Transferability of Safety Performance Functions: The Case of Urban Four-Lane Divided Roadways in Muscat

Rezapur-Shahkolai F., Afshari M., Doosti-Irani A., Bashirian S., Maleki S.

Road traffic injuries (RTIs) are a serious public health problem and are an important cause of disability and death worldwide. A significant proportion of traffic injuries and fatalities are among pedestrians. Reviewing and evaluating related interventions can be a practical step to implementing appropriate methods to prevent RTIs among pedestrians, a highly vulnerable group of road users.The search of articles was conducted in the electronic databases of Scopus, PubMed, ISI, Safety Lit, and CDC. Other papers were also reviewed using forward citation and backward citation. The search strategy was for studies examined from the first years of database creation until January 10, 2021, in all languages in journals with matched judgment according to the type of population, type of interventions, comparators, and results. Joanna Briggs Institute (JBI) checklists were used to determine articles' quality and assess possible biases depending on the type of study.The initial search resulted in finding 16,272 abstracts. Finally, 25 studies, including 17 randomized clinical trials (RCTs) studies, seven studies pretest/post-test interventions (PPI), and one study controlled pretest/post-test interventions (cPPI) met the inclusion criteria. Among these preventive interventional studies, 20 studies were conducted in high-income countries, three were in lower middle-income countries, one study was in upper middle-income country, and only one study was conducted in a low-income country. Most interventional studies in the field of prevention of RTIs (15 studies) had an educational/behavioral approach, designed to change pedestrian behavior and the use of this approach was also effective in improving pedestrian behavior. The legislation/law enforcement approach was used in one study and two studies used an engineering/technology approach. In studies with an engineering approach after engineering reforms, pedestrian injuries in children decreased by 37.5%. In seven studies, multifaceted interventions were used. The interventional studies that used this approach were able to improve pedestrian safe behavior.The majority of studies were based on educational/behavioral approaches and pedestrians' behavior improved notably. In addition, the majority of interventional studies were conducted in countries with high income levels. Therefore, programming for preventive interventions to attenuate RTIs is highly important in low/middle-income countries to reduce the risk of injury to vulnerable road users. These findings can be applied by policy-makers to develop educational, engineering, environmental, and law enforcement interventions and attenuate injuries sustained by pedestrians.

Al-Ahmadi H.M., Jamal A., Ahmed T., Rahman M.T., Reza I., Farooq D.

Crash prediction models (CPM) are mostly used for network screening in the road safety management process. The Highway Safety Manual (HSM) offers consistent and reliable CPMs for various roadway facilities that are commonly known as safety performance functions (SPFs). SPFs are statistical regression models that estimate the expected crash frequencies by crash severity, type, and facility types as a function of highway geometric characteristics and traffic exposure. They are vital in identifying high-frequency crash locations and assessing the effectiveness of safety countermeasures. HSM SPFs were originally developed using data collected from a selected few states in the USA. When applied to different jurisdictions, agencies can either develop local SPFs or calibrate the existing HSM base SPFs for local conditions depending on various trade-offs. This study aims to calibrate HSM-default SPFs for multilanes rural divided highway segments using three years of crash data (2017–2019) in the Kingdom of Saudi Arabia (KSA). In this regard, two highways (NHWY-80 and NHWY-85) in the eastern region were considered for the analysis. Crash and traffic data were procured from the MOT (Ministry of Transport), Riyadh, KSA. Geometric data was collected from MOT as well as google earth and field surveys. Calibration procedure as recommended by HSM was followed to obtain the local calibration factors. The Interactive Highway Safety Design Model (IHSDM) calibrator tool was used for the analysis. SPFs calibration results revealed that HSM predictive methodology consistently overpredicts all types of crashes (i.e., total, fatal and injury, and property damage crashes) on both highways. The estimated calibration factors ranged from 0.53 to 0.78. Various goodness of fit (GOF) measures (like MAD, MSPE, MPB) were used for quality assessment of calibrated SPFs. Methods used in this study could be beneficially practiced in any jurisdiction. Calibrated SPFs provide a favorable alternative and replacement of HSM-default SPFs, thereby making the crash predictions more accurate and thus helps in better decision-making related to highway safety.

Al-Aamri A.K., Hornby G., Zhang L., Al-Maniri A.A., Padmadas S.S.

Objective: Road traffic crashes (RTCs) are a major global public health problem and cause substantial burden on national economy and healthcare. There is little systematic understanding of the geography of RTCs and the spatial correlations of RTCs in the Middle-East region, particularly in Oman where RTCs are the leading cause of disability-adjusted life years lost. The overarching goal of this paper is to evaluate the spatial and temporal dimensions, identifying the high risk areas or hot-zones where RTCs are more frequent, using the geocoded data from the Muscat governorate. Data: This study is based on data drawn from the Royal Oman Police (ROP) sample iMAAP database and the National Road Traffic Crash (NRTC) database, managed by the ROP and made available for research use by The Research Council of the Sultanate of Oman. The data covered the period from 1st January 2010 to 2nd November 2014. Only RTCs occurred in Muscat Governorate were included in the study. The study is based on 12,438 registered incidents, however, due to disconnections found on road network, RTCs occurred on disconnected parts were removed and the final analysis considered only 9,357 incidents. Methods: We considered an adjacency network analysis integrating GIS and RTC data using robust estimation techniques including: Kernel Density Estimation (KDE) of both 1-D and 2-D space dimensions, Network-based Nearest Neighbour Distance (Net-NND), Network-based K-Function, Random Forest Algorithm (RF) and spatiotemporal Hot-zone analysis. Findings: The analysis highlight evidence of spatial clustering and recurrence of RTC hot-zones on long roads demarcated by intersections and roundabouts in Muscat. The findings confirm that road intersections elevate the risk of RTCs than other effects attributed to road geometry features. The results from GIS application of NRTC data are validated using the sample data generated by iMAAP database. Conclusion: The findings of this study provide statistical evidence and confirm our research hypothesis that road intersections (roundabouts, crosses and bridges) represent higher risk of causing RTCs than other road geometric features. The results also demonstrate systematic quantitative evidence of spatio-temporal patterns associated with the crash risk over different locations on road networks in Muscat. More importantly, the findings clearly pinpoint the importance and influence of the road and traffic related features in road crash spatial analysis.

Feng M., Wang X., Lee J., Abdel-Aty M., Mao S.

Safety performance function (SPF) has been a vital tool in traffic safety evaluation including finding contributing factors to crashes, identifying hotspots, and assessing safety effects of countermeasures. In the United States (U.S.), the Highway Safety Manual provides a number of SPFs for a variety of road facilities. Due to the limited availability of traffic data in many regions, the transferability of SPFs has been an important topic in traffic safety analysis and has been evaluated by several studies. Nevertheless, the international transferability of freeway SPFs and the applicability of transferred SPFs on hotspot identification has been rarely investigated. Based on data from two Chinese cities, Shanghai and Suzhou, and three U.S. states, Texas, New York and Florida, this study analyzes the transferability of freeway SPFs between Chinese and U.S. regions. These SPFs are then transferred to the other country and their performance on hotspot identification is investigated. SPFs were developed in the frameworks of Poisson, Poisson-lognormal and negative binomial regressions for the five localities separately, and were calibrated using the calibration functions before being transferred. Without calibration, the poor model transferability was found between the two countries, while after calibration, the transferred SPFs between Shanghai/Suzhou and Texas/New York showed satisfactory performance on both model fitting and hotspot identification. However, the transferability of SPFs between Florida and the Chinese cities turned out to be unsatisfactory regardless of whether being calibrated or not, which was attributable to the considerable difference in traffic flow. The findings of this study are expected to be a good reference for researchers and practitioners who want to understand the transferability and applicability of SPFs in the international context.

Farrag S.G., Outay F., Yasar A.U., Janssens D., Kochan B., Jabeur N.

In addition to their environmental impact, road traffic congestions have been recognized to seriously affect commuters’ safety as well as the performance of transportation systems. To address these issues, various Traffic Incident Management (TIM) systems have been implemented. Recent systems are particularly focusing on the integration of promising emergent technologies such as the Internet of Things. However, thorough studies are still necessary to make sure that these technologies are compatible with existing systems and effective within their context of use. The main goal of this research is to develop a smart TIM system which is based on Car2X communications and which aims at improving both traffic safety, commuters’ mobility, and gas emissions. To assess the effectiveness of our solution, we use the following measures: stops delay, stops all, vehicle delay, travel time, gas emissions, and fuel consumption. This paper also outlines how we use our simulation platform (which was developed based on VISSIM and Python) to quantify the benefits of using Car2X communications. We run our simulations on Muscat Expressway in the Sultanate of Oman. Results are promising and include (1) the travel time decreased by 6%; (2) the average stop delay and vehicle stops were reduced by at least 9% and 27% respectively; and (3) there is a total decrease in fuel consumption and carbon monoxide emissions by approximately 16%.

Matarage I.C., Dissanayake S.

The Safety Performance Functions (SPFs) in the Highway Safety Manual (HSM) are statistical formulas developed based on limited data gathered from selected few states. Therefore, the HSM recommends calibrating the HSM-default SPFs to local conditions or developing local SPFs to enhance the accuracy of crash prediction. This paper demonstrates the HSM calibration procedure for freeway segments and the quality assessment of the calibration process using Kansas freeway geometric and crash data. A minimum sample size of 446 freeway segments was calculated corresponding to 95% confidence level and 5% error; consequently, data for 521 freeway segments were collected and utilized in this freeway calibration. Results indicated that the HSM methodology overpredicts fatal and injury crashes and underpredicts property damage only crashes for freeway segments in Kansas. Results of quality assessment of the calibration process showed that estimated calibration factors were satisfactory for all freeway types considered in this study.

Wu L., Lord D., Geedipally S.R.

Horizontal curves have been identified as experiencing more crashes than tangent sections on roadways, especially on rural two-lane highways. The first edition of the Highway Safety Manual provides crash modification functions (CM functions) for curves on rural two-lane highways. The CM functions proposed in the manual may suffer from both outdated data and analysis technique. Before-and-after studies are usually the preferred method for estimating the safety effects of treatments. Unfortunately, this method is not feasible for curves. Previous studies have frequently used regression models for developing CM functions for horizontal curves. As recently documented in the literature, some potential problems exist with using regression models to develop crash modification factors. This research utilized a cross-sectional study to develop curvature CM functions. Curves located on Texas rural two-lane undivided highways were divided into a number of bins based on the curve radius. Safety was predicted with the assumption that these curves had been tangents. The observed number of crashes that occurred on the curves was compared with the dummy tangents and for different bins. The results showed that the horizontal curve radius has a significant role in the risk of a crash. From these results, a new CM function was developed. The prediction performance of the Highway Safety Manual CM function was compared with the new CM function in this study and another function that was recently proposed in the literature. It was found that the new CM function documented in this study outperformed both.

Srinivasan R., Colety M., Bahar G., Crowther B., Farmen M.

This paper describes a study that used data from rural two-lane roads in Arizona to illustrate two issues. One issue is the importance of selecting an appropriate sample size for calibrating the Highway Safety Manual (HSM) predictive models based on the desired accuracy of the calibration factor, instead of relying on the guidance from the HSM (the HSM recommends 30 to 50 sites with at least 100 crashes per year but provides no statistical basis for this guidance). The second issue is the usefulness of estimating calibration functions when individual calibration factors do not provide a proper fit of the local data. On the basis of the outcome of the exploration of these two issues, this study recommends a simple calibration function for predicting total crashes on rural two-lane roads in Arizona. This paper also provides a brief overview of a procedure in Microsoft Excel that can be used by practitioners (after appropriate training) to estimate simple calibration functions.

Kaaf K.A., Abdel-Aty M.

The first edition of the Highway Safety Manual (HSM) provides a number of safety performance functions (SPFs) that can be used to predict severe collisions on a roadway network. This paper examines the calibration of the HSM SPFs for urban four-lane divided roadway segments with angle parking in Riyadh, Saudi Arabia, and the development of new SPFs. This study first calibrated the HSM SPFs by using HSM default crash modification factors (CMFs) and then proposed new local CMFs, which adjust the estimation of calibration factors by using fatal and injury crash data. In addition, new forms for specific SPFs were further evaluated to identify the best model by using the Poisson–gamma regression technique. It was found that the jurisdiction-specific SPFs provided the best fit of the data used in this study and would be the best SPFs for predicting severe collisions in the city of Riyadh. The best fatal and injury model describes the mean crash frequency as a function of the natural logarithm of the annual average daily traffic, segment length, speed limit, and driveway density. The study found that the HSM calibration using Riyadh local CMFs outperformed the calibration method using HSM default values. Based on these results, potential countermeasures were proposed to reduce the number of severe crashes on Riyadh urban roads, and the potential for HSM application in Saudi Arabia is addressed.

Lu J., Haleem K., Alluri P., Gan A., Liu K.

• We separated freeways into basic segments and interchange influence areas. • Both urban and rural freeways were used in the analysis. • We compared local SPFs with calibrated default SPFs in SafetyAnalyst . • We used four years of crashes (2007–2010) from Florida in developing the models. • Local SPFs generally produced better models than calibrated SPFs in SafetyAnalyst . Safety performance functions (SPFs) are a required input to the newly released SafetyAnalyst software tool. Although SafetyAnalyst provides default SPFs that were developed based on data from multiple states in the United States, agencies have the option to calibrate local SPFs to better reflect their local conditions. However, the benefit from local calibration of SPFs is unclear and may vary from state to state. Using statistical goodness-of-fit measures and visual plots, this paper compares the performance of locally-calibrated SPFs using Florida data with the default SPFs from SafetyAnalyst for both freeway interchange influence areas and basic segments. An interchange influence area is one that extends 0.3 miles upstream and downstream of the respective gore point. Unlike for intersections, an automatic process for segmenting the influence areas for interchanges based on the above definition is relatively complex. Therefore, this paper also describes a spatial method to automatically segment a freeway facility into interchange influence areas and basic segments. Using four years of local crash data (2007–2010) from Florida, SPFs for both types of segments for both urban and rural areas were developed using the negative binomial regression model. The results showed that Florida-specific SPFs generally produced better-fitted models than the calibrated SafetyAnalyst default SPFs. This was clear in that the majority of Florida-specific models had higher Freeman–Tukey R -square ( R FT 2 ), as well as lower mean absolute deviance (MAD) and mean square prediction error (MSPE) estimates. Overall, the results suggest that agencies implementing SafetyAnalyst could improve their crash prediction by developing local-specific SPFs.

Mehta G., Lou Y.

One critical component of the Highway Safety Manual (HSM) statistical methods is the safety performance function (SPF). SPFs are essentially regression models that correlate quantitatively the expected number of crashes with traffic exposure and geometric characteristics of the road. As part of a project performed by the University of Alabama to facilitate implementation of the new HSM procedures in the state, this study aims to evaluate the applicability of HSM predictive methods to Alabama data and to develop state-specific statistical models for two facility types: two-lane, two-way rural roads and four-lane divided highways. This study first calibrates HSM base SPFs by using two approaches: the method recommended by the HSM and a newly proposed approach that treats the estimation of calibration factors as a special case of a negative binomial regression. In addition, new forms of state-specific SPFs are further investigated by using Poisson-gamma regression techniques. Four new functional forms are studied in this project. The prediction capabilities of the two calibrated models and the four newly developed state-specific SPFs are evaluated with a validation data set. Five performance measures are considered for model evaluation. The study is able to identify a particular state-specific SPF that fits the Alabama data well and outperforms other models, including the calibrated SPFs. The best model describes the mean crash frequency as a function of annual average daily traffic, segment length, lane width, year, and speed limit. The study finds that the HSM-recommended method for calibration factor estimation also performs well.

Cafiso S., Di Silvestro G., Di Guardo G.

In the framework of the European Union (EU) Directive on Road Infrastructure Safety Management, this paper will focus on ‘ranking of high accident concentration sections’ by analyzing the opportunity to adopt the Highway Safety Manual (HSM) approach. With this goal, a pilot project of network screening for Italian motorways using HSM is presented. The more appropriate HSM procedures are selected and calibrated that take into account Italian national factors. Results are evaluated using Safety Performance Functions (SPF) as a comparison that is developed directly with data for the specific Italian road infrastructure, which is assumed to provide more reliable estimates than HSM calibration.

Sacchi E., Persaud B., Bassani M.

The Highway Safety Manual (HSM) provides an algorithm and associated knowledge to predict crashes on different types of facilities. This algorithm requires calibration to current local conditions through a procedure prescribed in the HSM to enhance its transferability. However, no procedure assesses the transferability. To fill this void, this paper focuses on a methodology to assess the transferability of the key HSM algorithm components—the baseline safety performance function and the crash modification factors (CMFs)—and uses the Italian road network as an illustrative case study. The calibration of the HSM crash prediction model is investigated with a data set for two-lane two-way rural highways to demonstrate some tools that could be used by jurisdictions around the world to assess the validity and compatibility of the CMFs and the base models, as well as the performance of the complete algorithm. A comparison with the results from a similar study carried out in Canada is provided to supplement the conclusions on the transferability of the HSM algorithm outside the United States.

Brimley B.K., Saito M., Schultz G.G.

This paper documents the calibration of the Highway Safety Manual (HSM) safety performance function (SPF) for rural two-lane two-way roadway segments in Utah and the development of new SPFs through negative binomial regression. Crash data from 2005 to 2007 on 157 selected study segments in Utah provided a 3-year frequency of observed crashes to calibrate the HSM SPF and develop new models. The calibration factor for the HSM SPF for rural two-lane two-way roads in Utah is 1.16, indicating that the original HSM model under predicts crashes in Utah. The HSM suggests that jurisdiction-specific SPFs may predict crashes with greater reliability than calibrated SPFs. The following variables were significant in each of the four models developed by this research: annual average daily traffic (AADT), segment length, speed limit, and the percentage of AADT composed of multiple-unit trucks. AADT and segment length are used in the HSM SPF; speed limit and the percentage of AADT composed of multiple-unit trucks were found to correlate significantly with observed crash frequencies. The fourth negative binomial model developed in the study would be the best SPF to predict crashes on rural highways in Utah. As encouraged by the HSM and contemporary research, the empirical Bayes method can be applied with each jurisdiction-specific SPF because the analysis provided an overdispersion parameter for each model.

Xie F., Gladhill K., Dixon K.K., Monsere C.M.

The recently released Highway Safety Manual (HSM) published by AASHTO provides a comprehensive set of tools for evaluating and identifying opportunities to improve safety for highway facilities. Included in the HSM is a quantitative method for predicting crashes on the basis of recently developed scientific approaches. These predictive methods currently exist for three facility types: rural two-lane roads; rural multilane highways; and urban and suburban arterial highways. To enhance precision, each HSM predictive method should be calibrated for location conditions. This paper demonstrates the HSM calibration procedure for total crashes in Oregon. The research identified three critical data collection limitations on information about pedestrian volumes, minor road traffic volumes at rural locations, and minimum sample size for underrepresented crash locations. Most of the calibration factors for Oregon were determined to be considerably lower than the expected value of approximately 1, and this observation was attributed to Oregon crash reporting thresholds and procedures. The paper includes an evaluation of crash severity distribution methods and an assessment of the significance of collision type distributions on the overall predicted crashes.

Pathivada B.K., Banerjee A., Haleem K.

Zhang C., He J., Wang H., Ye Y., Yan X., Wang C., Zhang X.