Sara Ferreira

A note on modeling road accident frequency: A flexible elasticity model

Count data models and their variants have been widely applied in accident modeling. The traditional log-linear function is used to represent the relationship between explanatory variables and the dependent variable (accident frequency). However, this function assumes constant elasticity for the estimation parameters, which is a limitation in the analysis of the effects of explanatory variables on accident risk. Although interaction effects between explanatory variables have been studied in the road safety context (where they are normally assessed by logistic regression), no one has yet examined the possibility of using a flexible function form allowing non-constant elasticity values. This paper seeks to explore the use of the translog function usually used in the economics context to allow the elasticity to vary with the values of other explanatory variables. Therefore, the objective of this study was to evaluate the application of the translog function to accident modeling and to compare the results with those of the traditional log-linear function negative binomial (NB) model. The results show that, in terms of goodness-of-fit statistics and residual analysis, the NB model with the translog function performs better than the traditional NB model. Additional evaluations in terms of predictive performance, hotspot identification and uncertainty associated with the estimated values were taken into account. Although this study is exploratory in nature, it suggests that the translog function has considerable potential for modeling accident observations. It is hoped that this novel accident modeling methodology will open the door to the reliable interpretation and evaluation of the influence of explanatory variables on accident frequency.

Categorical modeling to evaluate road safety at the planning level

The most efficient strategy to ensure long-term road network safety is to integrate safety analysis into the planning process of a network or a corridor. Safety planning decision-support tool outcomes should be reliable and realistic, taking into account the main characteristics of this particular level, which is characterized by scant and generalized data. However, the tools developed and presented in previous studies are based on models with a quantitative response. To develop a more suitable tool while maintaining a measure assessment character, this work presents a qualitative response model whose outcome is the risk of occurring three degrees of hazards: low, medium, and high. In this study, an ordered probit model was applied to an urban road network using Porto city data. Hazard categories were defined using accident frequency to reflect a measure of the safety of the road network studied. The developed model provides a safety risk analysis considering road data that are easy to gather or estimate at the planning level. In addition, an analysis of hypothetical scenarios for two sample segments is presented to illustrate an application of the categorical model in typifying an accident risk analysis at a planning level.

Risk factors affecting injury severity determined by the MAIS score

ABSTRACT Objective: Traffic crashes result in a loss of life but also impact the quality of life and productivity of crash survivors. Given the importance of traffic crash outcomes, the issue has received attention from researchers and practitioners as well as government institutions, such as the European Commission (EC). Thus, to obtain detailed information on the injury type and severity of crash victims, hospital data have been proposed for use alongside police crash records. A new injury severity classification based on hospital data, called the maximum abbreviated injury scale (MAIS), was developed and recently adopted by the EC. This study provides an in-depth analysis of the factors that affect injury severity as classified by the MAIS score. Method: In this study, the MAIS score was derived from the International Classification of Diseases. The European Union adopted an MAIS score equal to or greater than 3 as the definition for a serious traffic crash injury. Gains are expected from using both police and hospital data because the injury severities of the victims are detailed by medical staff and the characteristics of the crash and the site of its occurrence are also provided. The data were obtained by linking police and hospital data sets from the Porto metropolitan area of Portugal over a 6-year period (2006–2011). A mixed logit model was used to understand the factors that contribute to the injury severity of traffic victims and to explore the impact of these factors on injury severity. A random parameter approach offers methodological flexibility to capture individual-specific heterogeneity. Additionally, to understand the importance of using a reliable injury severity scale, we compared MAIS with length of hospital stay (LHS), a classification used by several countries, including Portugal, to officially report injury severity. To do so, the same statistical technique was applied using the same variables to analyze their impact on the injury severity classified according to LHS. Results: This study showed the impact of variables, such as the presence of blood alcohol, the use of protection devices, the type of crash, and the site characteristics, on the injury severity classified according to the MAIS score. Additionally, the sex and age of the victims were analyzed as risk factors, showing that elderly and male road users are highly associated with MAIS 3+ injuries. The comparison between the marginal effects of the variables estimated by the MAIS and LHS models showed significant differences. In addition to the differences in the magnitude of impact of each variable, we found that the impact of the road environment variable was dependent on the injury severity classification. Conclusions: The differences in the effects of risk factors between the classifications highlight the importance of using a reliable classification of injury severity. Additionally, the relationship between LHS and MAIS levels is quite different among countries, supporting the previous conclusion that bias is expected in the assessment of risk factors if an injury severity classification other than MAIS is used.

Linking Police and Hospital Road Accident Records How Consistent Can It Be

This paper presents the description of various steps to be applied in the linkage of road traffic accident records through a case study from the city of Porto, Portugal. The complexity of this process stems primarily from several issues found in the data sets: mistakes and missing values were frequently detected, and only a few common data fields could be matched by the linkage process. This study used the application of a mixed deterministic and weight-based probabilistic method to link police and hospital records. The tolerance calibration and weights computation were critical for the final linkage rate as well as for the correct matching of the results. The results obtained lay within the range of rates found by other authors. Furthermore, to improve the record linkage results, a validation process based on the emergency ambulance data was performed. Despite missing values, 98% of the matched records were verified as true matches. Finally, a preliminary investigation of bias after data linkage is described; it shows that the variables selected for comparison indicate similar statistical values. The main outcome of this study is a road accident linkage process that can be adapted, developed, and applied in different contexts and that aims to promote development of police, hospital, and emergency ambulance data in Portugal and other countries. Additional development is planned for each step presented in this paper.

Hot-Spot Identification: Categorical Binary Model Approach

An alternative methodology is presented for hot-spot identification based on a probabilistic model. In this method, the ranking criterion for hot-spot identification conveys the probability of a sites being a hot spot or not being a hot spot. A binary choice model is used to link the outcome to a set of factors that characterize the risk of the sites under analysis on the basis of two categories (0/1) for the dependent variable. The proposed methodology consists of two main steps. After a threshold value for the number of accidents is set to distinguish hot spots from safe sites (Category 1 or 0, respectively), a binary model based on this classification is applied. This model allows the construction of a site list ordered by using the probability of a sites being a hot spot. In the second step, the selection strategy can target a fixed number of sites with the greatest probability or all sites exceeding a specific probability, such as .5. To demonstrate the proposed methodology, simulated urban intersection data from Porto, Portugal, covering 5 years are used. The results of the binary model show a good fit. To evaluate and compare the probabilistic method with other commonly used methods, the performance of each method is tested by its power to detect true hot spots. The test results indicate the superiority of the proposed method. This method is simple to apply, and critical issues such as assumptions of a prior distribution effect and the regression-to-the-mean phenomenon are overcome. Further, the model provides a realistic and intuitive perspective.

An Analysis of the Injury Severity of Motorcycle Crashes in Brazil Using Mixed Ordered Response Models

ABSTRACT The Brazilian traffic environment has experienced a disproportionate growth in motorcycle use over the last 15 years. Unfortunately the same trend has been observed for crash frequency and severity in the category in part by their relative exposure as well as vulnerability. This study investigates factors that influence the severity of motorcycle accidents in urban streets of Fortaleza. Traditional and mixed ordered logit models were calibrated using a sample of 3,232 observations of traffic accidents from 2004 to 2011. Physical levels of injury inflicted to motorcyclists were grouped as no apparent injury, slight injury, serious injury, and fatal injury. The models were developed using several variables as risk factors. Results suggested that motorcyclists using helmets reduced their chances by 9% of suffering severe and fatal injuries after the crash. Accidents during the daylight as well as on weekdays presented lower risk of resulting in fatal injuries. Also, crashes involving motorcyclists older than age 61 years have 22% more probability of resulting in severe and fatal injuries as compared to young riders. Most of these findings can be associated with commonly reported risky behavior from motorcyclists such as speeding, improper lane changes, and red light running.

Urban Road Planning: A Safety Perspective

The planning process is the first stage to incorporate the safety analysis of the road urban network. This paper provides a tool that allows the analysis of the potential accident risk at urban segments using information available at road planning process. The main points decided in the planning stage - land-use and the road function classification – were used as explanatory variables. Other variables, as control variables, were included such as traffic flow, segment length, minor intersections density and time trend. Generalized linear modeling techniques were used to model accident data from Porto, Portugal. The models outcomes demonstrated that all the variables are statistically significant and according to the expectation. Using the “percentage explained” by the model it can be concluded that although the traffic flow was the strongest variable, the land-use and road function classification have improved the “percentage explained” by the model.

Urban Road Network Safety Model at the Transportation Planning Process

The transportation planning process is considered to be the initial step to incorporate the safety analyses of a transport system. Recently, tools have been created for this incorporation, although the vast majority has been applied at an area level. Despite the capacity of these tools to provide safety information, they are unable to identify hazardous locales in the network. The primary objective of this article was to develop a tool based on accident prediction models, which enable the forecasting of the highest accident-risk entities (i.e., nodes or links) using the information available as well as meeting the main issues that are decided at the planning level. Analyzing the influence of macrolevel road design variables on accident risk was the second objective, providing knowledge for planners and decision makers. To accomplish these objectives, generalized linear modeling techniques were employed to develop models for nodes and links using data from Porto, Portugal. The result analysis demonstrates that the additional macrolevel variables are statistically significant and improves the model. Thus, the results lead to a general conclusion that the proposed models are the suitable tool because they embody the balance between the needs and constraints inherent to the planning level.

Emergency Medical Service Response: Analyzing Vehicle Dispatching Rules

In an era of information and advanced computing power, emergency medical services (EMS) still rely on rudimentary vehicle dispatching and reallocation rules. In many countries, road conditions such as traffic or road blocks, exact vehicle positions, and demand prediction are valuable information that is not considered when locating and dispatching emergency vehicles. Within this context, this paper presents an investigation of different EMS vehicle dispatching rules by comparing them using various metrics and frameworks. An intelligent dispatching algorithm is proposed, and survival metrics are introduced to compare the new concepts with the classic ones. This work shows that the closest idle vehicle rule (classic dispatching rule) is far from optimal and even a random dispatching of vehicles can outperform it. The proposed intelligent algorithm has the best performance in all the tested situations where resources are adequate. If resources are scarce, especially during peaks in demand, dispatching delays will occur, degrading the system’s performance. In this case, no conclusion could be drawn as to which rule might be the best option. Nevertheless, it draws attention to the need for research focused on managing dispatch delays by prioritizing the waiting calls that inflict the higher penalty on the system performance. Finally, the authors conclude that the use of real traffic information introduces a considerable gain to the EMS response performance.

Exploring clinical metrics to assess the health impact of traffic injuries

ABSTRACT In order to allow a deep knowledge of the nonfatal injuries, recently the European Commission adopted the maximum abbreviated injury scale classification which is based on medical diagnosis. This classification will open the door to a new source of information based on international hospital data such as diagnosis-related group and international classification of diseases. In this study, we seek to explore these clinical metrics, which are used to describe the diagnosis and the medical treatment, and to infer consequences of crashes mainly through the costs and severity. Therefore, statistical analyses were applied using generalized linear models selected depending on the type of response variable, i.e. discrete or continuous. Relationships between these metrics were identified revealing for instance that head is the region of the body associated with high severity as well as to higher health care costs. Additionally, a discussion is presented regarding study results and future developments of clinical metrics are pointed out.