Exploration and comparison of crash modification factors for multiple treatments on rural multilane roadways

As multiple treatments (or countermeasures) are simultaneously applied to roadways, there is a need to assess their combined safety effects. Due to a lack of empirical crash modification factors (CMFs) for multiple treatments, the Highway Safety Manual (HSM) and other related studies developed various methods of combining multiple CMFs for single treatments. However, the literature did not evaluate the accuracy of these methods using CMFs obtained from the same study area. Thus, the main objectives of this research are: (1) develop CMFs for two single treatments (shoulder rumble strips, widening shoulder width) and one combined treatment (shoulder rumble strips + widening shoulder width) using before–after and cross-sectional methods and (2) evaluate the accuracy of the combined CMFs for multiple treatments estimated by the existing methods based on actual evaluated combined CMFs. Data was collected for rural multi-lane highways in Florida and four safety performance functions (SPFs) were estimated using 360 reference sites for two crash types (All crashes and Single Vehicle Run-off Roadway (SVROR) crashes) and two severity levels (all severity (KABCO) and injury (KABC)).     

Assessment of safety effects for widening urban roadways in developing crash modification functions using nonlinearizing link functions

Since a crash modification factor (CMF) represents the overall safety performance of specific treatments in a single fixed value, there is a need to explore the variation of CMFs with different roadway characteristics among treated sites over time. Therefore, in this study, we (1) evaluate the safety performance of a sample of urban four-lane roadway segments that have been widened with one through lane in each direction and (2) determine the relationship between the safety effects and different roadway characteristics over time. Observational before–after analysis with the empirical Bayes (EB) method was assessed in this study to evaluate the safety effects of widening urban four-lane roadways to six-lanes. Moreover, the nonlinearizing link functions were utilized to achieve better performance of crash modification functions (CMFunctions). The CMFunctions were developed using a Bayesian regression method including the estimated nonlinearizing link function to incorporate the changes in safety effects of the treatment over time. Data was collected for urban arterials in Florida, and the Florida-specific full SPFs were developed and used for EB estimation.     

Developing crash modification functions to assess safety effects of adding bike lanes for urban arterials with different roadway and socio-economic characteristics

Although many researchers have estimated crash modification factors (CMFs) for specific treatments (or countermeasures), there is a lack of studies that explored the heterogeneous effects of roadway characteristics on crash frequency among treated sites. Generally, the CMF estimated by before–after studies represents overall safety effects of the treatment in a fixed value. However, as each treated site has different roadway characteristics, there is a need to assess the variation of CMFs among the treated sites with different roadway characteristics through crash modification functions (CMFunctions). The main objective of this research is to determine relationships between the safety effects of adding a bike lane and the roadway characteristics through (1) evaluation of CMFs for adding a bike lane using observational before–after with empirical Bayes (EB) and cross-sectional methods, and (2) development of simple and full CMFunctions which are describe the CMF in a function of roadway characteristics of the sites. Data was collected for urban arterials in Florida, and the Florida-specific full SPFs were developed. Moreover, socio-economic parameters were collected and included in CMFunctions and SPFs (1) to capture the effects of the variables that represent volume of bicyclists and (2) to identify general relationship between the CMFs and these characteristics. In order to achieve better performance of CMFunctions, data mining techniques were used.     

Assessing the safety effects of multiple roadside treatments using parametric and nonparametric approaches

This study evaluates the safety effectiveness of multiple roadside elements on roadway segments by estimating crash modification factors (CMFs) using the cross-sectional method. To consider the nonlinearity in crash predictors, the study develops generalized nonlinear models (GNMs) and multivariate adaptive regression splines (MARS) models. The MARS is one of the promising data mining techniques due to its ability to consider the interaction impact of more than one variables and nonlinearity of predictors simultaneously. The CMFs were developed for four roadside elements (driveway density, poles density, distance to poles, and distance to trees) and combined safety effects of multiple treatments were interpreted by the interaction terms from the MARS models. Five years of crash data from 2008 to 2012 were collected for rural undivided four-lane roadways in Florida for different crash types and severity levels. The results show that the safety effects decrease as density of driveways and roadside poles increase. The estimated CMFs also indicate that increasing distance to roadside poles and trees reduces crashes. The study demonstrates that the GNMs show slightly better model fitness than negative binomial (NB) models. Moreover, the MARS models outperformed NB and GNM models due to its strength to reflect the nonlinearity of crash predictors and interaction impacts among variables under different ranges. Therefore, it can be recommended that the CMFs are estimated using MARS when there are nonlinear relationships between crash rate and roadway characteristics, and interaction impacts among multiple treatments.     

Using multivariate adaptive regression splines (MARS) to develop crash modification factors for urban freeway interchange influence areas

Crash modification factors (CMFs) are used to measure the safety impacts of changes in specific geometric characteristics. Their development has gained much interest following the adoption of CMFs by the recently released Highway Safety Manual (HSM) and SafetyAnalyst tool in the United States. This paper describes a study to develop CMFs for interchange influence areas on urban freeways in the state of Florida. Despite the very different traffic and geometric conditions that exist in interchange influence areas, most previous studies have not separated them from the rest of the freeway system in their analyses. In this study, a promising data mining method known as multivariate adaptive regression splines (MARS) was applied to develop CMFs for median width and inside and outside shoulder widths for “total” and “fatal and injury” (FI) crashes. In addition, CMFs were also developed for the two most frequent crash types, i.e., rear-end and sideswipe. MARS is characterized by its ability to accommodate the nonlinearity in crash predictors and to allow the impact of more than one geometric variable to be simultaneously considered. The methodology further implements crash predictions from the model to identify changes in geometric design features. Four years of crashes from 2007 to 2010 were used in the analysis and the results showed that MARS's prediction capability and goodness-of-fit statistics outperformed those of the negative binomial model. The influential variables identified included the outside and inside shoulder widths, median width, lane width, traffic volume, and shoulder type. It was deduced that a 2-ft increase in the outside and inside shoulders (from 10 ft to 12 ft) reduces FI crashes by 10% and 33%, respectively. Further, a 42-ft reduction in the median width (from 64 ft to 22 ft) increases the rear-end, total, and FI crashes by 473%, 263%, and 223%, respectively.     

Accounting for heterogeneity among treatment sites and time trends in developing crash modification functions

Collision modification factors (CMFs) are commonly used to quantify the impact of safety countermeasures. The CMFs obtained from observational before–after (BA) studies are usually estimated by averaging the safety impact (i.e., index of effectiveness) for a group of treatment sites. The heterogeneity among the treatment locations, in terms of their characteristics, and the effect of this heterogeneity on safety treatment effectiveness are usually ignored. This is in contrast to treatment evaluations in other fields like medical statistics where variations in the magnitude (or in the direction) of response to the same treatment given to different patients are considered.     

Relating crash frequency and severity: Evaluating the effectiveness of shoulder rumble strips on reducing fatal and major injury crashes

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.     

Exploring the transferability of safety performance functions

Safety performance functions (SPFs), by predicting the number of crashes on roadway facilities, have been a vital tool in the highway safety area. The SPFs are typically applied for identifying hot spots in network screening and evaluating the effectiveness of road safety countermeasures. The Highway Safety Manual (HSM) provides a series of SPFs for several crash types by various roadway facilities. The SPFs, provided in the HSM, were developed using data from multiple states. In regions without local jurisdiction based SPFs it is common practice to adopt national SPFs for crash prediction. There has been little research to examine the viability of such national level models for local jurisdictions. Towards understanding the influence of SPF transferability, we examine the rural divided multilane highway models from Florida, Ohio, and California. Traffic, roadway geometry and crash data from the three states are employed to estimate single-state SPFs, two-state SPFs and three-state SPFs. The SPFs are estimated using the negative binomial model formulation for several crash types and severities. To evaluate transferability of models, we estimate a transfer index that allows us to understand which models transfer adequately to other regions. The results indicate that models from Florida and California seem to be more transferable compared to models from Ohio. More importantly, we observe that the transfer index increases when we used pooled data (from two or three states). Finally, to assist in model transferability, we propose a Modified Empirical Bayes (MEB) measure that provides segment specific calibration factors for transferring SPFs to local jurisdictions. The proposed measure is shown to outperform the HSM calibration factor for transferring SPFs.     

Methodology to develop crash modification functions for road safety treatments with fully specified and hierarchical models

Crash modification factors (CMFs) for road safety treatments are developed as multiplicative factors that are used to reflect the expected changes in safety performance associated with changes in highway design and/or the traffic control features. However, current CMFs have methodological drawbacks. For example, variability with application circumstance is not well understood, and, as important, correlation is not addressed when several CMFs are applied multiplicatively. These issues can be addressed by developing safety performance functions (SPFs) with components of crash modification functions (CM-Functions), an approach that includes all CMF related variables, along with others, while capturing quantitative and other effects of factors and accounting for cross-factor correlations. CM-Functions can capture the safety impact of factors through a continuous and quantitative approach, avoiding the problematic categorical analysis that is often used to capture CMF variability. There are two formulations to develop such SPFs with CM-Function components – fully specified models and hierarchical models. Based on sample datasets from two Canadian cities, both approaches are investigated in this paper. While both model formulations yielded promising results and reasonable CM-Functions, the hierarchical model was found to be more suitable in retaining homogeneity of first-level SPFs, while addressing CM-Functions in sub-level modeling. In addition, hierarchical models better capture the correlations between different impact factors.     

Finite mixture modeling approach for developing crash modification factors in highway safety analysis

This study aimed to investigate the relative performance of two models (negative binomial (NB) model and two-component finite mixture of negative binomial models (FMNB-2)) in terms of developing crash modification factors (CMFs). Crash data on rural multilane divided highways in California and Texas were modeled with the two models, and crash modification functions (CMFunctions) were derived. The resultant CMFunction estimated from the FMNB-2 model showed several good properties over that from the NB model. First, the safety effect of a covariate was better reflected by the CMFunction developed using the FMNB-2 model, since the model takes into account the differential responsiveness of crash frequency to the covariate. Second, the CMFunction derived from the FMNB-2 model is able to capture nonlinear relationships between covariate and safety. Finally, following the same concept as those for NB models, the combined CMFs of multiple treatments were estimated using the FMNB-2 model. The results indicated that they are not the simple multiplicative of single ones (i.e., their safety effects are not independent under FMNB-2 models). Adjustment Factors (AFs) were then developed. It is revealed that current Highway Safety Manual’s method could over- or under-estimate the combined CMFs under particular combination of covariates. Safety analysts are encouraged to consider using the FMNB-2 models for developing CMFs and AFs.     

Development of crash modification factors for changing lane width on roadway segments using generalized nonlinear models

This study evaluates the effectiveness of changing lane width in reducing crashes on roadway segments. To consider nonlinear relationships between crash rate and lane width, the study develops generalized nonlinear models (GNMs) using 3-years crash records and road geometry data collected for all roadway segments in Florida. The study also estimates various crash modification factors (CMFs) for different ranges of lane width based on the results of the GNMs. It was found that the crash rate was highest for 12-ft lane and lower for the lane width less than or greater than 12 ft. GNMs can extrapolate this nonlinear continuous effect of lane width and estimate the CMFs for any lane width, not only selected lane widths, unlike generalized linear models (GLMs) with categorical variables. The CMFs estimated using GNMs reflect that crashes are less likely to occur for narrower lanes if the lane width is less than 12 ft whereas crashes are less likely to occur for wider lanes if the lane width is greater than 12 ft. However, these effects varied with the posted speed limits as the effect of interaction between lane width and speed limit was significant. The estimated CMFs show that crashes are less likely to occur for lane widths less than 12 ft than the lane widths greater than 12 ft if the speed limit is higher than or equal to 40 mph. It was also found from the CMFs that crashes at higher severity levels (KABC and KAB) are less likely to occur for lane widths greater or less than 12 ft compared to 12-ft lane. The study demonstrates that the CMFs estimated using GNMs clearly reflect variations in crashes with lane width, which cannot be captured by the CMFs estimated using GLMs. Thus, it is recommended that if the relationship between crash rate and lane width is nonlinear, the CMFs are estimated using GNMs.     

Prioritizing Highway Safety Manual’s crash prediction variables using boosted regression trees

The Highway Safety Manual (HSM) recommends using the empirical Bayes (EB) method with locally derived calibration factors to predict an agency’s safety performance. However, the data needs for deriving these local calibration factors are significant, requiring very detailed roadway characteristics information. Many of the data variables identified in the HSM are currently unavailable in the states’ databases. Moreover, the process of collecting and maintaining all the HSM data variables is cost-prohibitive. Prioritization of the variables based on their impact on crash predictions would, therefore, help to identify influential variables for which data could be collected and maintained for continued updates. This study aims to determine the impact of each independent variable identified in the HSM on crash predictions. A relatively recent data mining approach called boosted regression trees (BRT) is used to investigate the association between the variables and crash predictions. The BRT method can effectively handle different types of predictor variables, identify very complex and non-linear association among variables, and compute variable importance. Five years of crash data from 2008 to 2012 on two urban and suburban facility types, two-lane undivided arterials and four-lane divided arterials, were analyzed for estimating the influence of variables on crash predictions. Variables were found to exhibit non-linear and sometimes complex relationship to predicted crash counts. In addition, only a few variables were found to explain most of the variation in the crash data.     

Full Bayesian evaluation of the safety effects of reducing the posted speed limit in urban residential area

Full Bayesian (FB) before–after evaluation is a newer approach than the empirical Bayesian (EB) evaluation in traffic safety research. While a number of earlier studies have conducted univariate and multivariate FB before–after safety evaluations and compared the results with the EB method, often contradictory conclusions have been drawn. To this end, the objectives of the current study were to (i) perform a before–after safety evaluation using both the univariate and multivariate FB methods in order to enhance our understanding of these methodologies, (ii) perform the EB evaluation and compare the results with those of the FB methods and (iii) apply the FB and EB methods to evaluate the safety effects of reducing the urban residential posted speed limit (PSL) for policy recommendation. In addition to three years of crash data for both the before and after periods, traffic volume, road geometry and other relevant data for both the treated and reference sites were collected and used. According to the model goodness-of-fit criteria, the current study found that the multivariate FB model for crash severities outperformed the univariate FB models. Moreover, in terms of statistical significance of the safety effects, the EB and FB methods led to opposite conclusions when the safety effects were relatively small with high standard deviation. Therefore, caution should be taken in drawing conclusions from the EB method. Based on the FB method, the PSL reduction was found effective in reducing crashes of all severities and thus is recommended for improving safety on urban residential collector roads.     

Effects on speed and safety of point-to-point speed enforcement systems: Evaluation on the urban motorway A56 Tangenziale di Napoli

In this paper, we evaluated the effects on speed and safety of the point-to-point (P2P) speed enforcement system activated on the urban motorway A56 in Italy. The P2P speed enforcement is a relatively new approach to traffic law enforcement that involves the calculation of the average speed over a section. To evaluate the speed effects, we performed a before–after analysis of speed data investigating also effects on non-compliance to speed limits. To evaluate the safety effects, we carried out an empirical Bayes observational before-and-after study.     

How the choice of safety performance function affects the identification of important crash prediction variables

Across the nation, researchers and transportation engineers are developing safety performance functions (SPFs) to predict crash rates and develop crash modification factors to improve traffic safety at roadway segments and intersections. Generalized linear models (GLMs), such as Poisson or negative binomial regression, are most commonly used to develop SPFs with annual average daily traffic as the primary roadway characteristic to predict crashes. However, while more complex to interpret, data mining models such as boosted regression trees have improved upon GLMs crash prediction performance due to their ability to handle more data characteristics, accommodate non-linearities, and include interaction effects between the characteristics.An intersection data inventory of 36 safety relevant parameters for three- and four-legged non-signalized intersections along state routes in Alabama was used to study the importance of intersection characteristics on crash rate and the interaction effects between key characteristics. Four different SPFs were investigated and compared: Poisson regression, negative binomial regression, regularized generalized linear model, and boosted regression trees. The models did not agree on which intersection characteristics were most related to the crash rate. The boosted regression tree model significantly outperformed the other models and identified several intersection characteristics as having strong interaction effects.     

Determining the potential safety benefit of improved lighting in three pedestrian crash scenarios

The influence of light level was determined for three pedestrian crash scenarios associated with three adaptive headlighting solutions-curve lighting, motorway lighting, and cornering light. These results were coupled to corresponding prevalence data for each scenario to derive measures of annual lifesaving potential. For each scenario, the risk associated with light level was determined using daylight saving time (DST) transitions to produce a dark/light interval risk ratio; prevalence was determined using the corresponding annual crash rate in darkness for each scenario. For curve lighting, pedestrian crashes on curved roadways were examined; for motorway lighting, crashes associated with high speed roadways were examined; and for cornering light, crashes involving turning vehicles at intersections were examined. In the curve analysis, lower dark/light crash ratios were observed for curved sections of roadway compared to straight roads. In the motorway analysis, posted speed limit was the dominant predictor of this ratio for the fatal crash dataset; road function class was the dominant predictor of the ratio for the fatal/nonfatal dataset. Finally, in the intersection crash analysis, the dark/light ratio for turning vehicles was lower than for nonturning vehicles; and the ratio at intersections was lower than at non-intersections. Relative safety need was determined by combining the dark/light ratio with prevalence data to produce an idealized measure of lifesaving potential. While all three scenarios suggested a potential for safety improvement, scenarios related to high speed roadway environments showed the greatest potential.     

Comparison of safety effect estimates obtained from empirical Bayes before–after study,propensity scores-potential outcomes framework,and regression model with cross-sectional data

A variety of different study designs and analysis methods have been used to evaluate the performance of traffic safety countermeasures. The most common study designs and methods include observational before–after studies using the empirical Bayes method and cross-sectional studies using regression models. The propensity scores-potential outcomes framework has recently been proposed as an alternative traffic safety countermeasure evaluation method to address the challenges associated with selection biases that can be part of cross-sectional studies. Crash modification factors derived from the application of all three methods have not yet been compared. This paper compares the results of retrospective, observational evaluations of a traffic safety countermeasure using both before–after and cross-sectional study designs.     

Effect of shoulder width, guardrail and roadway geometry on driver perception and behavior

Roadway design is one of the most significant factors that affect driving behavior and perceived safety. The current study tests the combined effects of three roadway design elements – shoulders width, guardrail existence and roadway geometry (curvature) – on objective driving measures (speed and lane position), and subjective measures (perceived safe driving speed and estimated road safety).Twenty two drivers participated in an experiment with a driving simulation. In the first part objective driving data were collected, and in the second part subjective paper–pencil evaluations were requested of the perceived safety of 30 different scenarios that were previously experienced in the simulator. The scenarios consisted of the various combinations of the three roadway design elements.The results showed a significant effect of roadway geometry on both objective and subjective measures. The shoulders width had a significant effect on actual speed, on lane position, and on perceived safe driving speed, but only when a guardrail was present.These findings illustrate the perceptual role of a guardrail in defining the perceived safety margins that various shoulder widths provide. When a guardrail is absent, the width of the shoulder loses much of its benefits and effects on driving behavior. The results also demonstrate that roadway geometry can be used to reduce driving speeds, but at the same time it can have a negative effect on maintaining a stable lane position in sharp curves. Thus, controlling the width of road shoulders and the placement of guardrails seems to be a safer approach to speed and lane position control.