Search

• Paper estimates a CMF for high friction surface treatment (HFST) applied to horizontal curves on undivided two-lane roads. • Propensity score matching is incorporated into EB before-after methodology to select more appropriate reference group. • Results indicate significant reductions in crash frequency (of all types considered) when applying HFST. Horizontal curves are locations that, as a result of the changing alignment, may be a contributing factor in roadway departure crashes. One low-cost countermeasure to mitigate crashes at these locations is the installation of the high friction surface treatment (HFST), which increases roadway friction and is intended to help keep drivers on the roadway when traversing a horizontal curve. This treatment has been implemented at numerous curves in Pennsylvania, but the overall safety effectiveness is not known. The purpose of this study is to estimate a suite of Crash Modification Factors (CMFs) for HFST applied to curve sections of undivided two-lane roadways. A novel combination of the empirical Bayes observational before-after study design and propensity score matching was used to estimate CMFs for multiple crash types, crash severities, and roadway settings (urban and rural). Propensity score matching was implemented to identify the most appropriate reference group to use within the empirical Bayes methodology. The results indicate that the installation of HFST is associated with a statistically significant decrease in all crash types and severities considered. [ABSTRACT FROM AUTHOR]

Past design consistency research has demonstrated the relationship between operating speeds and geometric design features on two-lane rural highways. However, little is known about the relationship between geometric design consistency and traffic safety. In this study, design consistency is referred to as the difference between operating speed and inferred design speed, and design consistency density is measured to account for the effect of elements upstream and downstream of the study element. To perform the design consistency-safety evaluation in the present study, geometric design, roadway inventory, crash, and operating speed data were collected along two case-study highways in central Pennsylvania (U.S. 322 and PA 350). Several count regression model formulations were used to explore the statistical association between design consistency and total crash frequency. A statistically significant positive association between geometric design consistency and safety was found. Design consistency surrounding the study elements was also found to increase the expected crash frequency in the study element. The significant effects of design consistency and design consistency density measures on crash frequency for the two case-study sites indicate that there is value in further considering the relationship between design consistency and safety in future research, particularly in the context of the highway safety manual crash-prediction algorithm. [ABSTRACT FROM AUTHOR]

Abstract: Introduction: Crossover and rollover crashes in earth-divided, traversable medians on rural divided highways can lead to severe injury outcomes. This study estimated severity models of these two crash types. Vehicle, driver, roadway, and median cross-section design data were factors considered in the models. A unique aspect of the data used to estimate the models were the availability of median cross-slope data, which are not commonly included in roadway inventory data files. Methods: A binary logit model of cross-median crash severity and a multinomial logit model of rollover crash severity were estimated using five years of data from rural divided highways in Pennsylvania. Results: The highest probability of a fatal or major injury in cross-median and rollover crashes was found to occur in cases when a driver was not wearing a seatbelt. While flatter cross-slopes and narrower medians were associated with more severe cross-median crash outcomes, steeper cross-slopes and narrower medians significantly increased rollover crash severity outcomes. The presence of horizontal curves was associated with increased probabilities of high-severity outcomes in a median rollover crash. Impact on Industry: Modeling results in this study confirmed that cross-median and median rollover crash severity outcomes are associated with median cross-section design characteristics. Based on the estimated models, it appears that flatter and narrower medians lead to more severe injury outcomes in cross-median crashes. Steeper median cross-slopes and narrower medians were associated with higher probabilities of more severe outcomes in median rollover crashes. The results presented in this study suggest that there is a trade-off between median cross-section design and cross-median and rollover crashes in earth-divided, traversable medians on rural divided highways. While the severity models can be included in a framework to develop design guidance in relation to this trade-off, models of crash frequency should also be considered. [Copyright &y& Elsevier]

Two-lane rural highways attract trips between traffic generators such as residential villages or small commercial areas and are generally uninterrupted flow facilities that provide a high level of mobility with high posted speeds limits. When passing through developed areas, the posted speed limit may be lowered to encourage reduced vehicle operating speeds. No published guidelines currently exist for the design of transition zones that connect high- to low-speed operating environments on two-lane rural highways. The objective of this research was to collect operating speed and roadway characteristic data along two-lane rural highway transition zones in Pennsylvania to explore the roadway, roadside, and traffic control factors that are associated with driver speed differentials. Single- and multilevel models were estimated and compared. The posted speed limit reduction as well as a change in the paved shoulder width, total number of driveways, various advance warning signs, the transition zone length, and the presence of horizontal curves were shown to increase the expected speed reduction in transition zones in the models. Additionally, drivers entering transition zones at higher speeds were found to have greater speed reductions than drivers entering the transition zone at lower speeds. [ABSTRACT FROM AUTHOR]

The Pennsylvania Department of Transportation collected periodic pavement marking retroreflectivity data on 88 roadway segments over a 1-year period from May 2007 through May 2008. The purpose of this effort was to use the data collected to develop a series of pavement marking degradation models that could be used to implement a management system. In this study, panel data models were used to estimate the service life of both waterborne and epoxy pavement markings because of the repeated retroreflectivity measurements recorded over time. It was found that pavement marking retroreflectivity decreases as the age of both epoxy and waterbornepaint pavement markings increase. White pavement markings were shown to have higher estimated service lives than yellow pavement markings. Traffic exposure was found to be negatively correlated with pavement marking retroreflectivity in the waterborne-paint analysis, but traffic exposure was not statistically significant in the epoxy pavement marking retroreflectivity model. A life cycle cost analysis showed that waterborne paints are more cost-effective than epoxy pavement markings if applied on a statewide basis. [ABSTRACT FROM AUTHOR]

The interactive highway safety design model (IHSDM) is a suite of software analysis tools used to evaluate safety and operational effects of geometric design on two-lane rural highways. IHSDM checks existing or proposed two-lane highway designs against relevant design policy values and provides estimates of a design’s expected safety and operational performance. This paper describes efforts to apply IHSDM to two existing rural roadways (one an arterial and the other a collector) in Pennsylvania and to compare the crash prediction and speed profile results to observed speed and crash data for these same two roadway segments. When considering both case study roadways together, the use of site-specific crash history data, adjusted using the empirical Bayes procedure, and a county-level calibration factor, produced crash frequency predictions that most closely matched the observed crash frequencies. The design consistency module output on the arterial roadway closely matched the observed speeds. On the collector roadway, the predicted and observed speeds did not match closely, thus, underscoring the need to expand the design consistency module capabilities to consider low-speed, complex alignments. [ABSTRACT FROM AUTHOR]

Data from a controlled nighttime driving experiment along a complex, two-lane rural highway in Pennsylvania were used to investigate participant speed and lateral vehicle position. Additionally, the association of pavement marking visibility and the roadway geometry on speed and lateral vehicle position was investigated. The modeling methodology employed considered a variety of single equation and systems methods to estimate models for speed and lateral vehicle position. These include ordinary least-squares regression models, random effects panel data models, a three-stage least-squares technique, and seemingly unrelated regression. These methods were used to determine if there was an endogenous relationship between the speed and lateral vehicle position metrics, and to account for the repeat observations on the research participants in the experiment. The results suggest that endogeneity did not exist in the speed and lateral vehicle position measures in the present experiment. A single-equation random effects panel regression model was an improvement over the ordinary least-squares regression model for the change in speed metric. However, the ordinary least-squares regression model was appropriate for the lateral vehicle position metric. A seemingly unrelated regression model was also specified to account for contemporaneous correlation across the change in speed and lateral vehicle position models. It was found that the roadside hazard rating, horizontal curve direction, vertical grade, horizontal curve radius, upstream horizontal curve radius, and approach tangent length all were associated with vehicle speed on the complex, two-lane rural highway. The horizontal curve direction, roadside hazard rating, vertical grade, horizontal curve radius, and approach tangent length all were associated with lateral vehicle position in the present experiment. [ABSTRACT FROM AUTHOR]

Abstract: Median barrier warrant criteria were developed in the 1970s and generally remain unchanged today. Vehicle travel, including both traffic volumes and operating speeds, have increased over this same time period. Encroachments into the median, and subsequent collisions with vehicles traveling in the opposite travel lanes, result in high severity crashes. Median barrier is typically used to prevent cross-median crashes; median barrier selection is based on median width and traffic volumes. Quantifiable information regarding the effects of median barrier installation and its placement on crash frequency is limited. This paper investigates median barrier crash frequency on Pennsylvania Interstate highways, including separate models for the Turnpike and all other Interstate-designated highways. Negative binomial regression models were used to develop predictive crash frequency tools. Traffic volume, horizontal alignment, interchange ramp presence, and median barrier offset distance from the travel lanes were used to estimate median barrier crash frequency. The analytical methodology developed in this research can be used, in concert with other prediction models, to assess the consequences of median barrier placement decisions. [Copyright &y& Elsevier]

• New SPF functional form proposed that allows traffic volume to have different elasticities in different ranges. • Proposed form allows for continuous relationship between predicted crash frequency and traffic volume. • Proposed models shown to better fit observed data that traditional and other forms. Statistical models of expected crash frequency are referred to as Safety Performance Functions (SPFs) in the first edition of the American Association of State Highway and Transportation Officials' Highway Safety Manual (HSM). The SPFs in the HSM specify expected annual crash frequencies as a function of various roadway and roadside features, with the most important predictor variable being traffic volume, which serves as a measure of vehicle exposure to crashes. Traffic volumes are typically measured using the average annual daily traffic and are incorporated into the SPFs using a natural logarithm transformation. This specification suggests that the relationship between expected crash frequency and traffic volume increases non-linearly with a constant elasticity over the range of observed values. While researchers concur that the relationship between expected crash frequencies and traffic volume is non-linear, further exploration of the functional form of this relationship may offer additional insights concerning the association between safety performance and vehicle exposure. This paper proposes an alternative functional form for the traffic volume variable in SPFs that allows for different elasticities between traffic volume and expected crash frequency within different traffic volume ranges, while preserving the same general non-linear relationship in existing HSM SPFs. Although other forms—like the Hoerl function—have been proposed in the literature, the proposed model allows for natural breakpoints in the traffic volume for which roadway or geometric features might have varying effects on low- or high-volume roads. The proposed functional form was applied to SPFs developed for two-lane rural roadways in Pennsylvania. Comparisons with SPFs developed using the traditional and Hoerl functional forms suggest that this proposed functional form offers an improved fit and predictive performance, and thus might be considered for the development of future SPFs. [ABSTRACT FROM AUTHOR]

• Horizontal Curve Warning Pavement Markings in Pennsylvania were Evaluated. • Empirical Bayes Before-After Analysis was Used. • Multiple crash types and severities along horizontal curves were considered. • Findings indicate that the pavement markings reduce run-off road crashes by 23–35 %. Roadway departure crashes contribute to a large proportion of fatal and injury crashes in the United States. These crash types are more likely to occur along horizontal curve sections of a roadway. Countermeasures that prevent vehicles from departing the roadway is one method to mitigate roadway departure crashes. Pennsylvania has deployed on-pavement horizontal curve warning markings in advance of horizontal curves on two-lane rural highways as a roadway departure crash reduction strategy. This study used an Empirical Bayes (EB) before-after study design to evaluate the safety effects of the horizontal curve warning pavement markings. A total of 263 treatment sites and more than 21,000 reference sites were included in the evaluation. Crash modification factors were developed for total, fatal plus injury, run-off-road, nighttime, nighttime run-off-road, and nighttime fatal plus injury crashes. The point estimates for each of these crashes ranged from 0.65 to 0.77 – the results were statistically significant for total and fatal plus injury crashes at the 95th-percentile confidence level. [ABSTRACT FROM AUTHOR]

• Model-based recursive partitioning (MOB) produces similar crash frequency model estimation compared to heterogeneity model. • MOB algorithm improves predictive power relative to traditional safety and heterogeneity models based on hold-out sample. • MOB is promising alternative to traditional count regression models to identify covariate subgroup patterns in safety data. • The posted speed limit and the presence of a passing zone were key splitting variables in the MOB algorithm. • Traffic volume and horizontal curve density were identified as random parameters in heterogeneity model. Count regression models have been applied widely in traffic safety research to estimate expected crash frequencies on road segments. Data mining algorithms, such as classification and regression trees, have recently been introduced into the field to overcome some of the assumptions associated with statistical models. However, these data-driven algorithms usually provide non-parametric output, making it difficult to draw statistical inference or to evaluate how independent variables are associated with expected crash frequencies. In this paper, the model-based recursive partitioning (MOB) algorithm is applied in a crash frequency application. The algorithm incorporates the concept of recursive partitioning data in tree models and develops user-defined statistical models as outputs. The objective of this paper is to explore the potential of the MOB algorithm as a methodological alternative to parametric modeling methods in crash frequency analysis. To accomplish the objective, a standard negative binomial (NB) regression model, a NB model developed using the MOB algorithm, adjusted NB models which incorporate variables identified by the MOB algorithm, and a random parameters NB model are compared using 8 years of data collected from two-lane rural highways in Pennsylvania. The models are compared in terms of data fitness, sign and magnitude of statistical association between the independent and dependent variables, and predictive power. The results show that the MOB-NB model yields better data fitness than other NB models, and provides similar performance to the RPNB model, suggesting that the MOB-NB model may be capturing unobserved heterogeneity by dividing the data into subgroups. The presence of a passing zone and posted speed limit are two covariates identified by the MOB algorithm that differentiate variable effects among subgroups. In addition, the MOB-NB model provides the highest prediction accuracy based on the training and test data sets, although the difference among models is small. The comparison results reveal that the MOB algorithm is a promising alternative to identify covariates, evaluate variable associations and instability, and make predictions in a crash frequency context. [ABSTRACT FROM AUTHOR]

The American Association of State Highway and Transportation Officials’ Highway Safety Manual (HSM) contains safety performance functions (SPFs) to predict annual crash frequencies for several roadway types. When applying these SPFs in a jurisdiction whose data were not used to develop the SPF, a calibration factor can be applied to adjust the expected crash frequency estimate to statewide or local conditions. Alternatively, the HSM suggests that transportation agencies may develop their own SPFs in lieu of applying the calibration factor to the HSM SPFs. However, the HSM does not provide guidance on the appropriate level of regionalization that should be adopted for either method, even though safety performance may vary considerably within a state. In light of this, the present study considers the development of local or regionalized SPFs for two-lane rural highways within the Commonwealth of Pennsylvania. Three regionalization levels were considered: statewide, engineering district and individual counties. The expected crash frequency for each level of regionalization was compared to the reported crash frequency over an eight-year analysis period. The results indicate that district-level SPFs with county-level adjustment factors provide better predictive accuracy than the development of a statewide SPF or application of the HSM-calibrated SPF. The findings suggest that there are significant differences in safety performance across engineering districts within Pennsylvania. As such, other state transportation agencies developing SPFs or using calibration factors may also consider how variations across jurisdictions will affect predicted crash frequencies. [ABSTRACT FROM AUTHOR]

• Simultaneous estimation of crash frequency by either type or severity considered. • Univariate and bivariate (without and with copulas) negative binomial regression models compared. • Bivariate models with copulas shown to outperform traditional bivariate and univariate models. • Bivariate model with normal copulas outperforms other copulas tested. Statistical models of crash frequency typically apply univariate regression models to estimate total crash frequency or crash counts by various categories. However, a possible correlation between the dependent variables or unobserved variables associated with the dependent variables is not considered when univariate models are used to estimate categorized crash counts—such as different severity levels or numbers of vehicles involved. This may lead to inefficient parameter estimates compared to multivariate models that directly consider these correlations. This paper compares the results obtained from univariate negative binomial regression models of property-damage only (PDO) and fatal plus injury (FI) crash frequencies to models using traditional bivariate and copula-based bivariate negative binomial regression models. A similar comparison was made using models for the expected crash frequency of single- (SV) and multi-vehicle (MV) crashes. The models were estimated using two-lane, two-way rural highway segment-level data from an engineering district in Pennsylvania. The results show that all bivariate negative binomial models (with or without copulas) outperformed the corresponding univariate negative binomial models for PDO and FI, as well as SV and MV, crashes. Second, the statistical association of various traffic and roadway/roadside features with PDO and FI, as well as SV and MV crashes, were not the same relative to their corresponding relationships in the univariate models. The bivariate negative binomial model with normal copula outperformed all other models based on the goodness-of-fit statistics. The results suggest that copula-based bivariate negative binomial regression models may be a valuable alternative for univariate models when simultaneously modeling two disaggregate levels of crash counts. [ABSTRACT FROM AUTHOR]

• CMFs for adaptive traffic signal control (ATSC) systems developed using Empirical Bayes before-after study. • CMFs for ATSC system impacts at individual intersections and along entire adaptive corridors are estimated. • Temporal trends in safety performance of ATSC system also studied. • Results reveal minor increase in total crash frequency and decrease in fatal + injury crash frequency. Adaptive traffic signal control (ATSC) is a novel traffic management system that 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 light of this, 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). Temporal trends for intersection-level CMFs were examined using annual crash data in the after period. 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. In addition, the temporal trend analysis indicates that the safety effectiveness does not vary annually in the after period, suggesting no obvious novelty effect associated with ATSC. Finally, the magnitude of the corridor-level CMFs are slightly lower than the intersection-level CMFs, except for rear-end crashes. [ABSTRACT FROM AUTHOR]

• SPFs developed using negative binomial (NB) and random parameters negative binomial (RPNB) modeling frameworks. • Fit and predictive power compared on both within sample and out-of-sample observations. • RPNB model provides better predictions than NB model when applied to within sample observations. • NB model provides better predictions than RPNB model when applied to out-of-sample observations. Negative binomial (NB) regression is among the most common statistical modeling methods used to model crash frequencies due to its simple functional form and ability to handle over-dispersion commonly found in crash data. However, a drawback of this approach is that regression parameters are assumed to be the same across observations, which could contribute to biased parameter estimates. To alleviate this concern, the random parameters negative binomial (RPNB) model was recently proposed, which allows regression parameters to differ across observations following some known distribution. The resulting coefficients should be less biased, and thus the RPNB approach is believed to provide a more accurate relationship between independent variables and expected crash frequency. However, the prediction accuracy of the RPNB model relative to the standard NB model has not been thoroughly evaluated, particularly with respect to out-of-sample observations for which unique random parameters cannot be estimated. In this paper, the predictive power of the RPNB and NB models are examined using two-lane rural highway data from three engineering Districts in Pennsylvania. Multiple evaluation metrics are applied—root-mean-square error (RMSE) and mean absolute error (MAE), coefficients from calibration functions and cumulative residual (CURE) plots—to assess each model type. The results show that the RPNB model outperforms the NB model when applied to within sample observations (i.e., those used to estimate the model) by making use of the observation-specific coefficients. However, the predictive power of the RPNB model appears to be similar to or slightly less precise than the traditional NB model when applied to out-of-sample observations. Since the RPNB model is estimated using a simulation-based approach, sensitivity tests were also performed to see how the parameter estimates change with the number of Halton draws used to perform the simulation. For the sample sizes used in this paper, the estimates were fairly insensitive when more than 50 Halton draws were used. The findings suggest that the RPNB model is more reliable when applied to the same set of sites that were used to estimate the model but might not be as robust as the traditional NB model when applied to other sites. [ABSTRACT FROM AUTHOR]

To approach the goal of "Toward Zero Deaths," there is a need to develop an analysis paradigm to better understand the effects of a countermeasure on reducing the number of severe crashes. One of the goals in traffic safety research is to search for an effective treatment to reduce fatal and major injury crashes, referred to as severe crashes. To achieve this goal, the selection of promising countermeasures is of utmost importance, and relies on the effectiveness of candidate countermeasures in reducing severe crashes. Although it is important to precisely evaluate the effectiveness of candidate countermeasures in reducing the number of severe crashes at a site, the current state-of-the-practice often leads to biased estimates. While there have been a few advanced statistical models developed to mitigate the problem in practice, these models are computationally difficult to estimate because severe crashes are dispersed spatially and temporally, and cannot be integrated into the Highway Safety Manual framework, which develops a series of safety performance functions and crash modification factors to predict the number of crashes. Crash severity outcomes are generally integrated into the Highway Safety Manual using deterministic distributions rather than statistical models. Accounting for the variability in crash severity as a function geometric design, traffic flow, and other roadway and roadside features is afforded by estimating statistical models. Therefore, there is a need to develop a new analysis paradigm to resolve the limitations in the current Highway Safety Manual methods. We propose an approach which decomposes the severe crash frequency into a function of the change in the total number of crashes and the probability of a crash becoming a severe crash before and after a countermeasure is implemented. We tested this approach by evaluating the effectiveness of shoulder rumble strips on reducing the number of severe crashes. A total of 310 segments that have had shoulder rumble strips installed during 2002-2009 are included in the analysis. It was found that shoulder rumble strips reduce the total number of crashes, but have no statistically significant effect on reducing the probability of a severe crash outcome., (Copyright © 2014 Elsevier Ltd. All rights reserved.)