections on high-speed urban and suburban arterials (i.e., arterials with speed limits greater than or equal to 50 mph) Urban five-leg intersections with signal control Three-leg intersections where the through movements make turning maneuvers at the intersections Crossroad ramp terminals at single-point diamond interchanges Crossroad ramp terminals at tight diamond interchanges Development of severity distribution functions (SDFs) for use in combination with the SPFs to estimate crash severity as a function of geometric design elements and traffic control features was explored; but due to challenges and inconsistencies in developing and interpreting the SDFs, it was recommended for the second edition of the HSM that crash severity for the new intersection configurations and traffic control types be addressed in a manner consistent with existing methods in Chapters 10, 11, and 12 of the first edition of the HSM, without use of SDFs.
Development Of Safety Performance Functions For Two-Lane Rural Highways In The State Of Ohio
Abdulrahman Faden, University of DaytonShow Abstract
Deogratias Eustace, University of Dayton
The Highway Safety Manual (HSM), which is the guidance document for state departments of transportation (DOTs), was published in 2010, and one of its sections, called Part C of HSM, it involves the development of crash prediction models. However, HSM's default prediction models are not suitable for all states or jurisdictions because each state and jurisdiction have different characteristics, such as terrain, driver behaviors, weather conditions, etc. Hence, the principal objective of this study is to develop a prediction method for producing Ohio-specific SPF models to use for rural two-lane highways in the state of Ohio. Highway geometric data for almost 40,067 segments that have 21,666.03 miles and 79,481 total crashes that occurred for 4 consecutive years (2012-2015) were obtained from the Highway Safety Information System (HSIS) to create these new models using negative binomial regression and the pruned forward selection method by adding the interaction terms via JMP Pro software. The most critical variables used for analyzing and creating the best models for the state of Ohio are average annual daily traffic (AADT), segment length, lane width, shoulder width, posted speed limit, presence of curves and grades, which were proven to be statistically significant in developing SPFs. Besides, the standard goodness-of-fit parameters were chosen to evaluate the regression models was AIC. Two models were created for rural two-lane road segments in the state of Ohio, which can be used to predict all crash types and fatal and injury crashes.
Calibration and Development of Safety Performance Functions for Rural Two-Lane Two Way Roadways: A New Jersey Case Study
Abhinav Bhattacharyya, New York UniversityShow Abstract
Bekir Bartin, Ozyegin Universitesi
Kaan Ozbay, New York University
Chuan Xu, Southwest Jiaotong University
Hani Nassif, Rutgers University
Calibrating the safety performance functions (SPF) in the Highway Safety Manual (HSM) and developing jurisdiction-specific SPFs both require significant time, effort and resources, and detailed data from different sources. It is therefore crucial to identify all the readily available data sources and automatically gather much of the data required by the HSM. However, datasets maintained by the state transportation agencies are rarely comprehensive enough to meet all data requirements, often including errors and inconsistencies. This paper presents a detailed discussion of data needs and availability, data processing methods and approaches to gather the required data for calibration and development of SPFs using rural two-way two-lane rural roadway segments and intersections in New Jersey (NJ) as a case study. It is shown that generating a usable dataset from various different data sources is a rigorous task of data compiling, cleaning and processing, and requires a significant computer programming effort. While presenting the results of the SPF calibration and development process, this paper points to the importance of crash location information and its impact on analyses results. In addition, through past literature and best practices, this paper also discusses the practicality of the current manual data extraction practices, and argues that novel data extraction methods, such as the clustering approach used in this study, should be adopted to minimize labor-intensive and cost-prohibitive manual data collection processes and increase data accuracy. The choice between calibration and development of jurisdiction-specific SPFs is also discussed.
How did COVID-19 affect crash patterns in Florida? A GIS-based Spatiotemporal Investigation in Four Counties
Mohammadreza Koloushani (firstname.lastname@example.org), Florida A&M University-Florida State University College of EngineeringShow Abstract
Mahyar Ghorbanzadeh, Florida A&M University-Florida State University College of Engineering
Eren Ozguven, Florida A&M University-Florida State University College of Engineering
Alireza Ermagun, Mississippi State University
Mehmet Ulak, University of Twente
As a result of the COVID-19 pandemic, stay-at-home orders were issued in many states in the U.S. including the State of Florida. Due this order, many businesses have been closed or have started working virtually, and many other activities were cancelled. There is, consequently, a significant decrease in the number of trips, which might impact the pattern of crash density differently in counties with distinct characteristics. This study intends to investigate the impact of the COVID-19 on the spatiotemporal patterns of crash density in four demographically different counties in Florida: Escambia, Hillsborough, Leon, and Liberty. We propose a GIS-based method to examine whether the differences of crash density during the COVID-19 impacted dates are significantly and spatially significant than previous time periods. The statistical significance test results show that the mean crash density during the COVID-19 is statistically different from others. The Kernel Density Estimation (KDE)-based spatial analysis indicates that the crash density patterns vary from county to county based on demographic characteristics. The time series analyses also show that the curfew in the metropolitan area, namely the Hillsborough county, results in a higher number of crashes whereas we follow a more stable decreasing effect in other mid-size counties. The GIS-based results obtained from this study can help understand how changes in travel policies may affect traffic safety and inform policy-makers on safety outcomes of shifts in mobility patterns that are expected in the near future.
Accommodating for Systematic and Unobserved Heterogeneity in Panel Data:
Application to Macro-Level Crash Modeling
Tanmoy Bhowmik (tanmoy78@Knights.ucf.edu), University of Central FloridaShow Abstract
Shamsunnahar Yasmin, Queensland University of Technology
Naveen Eluru, University of Central Florida
The current research contributes to the burgeoning literature on multivariate models by proposing a hybrid model framework that (a) incorporates unobserved heterogeneity in a parsimonious framework and (b) allows for additional flexibility to accommodate for observed/systematic heterogeneity. Specifically, we estimate a Latent Segmentation Panel Multivariate Negative Binomial (LPMNB) to study the zonal level crash counts across different crash types. Further, we undertake a comparison exercise of the proposed hybrid LPMNB model with a Panel Mixed Negative Binomial model (PMNB) that accommodates for all unobserved heterogeneity via a simulation setting. The analysis is conducted using the zonal level crash records by different crash types from Central Florida region for the year 2016 considering a comprehensive set of exogenous variables. Based on the statistical data fit, we find that the segmented model (LPMNB) is a preferred choice as long as the framework is estimated in a closed form system. The comparison exercise is further augmented by computing several goodness of fit measures and the results offered by the LPMNB model highlight the value of the proposed model.
Developing an Index-based Methodology to Assess the Quality of SPF Calibration. A Multivariate Approach
Raul Avelar, Texas A&M Transportation InstituteShow Abstract
Karen Dixon, Texas A&M University
Boniphace Kutela, Texas A&M Transportation Institute
Samuel Klump, HDR
Wemple Elizabeth, HDR
Richard Storm, HDR
The calibration of Safety Performance Functions (SPFs) is a mechanism included in the Highway Safety Manual (HSM) for practitioners to adjust SPFs in the HSM for use in their respective jurisdictions. Critically, the quality of the calibration procedure must be assessed prior to using the calibrated SPFs. Multiple resources to aid practitioners in calibrating SPFs have been developed in the years following the publication of the HSM first edition. Similarly, the literature suggests multiple ways to assess the goodness of fit (GOF) of a calibrated SPF to a dataset from a given jurisdiction. This paper uses the calibration of multiple intersection SPFs to a large Mississippi safety database to examine the relations between GOF metrics. The goal is to develop a sensible single index that leverages the joint information from multiple GOF metrics. A factor analysis applied to the calibration results revealed three underlying factors that explain 76 percent of the variance of the GOF metrics. From the factor analysis results, the authors developed an index and performed a sensitivity analysis. The key metrics explaining index variation were found to be, in descending order: the deviation of the cumulative residual (CURE) plot from the 95 percent confidence area, the mean absolute deviation, the modified R-squared, and the value of the calibration factor. This paper also presents comparisons between the index and alternative scoring strategies, as well as an effort to verify the results using synthetic data. The developed index is recommended to assess the quality of the calibrated intersection SPFs.
Estimating Safety Effects of Adaptive Signal Control Technology Using the Full Bayesian Approach
John Kodi, Florida International UniversityShow Abstract
Angela Kitali, Florida International University
MD Sultan Ali, Florida International University
Priyanka Alluri, Florida International University
Thobias Sando, University of North Florida
Adaptive Signal Control Technology (ASCT) is a traffic management strategy that optimizes signal timing based on real-time traffic demand. Although the primary intent of ASCT is to improve the operational performance of signalized intersections, the technology may also have substantial safety benefits. This study explored the potential safety benefits of the ASCT strategy deployed at signalized intersections in Florida. An observational before-after full Bayes (FB) approach with a comparison-group was adopted to develop crash modification factors (CMFs) for total crashes, rear-end crashes, and specific crash severity levels (fatal plus injury (FI), and property damage only (PDO) crashes). The analysis was based on 20 intersections equipped with ASCT and their corresponding 40 comparison intersections without ASCT. The ASCT deployment was found to significantly reduce total crashes by 8.5% (CMF = 0.915), rear-end crashes by 8.5% (CMF = 0.915), and PDO crashes by 8.1% (CMF = 0.919). The 8.7% reduction in FI crashes (CMF = 0.913) was not significant at a 90% Bayesian credible interval. These findings provide researchers and practitioners with an effective means to quantify the safety benefits of the ASCT strategy and conduct economic appraisals of ASCT deployments.
Developing Short-Term Safety Performance Functions for Freeways at Different Aggregation Levels by Using Multi-State Microscopic Traffic Detector Data
Jinghui Yuan, University of Central FloridaShow Abstract
Mohamed Abdel-Aty, University of Central Florida
Jingwan Fu, University of Central Florida
Yina Wu, University of Central Florida
Lishengsa Yue, University of Central Florida
Naveen Eluru, University of Central Florida
Safety Performance Functions (SPFs) have been widely used by researchers and practitioners to conduct roadway safety evaluation. Traditional SPFs are usually developed by using annual average daily traffic (AADT) along with geometric characteristics. However, the high level of aggregation may lead to a failure of capturing the temporal variation in traffic volume, speed, weather, crashes, and other factors. In this study, short-term SPFs at different aggregation levels were developed based on microscopic traffic detector data from California, Florida, and Virginia. Five aggregation levels were considered: (1) annual average weekday hourly traffic (AAWDHT), (2) annual average weekend hourly traffic (AAWEHT), (3) annual average weekday peak/off-peak traffic (AAWDPT), (4) annual average day of the week traffic (AADOWT), and (5) annual average daily traffic (AADT). Model estimation results showed that the segment length and volume are significant across all the aggregation levels. Average speed is significant with a negative coefficient, and the standard deviation of speed was found to be positively associated with the crash frequency. It is noteworthy that the HOV operation status was found to have a positive effect on crash frequency across all the aggregation levels. The model comparison results in prediction performance showed that the AADOWT and AAWDPT models consistently performed slightly better than the other models, which implies that the differences between the day of the week and peak/off-peak periods should be considered in the development of crash prediction models.
MODELING THE EFFECTS OF DRIVE ERROR AND IMPAIRMENT ON CRASH INJURY SEVERITY
Mohammad Razaur Rahman Shaon (email@example.com), University of ConnecticutShow Abstract
Xiao Qin, University of Wisconsin, Milwaukee
Driver errors are widely cited as one of the critical reasons for crash occurrence in safety literature. Despite universal acceptance, the discussion of their effects on crash outcomes is limited. The primary objective of this study is to quantify the effects of driver errors in the crash injury severity model at urban intersections. To obtain research objectives, driver errors were categorized as sequential events in a driving task. Combinations of driver error categories were created and ranked based on their odds-ratios with injury severity levels. Furthermore, driver impairment was considered in the model to explore the compounding effects on crash consequences. Multiple ordered logit models were estimated to quantify the effect of driver errors and their interactions with driver impairment on crash injury at uncontrolled, sign-controlled, and signal-controlled intersections. Improved model performance was observed when driver error combinations were modeled along with typical crash variables. The exploration of multiple model formulations indicated that including driver impairment as an error category can yield informative inferences from both theoretical and modeling perspectives. As a result, appropriate countermeasures were recommended for major contributing factors to improve intersection safety. It is expected that this study can offer specific insights into explanatory variables and help safety professionals to develop effective countermeasures.
A Before-and-After Evaluation of Traditional Traffic Control Devices for Preventing Wrong-Way Driving at Freeway Off-Ramps
Qing Chang (firstname.lastname@example.org), Auburn UniversityShow Abstract
Md Atiquzzaman, Johnson Mirmiran and Thompson
Huaguo Zhou, Auburn University
Yukun Song, Auburn University
This paper presents a before-and-after study of the effectiveness of traditional traffic control devices (TCDs) on preventing Wrong-Way Driving (WWD) at two partial cloverleaf (parclo) off-ramp terminals in Alabama. These two locations were selected because (1) they were identified as the high-risk locations for WWD; and (2) the traditional TCDs have been improved to mitigate the WWD activities. WWD incident data was collected from more than 800 hours of video surveillance before-and-after the countermeasures implementations at each location. At I-65 Exit 284 Southbound (SB) off-ramp terminal, the pavement marking was improved, including (1) repainted double yellow line and newly painted left-turn skip strips, and (2) yield line for off-ramp right-turn lane and stop bar for the left-turn lane at the end of the exit ramp. At I-65 Exit 208 SB off-ramp terminal, a raised-curb channelizing island was implemented as the first stage of improvement. At the second stage, the additional signages were installed on the channelizing island, and the double yellow line on the crossroad was repainted. For I-65 Exit 284 SB ramp, the improvements reduced 60% of the total and approximately 75% of nighttime WWD incidents. For I-65 Exit 208 SB ramp, the channelizing island alone implemented at the first stage resulted in an approximately 80% increase in WWD incidents, however the number decreased by approximately 50% after the improvements at the second stage.
Localized Safety Performance Functions for Rural 3-Leg Stop-Controlled Intersections in Alabama
Zihe Zhang, University of AlabamaShow Abstract
Xiaobing Li, University of Alabama
Jun Liu, University of Alabama
Xing Fu, University of Alabama
Chenxuan Yang, University of Alabama
Steven Jones, University of Alabama
Safety Performance Functions (SPFs) are often used to predict the expected crash frequency for a proposed highway facility (intersection or road segment). Highway Safety Manual (HSM) provides a set of SPFs for practitioners to use and also recommends the development of jurisdiction-specific SPFs using local safety data. Because only limited variables are available in the data, SPF models need to handle the potential unobserved heterogeneity in data. Random-parameter models are often applied to develop SPFs; but the model outcomes are still to represent the regionwide relationships between crash frequency and factors. Given the interactions between traffic safety and the environment (social, culture as well as geography), the unobserved heterogeneity is likely related to the space. This study employs a spatial modeling approach, namely Geographically Weighted Negative Binomial Regression (GWNBR), to incorporate the spatial heterogeneity into SPF model specification. In contrast with models that provide regionwide SPFs (e.g., a state), the GWNBR model can generate an SPF for local areas (part of the region), called Localized SPFs (or L-SPFs). This study uses the 2014 to 2018 geo-referenced crash data from Alabama to develop L-SPFs for rural 3-leg stop-controlled intersections. The results show the L-SPFs estimated by the GWNBR model vary substantially across Alabama. For example, the coefficients for traffic volume (AADT) range from 0.121 to 0.919 across different parts of the state. Practitioners and decision-makers could use the L-SPFs to predict crash frequency in their local area, so the countermeasures and funds could be better allocated to reflect the local situations.
Road Geometry and Pavement Surface Condition Impacts on Roadway Departure Crashes
Justice Appiah (Justice.Appiah@VDOT.Virginia.gov), Virginia Transportation Research CouncilShow Abstract
Mo Zhao, Virginia Transportation Research Council
Bhaven Naik, Ohio University
Roadway departure (RD) crashes are one of the major causes of fatalities on rural roads in Virginia. Thus, research to develop and enhance understanding of potential influencing factors continues to be of interest in mitigating these crashes. SCRIM, a truck-based multifunctional roadway monitoring device that can simultaneously and continuously collect roadway surface condition and geometry data while being driven in the speed range of 15–53 mph, has been widely used in European countries for nationwide road surveys. The Virginia Department of Transportation (VDOT) started to conduct SCRIM surveys in 2018. These SCRIM surveys comprise a new source of data for pavement skid resistance, texture, and roadway alignment––all features that are not routinely collected or archived as part of current practice, but have potential to influence RD crashes. This study explored SCRIM data collected on 86 miles of rural highways in Virginia for relationships between pavement condition, road geometry, and RD crash frequency. The analysis was supplemented with additional information including archived pavement management and crash records that were obtained from VDOT sources. Standard statistical methods were used to investigate the relationship between RD crash frequency and likely influencing variables identified in the composite dataset. It was found that this relatively new data source holds promise for further insights into factors influencing RD crashes. In particular, the study found statistically significant association between RD crash frequency and geometric characteristics such as cross slope and curvature. A significant association was also found between crash frequency and pavement skid resistance.
Investigating Changes in Florida Traffic Crash Trends Due to COVID-19 Pandemic
Roodson Pierre, Florida A&M UniversityShow Abstract
Doreen Kobelo, Florida A&M University
Ntagwanko Kisabanzira, University of North Florida
John Kodi, Florida International University
Angela Kitali, Florida International University
Thobias Sando, University of North Florida
The novel COVID-19 pandemic has caused dramatic changes in almost everybody’s life in the world. Transportation safety is one of the sectors that experienced dramatic changes due to due to stay-at-home orders that were issued across the United States and in other countries, to combat the widespread of the pandemic. Variations in traffic pattern, travel behavior, and human behavior leads to significant changes in traffic safety. This study investigated the impacts of the COVID-19 pandemic on traffic crashes in Florida’s freeway (I-10, I-75, and I-95). Traffic crashes for the months of March and April for the years 2018 through 2020 along the selected corridors were analyzed to explore the impacts of the novel pandemic in crash safety. The analysis showed that since the first confirmed case in Florida, there was a decreasing trend in the total traffic crash frequency in the selected study corridor. Compared to the similar period in 2018 and 2019, the overall traffic crashes dropped significantly by up to 45.3% following the restrictions imposed to slow down the spread of the virus. Also, a decrease in the rear-end crashes and an increase in the run-off-road were observed. This study helps to understand the early impacts of the pandemic and may be useful in operational and strategic planning for future pandemics.
Investigating the Relationship between Vehicle on Shoulder and Crash: Correlation or Causation
Eugene Antwi Boasiako (email@example.com), Kentucky Transportation CabinetShow Abstract
Mei Chen, Kentucky Transportation Center
Xu Zhang, Kentucky Transportation Cabinet
Benjamin Blandford, Kentucky Transportation Center
Technological advancements have afforded road users the ability to contribute to traffic data. These road user generated data, termed crowdsourced traffic data, aid traffic managers to monitor roadway conditions and coordinate more effective incident management. Moreover, integrating crowdsourced traffic data with conventional traffic data provides additional information for traffic safety analysis. In this study, we focus our analysis on vehicles on the shoulder, using Waze vehicle-on-shoulder alerts, and their impact on safety of limited access highways in Kentucky. The analysis showed that about 36% of crashes had vehicle on shoulder present in their vicinity, defined as 0.25 miles upstream and downstream of a crash site and 30 minutes before crash occurrence. While congestion was associated with about 25% of crashes, 11.7% of crashes were attributable to both congestion and vehicles on the shoulder. As such, a statistically significant association was found between vehicles-on-shoulder, congestion and crashes. Based on crash narrative review, 1.8% of all crashes directly involved vehicles on shoulder and 23% of the crashes cited congestion as a contributor. However, there is little indication in the crash narratives on how vehicles on shoulder contributed to crashes, beyond their direct involvement, or how they contributed to congestion which may led to crashes. Though a small fraction of crashes were attributed to vehicles on shoulder, these crashes tended to be more severe on average than all interstate crashes. Data used in this study, and the analytical methods proposed, offer much-needed insights into the challenges posed by vehicles on roadway shoulders.
Development of Safety Performance Functions for Low Volume Rural Local Roads in Alabama
Samantha Islam (firstname.lastname@example.org), University of South AlabamaShow Abstract
Richard DZINYELA, University of South Alabama
Grace Toledo, University of South Alabama
The objective of this study was to develop safety performance functions (SPF's) for rural local roads in Alabama with low traffic volume (AADT<2000 vehicles per day) for three types of facilities: two-lane undivided roadway segments (LR2U), three-leg stop controlled intersections (3ST), and four-leg stop controlled intersections (4ST). The dataset used for the development of SPF’s was compiled from several sources and included data related to traffic crashes, volumes, roadway classification, geometry, cross-sectional features and roadway characteristics from all 67 counties throughout the state from 2012 to 2014. The total length of roadway segments that was used for SPF development was 1,036 miles and corresponding number of crashes were found to be 1,929 crashes. Similarly, the compiled dataset included 622 three-leg intersections (3ST) with 825 crashes and 325 four-leg intersection (4ST) with 539 crashes. Separate SPFs were developed for roadway segments and stop-controlled intersections for total (KABCO), fatal plus injury (KABC), and property damage only (PDO) crashes for the AADT ranges of 0-2000, 0-399, 400-1599, and 1600-2000 vehicles per day. Overall, it was found that the SPFs fit the data well. However, due to small number of sample sizes for three- and four-legged intersections, variables other than AADT’s for major and minor roads were not found to have significant impacts on crash frequencies. At the end of this paper, simple step-by-step examples have been provided for the users to familiarize them with the network screening process using these SPFs.
Development of Road Diet Segment and Intersection Crash Modification Factors
Linda Lim, University of VirginiaShow Abstract
Michael Fontaine, Virginia Transportation Research Council
Road diets can offer potential safety improvements for both pedestrians and vehicles. The additional space provided by reducing the number of vehicular through-lanes can be reallocated into other uses such as bicycle lanes, parking, sidewalks, transit use, turn lanes, curb extensions, parklets, or pedestrian refuge islands. This study evaluated the safety effectiveness of road diets in Virginia using the Empirical Bayes (EB) method, focusing on the common road diet conversion from a four-lane roadway to a three-lane roadway with added bike lanes. A total of 37 segment sites and 39 intersections were identified in Virginia where road diet installations were implemented between the years 2009 to 2018. The analysis showed segment crash modification factors (CMFs) of 0.65 for total crashes and 0.41 for fatal and injury (FI) crashes. Across all intersection types, the CMFs were 0.61 for total crashes and 0.59 for FI crashes. All CMFs were found to be statistically significant at a 95% confidence level. When intersections where separated into signalized and unsignalized intersections, no significant safety benefit was found for unsignalized intersections, however. Based on the results, it is concluded that road diets can potentially reduce crashes and public agencies should consider the safety benefits of road diets when justifying roadway improvements.
Impact of Injury-Based Safety Performance Functions on Network Screening
Praveen Vayalamkuzhi, University of California, BerkeleyShow Abstract
Aditya Medury, Indian Institute of Technology, Kanpur
Lin Yang, University of California, Berkeley
Offer Grembek, University of California, Berkeley
Venky Shankar, Texas Tech University
Safety Performance Functions (SPF) are proven to be the best resource for evaluating highway safety. However, agencies have a difficult time selecting the type of SPF from among the various options to use in analysis—whether all crashes will be included, or determining the impact of Property Damage Only (PDO) crashes in the safety analysis. Injury severities considered for network screening have important implications for how limited agency resources are used. This operational reality that dominates many agencies can inhibit their ability to systematically choose the appropriate SPF type for prediction and implementation. This paper describes the application and outcome of different crash-severity level SPFs to address such concerns through a set of SPFs—total crashes, fatal and severe crashes, and a combination of fatal, severe and visible injury crashes. Potential for Safety Improvement (PSI) based on the Empirical Bayes (EB) approach was then used for network screening to select the top one percent of hotspots within each facility type by injury severity level. The findings indicate differences in site characteristics across hotspots identified using different crash types, with total crashes favoring sites in heavier traffic urban areas. There is a greater need to understand the type of sites that yield high property damage only (PDO) crashes within network screening—for example, there is mixed evidence about the over-representation of PDO crashes among secondary crashes on highways. The analysis also revealed the differences in site demographic type and Annual Average Daily Traffic (AADT) across hotspots identified by different crash combinations.
Assessing the Predictability of Short Segment Crash Analysis in the State of South Carolina
Adika Iqbal, Clemson UniversityShow Abstract
Wayne Sarasua, Clemson University
Afshin Famili, Texas Department of Transportation
Jennifer Ogle, Clemson University
William Davis, Citadel Military College
Devesh Kumar, Clemson University
Saurabh Basnet, Clemson University
Emmanuel Adjei, Clemson University
The main objective of this research is to evaluate the predictability of a short segment peak search method with lengths of less than 0.1 miles for the statewide screening of midblock crash locations. Three different approaches (Based on HSM SPFs) are used to evaluate the short segment method. These approaches include state-specific SPFs, Driveway SPFs (using only AADT), and driveway SPFs with adjusted CMFs. Frequency-based identification of short segments stratified by six different roadway types (R2U, R4D, U2U, U4D, U3T, and U5T) has been compared with three SPF based screening methods to determine segments with the highest excess predicted average crash frequency. For short segment sites with highest crash frequencies (3 for U3T, U4D, and U2U; 4 for U5T and 2 for R4D and R2U), the comparison showed similar results (Top 90% agreement). Thus, should insufficient data be available to conclude SPFs, a frequency-based approach will likely identify the top sites. While this method works relatively well with top sites, the reliability of this method will wane with lower-ranked sites.
Evaluating the Safety Effectiveness of Restricted Crossing U-turn (RCUT) Intersections
Raunak Mishra, University of North Carolina, CharlotteShow Abstract
Srinivas Pulugurtha, University of North Carolina, Charlotte
The focus of this paper is on evaluating the safety effectiveness of restricted crossing U-turn (RCUT) intersections. Both unsignalized and signalized RCUT intersections were evaluated using the Empirical Bayes (EB) before-after evaluation method. The forty-two RCUT intersections considered in this research were converted from a two-way stop-controlled (TWSC) intersection or signalized intersection in the rural and suburban areas. The results show a 73.27% reduction in the total number of crashes and a 79.42% reduction in the number of fatal and injury crashes at unsignalized stop-controlled RCUTs in the rural area. In the suburban areas, a 64.86% reduction in the total number of crashes and a 73.39% reduction in the number of fatal and injury crashes was observed. Further, a 10.15% and a 31.08% reduction in the total number of crashes, and an 84.26% and 31.13% reduction in the number of fatal and injury crashes was observed at signalized RCUTs in the rural and suburban areas, respectively. Overall, the unsignalized RCUTs in the rural areas with a larger sample size were found to be more safer than was observed by researchers in the past. These findings are useful to researchers and practitioners for making informed decisions and implementing RCUTs from a safety perspective.
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.