Crash Analyses Research Articles

Reliability of Self-Reported Road Crash and Violation Data of Professional Drivers: The Case of Qatar

Having accurate road traffic crashes (RTCs) and violations data is essential for establishing more realistic crash and casualty reduction programs. Although crash data commonly suffer from issues of accuracy and comprehensiveness, these factors are often disregarded in crash analyses. This research aims to evaluate the trustworthiness of self-reported RTCs and violations data in relation to actual record data among professional taxi and bus drivers in the state of Qatar. The analysis will account for divergences across socio-demographic features and driver types. A statistical analysis i.e. Wilcoxon signed-rank test was used on the collected data (one year self-reported and two years actual data) of 566 participants, comprising 361 taxi drivers and 206 bus drivers who were employed by Karwa driving school. The results revealed significant differences between self-reported RTCs and traffic violation data and actual records. The accuracy of self-reporting also varies depending on the type of driver and across various socio-demographic categories. The variations observed in the data have significant inferences for the overall reliability of the data and their influence on crash analysis.

Did COVID-19 mandates influence driver distraction Behaviors? A case study in New Jersey

The global pandemic, Coronavirus (COVID), impacts almost everything, and human driving behavior is no exception. Since its breakdown in March 2020, transportation researchers have pondered on this matter to investigate the impact of COVID on drivers' travel behavior. Most of these studies are self-reported surveys, 'before and during' crash analyses, and short-duration observational studies. With a wider scope from an observational lens, we aim to investigate the impact of COVID-19 on drivers' distraction behaviors from an extensive observational database of distracted driving in New Jersey. A moving observational vehicle equipped with mounted cameras was utilized to collect the drivers' behavior across the high crash corridors and important highways in New Jersey. Samples collected during the COVID mandates and the post-COVID period were analyzed to assess the impact of COVID on drivers' distraction behavior at various temporal setups and roadway geometry. The results indicate a significant increase in cellphone use, eating, and fidgeting/grooming during the post-COVID period. However, behaviors like radio/reaching objects and talking to passengers decreased significantly after COVID. Utilizing the Mann-Whitney U test, this study conducted significance tests to identify the impacts of various temporal and geometric roadway features on each distraction type. Summer season, toll roads and roads with no signal had reduced distraction during the post-COVID compared to the COVID period. Both COVID and post-COVID period showed significant patterns of change of driver behavior within the temporal variables and road geometry. This research's findings would enable comprehension of human behavioral change for COVID mandates.

Application of Macro Element Method (MEM) for faster automotive crash safety design during concept stage

The non-linear nature of vehicle crash safety simulations using the Finite Element Method (FEM) usually requires a significant investment in computational time and resources, in addition to the requirement of full CAD (Computer Aided Design) data availability, which is challenging to obtain during the concept stage. The current work explores the usage of Macro Element Method (MEM) based on simplified Super Beam Element (SBE), for which the shape functions are developed based on experimental observations rather than algebraic equations. MEM crash analysis run time doesn’t exceed several seconds and is one of the underutilised analytical models which can be used to quickly explore new design directions during the early concept stage of the vehicle and has higher potential to speed up the product development process. In this paper, initially, a calibration study was done to validate the MEM at a component level to compare the results with FEM and experimental values. FEM and MEM simulation models were developed and evaluated for a metallic thin-walled square frusta with varying semi-apical angles subjected to axial compression. The results were compared, and it was found that MEM results show more than 97% correlation compared to FEM and experimental results. Subsequently, a design modification was introduced in the component and the results show close degree of correlation between MEM and FEM. Similar process was followed for full vehicle level simulations, such as full width frontal crash, side pole crash analyses to understand the robustness of the method in applying it to product development environment.

Adjunctive Damage Model to Describe the Interaction of Different Defect Types in Textile Composites on the Strain-Rate-Dependent Material Behaviour

Textile composites are predestined for crash-loaded lightweight structures due to their adjustable energy absorption capacity, but they can exhibit different types of defects that occur during production (voids) and in operation (fatigue). The influence of such defects, especially the interaction of several defect types on the strain-rate-dependent material behaviour, is still insufficiently researched and can represent a safety risk. Therefore, this paper presents a phenomenological model that can be used to mathematically describe the strain-rate-dependent stress-strain behaviour of nominally defect-free and defect-affected textile composites. An adjunctive damage model in the sense of continuum damage mechanics is introduced, which also considers the interaction of both defect types for the first time. For the model validation, extensive experimental tests on glass fibre non-crimp fabrics reinforced epoxy (GF-NCF/EP) composites are performed. The focus is put on the influence of voids and fatigue-related pre-damage under subsequent tensile loading at strain rates up to 40 s−1. The theoretical studies show a good coincidence with the experimental results. The novel model provides a method for the efficient generation of material maps for numerical highly dynamic crash and impact analyses for defect-free and defective textile composites. As a result, a flexible and practice-oriented model approach is available, which makes a significant contribution to an improved understanding of materials and enables a future defect-tolerant design of textile composites.

Assessing bicycle crash risks controlling for detailed exposure: A Copenhagen case study

A better understanding of factors associated with bicycle crashes can inform future efforts to limit crash risks. Many previous studies have analysed crash risk based on crash databases. However, these can only provide conditional information on crash risks. A few recent studies have included aggregate flow measures in their crash risk analyses. This study incorporates detailed bicycle flow to investigate factors related to bicycle crashes. Specifically, the study assesses the relative crash risk given various conditions by applying Palm distributions to control for exposure.The study specifically investigates the relationship between weather and time conditions and the relative risk of bicycle crashes at a disaggregate level. The study uses bicycle crash data from police reports of bicycle crashes from 2017–2020 in the greater Copenhagen area (N = 4877).The relations between the bicycle crash risk and the air temperature and wind speeds are found to be highly non-linear. The relative risk of bicycle crashes is elevated at low and high temperatures (0 °C ¿ x, x ¿ 21 °C). The results also show how decreasing visibility relates to increasing bicycle crash risk. Meanwhile, cycling during the early morning peak (7–8) and afternoon peak hours (15–18) is related to an increased risk of bicycle crashes. While some of the effects are likely spurious, they highlight specific conditions associated with higher relative risk. Finally, the results illustrate the increased risk at weekend night times when cyclists are likely to bike under the influence of alcohol.In conclusion, the analysis confirms that visibility, slippery surfaces, and intoxication are all factors associated with a higher risk of bicycle crashes. Hence, it is relevant to consider how infrastructure planning and preventive measures can modify the bicycle environment to minimise these risks.

Damage surrogate models for real-time flooding risk assessment of passenger ships

Real-time assessment of flooding risk associated with the collision between two ships, requires a fast estimation of damage dimensions and associated survivability. The state-of-the-art frameworks for risk assessment on passenger ships do not consider a direct evaluation of damages through crash simulations but refer to probabilistic considerations, modelling damage characteristics according to statutory marginal distributions of damage breaches too old to be any longer relevant. In any case, such an approach is not possible for the real-time risk assessment process developed in project FLARE, aimed at promoting the employment of first-principles tools for risk evaluation. In this spirit, the present work investigates the possibility of developing a database-oriented damage breach model, which employs direct crash analyses with the super-element code SHARP. However, the sole usage of crash simulations is not suitable for real-time applications. Therefore, starting from the collision simulation database, surrogate models have to be derived for real-time application. In this specific case, three different strategies have been used for the models creation namely: multiple linear regressions, neural networks and decision trees. Here, the strategy to build the database and application on a reference passenger ship is described, highlighting the differences in accuracy and calculation time between the proposed surrogate models.

A Review of Surrogate Safety Measures Uses in Historical Crash Investigations

Historical road crash data are the main indicator for measuring road safety outcomes. Over the past few decades, significant efforts have been made in obtaining and exploiting Surrogate Safety Measures (SSMs). SSMs have the potential to provide excellent sustainable road safety indicators and proxy measurements which can complement traditional historical crash analyses or even substitute them. By using SSMs, crash data collection demands can be bypassed and areas can be investigated before crashes occur. Due to such advantages, the objective of the present research is to provide a review of the scientific literature regarding studies exploiting SSMs for historical crash record investigations. Specifically, 34 studies were examined, providing insights on the different types of SSMs collected under real road environment conditions, the way they are collected, their connection with specific road crash types, and the type of the developed statistical models are examined and discussed. Particular focus is also placed on the temporal dimension of the collection period of both SSMs and road crashes. Finally, the overall trends deriving from the reviewed studies are summarized and future research directions are provided.

Experimental and numerical investigation of crashworthiness performance for optimal automobile structures using response surface methodology and oppositional based learning differential evolution algorithm

Abstract In this study, experimental and numerical crash analyses are carried out to reach an optimum bumper beam and energy absorber design for a passenger car. Design parameters have been created to determine the most crash-efficient bumper beam and energy absorber models. The models that are formed by using Taguchi tables are subjected to crash analysis, and the responses are obtained to find an optimal design. Response surface methodology is used to approximate the structural responses in crash analysis, and the optimum bumper beam and energy absorber models are obtained by the differential evolution algorithm. The optimum model is subjected to crash analysis in the Hyperform software without considering the sheet metal forming effect. Besides, the model is analyzed by incorporating forming history into the crash analysis. As a result of the numerical analysis, a new energy absorber and bumper beam model with the better crash performance and weight reduction are obtained.

Neural network modelling of mechanical joints for the application in large-scale crash analyses

This paper presents an artificial neural network (NN) modelling approach to represent a connector model in large-scale finite element explicit crash simulations. The NN model was established to describe the local force–deformation response of point connectors in automotive applications, namely self-piercing rivets and flow-drill-screws. The study is limited to two-sheet connections and to the use of feedforward NNs. Successive loading and unloading of the joints is not studied. Various architectures and complexities of the feedforward NNs were evaluated and trained based on data generated from a constraint model found in the literature. This forms a proof of concept for implementing a modelling technique not based on physics-motivated constitutive equations. The impact of the network complexity and training data diversity was investigated. The NN model was implemented as a cohesive zone model for incremental force prediction in an explicit finite element code. In order to have a wide selection of joint types, five different joint configurations including self-piercing rivets (SPR) and flow-drill screws (FDS) were investigated. Numerical results from the NN model were compared to physical tests from all joint configurations. It was shown that a rather basic machine learning technique like a feedforward NN was able to reproduce path-dependent force–deformation behaviour for the application in explicit FE solvers.

Network-Level Safety Metrics for Overall Traffic Safety Assessment: A Case Study

Driving safety analysis has recently experienced unprecedented improvements thanks to technological advances in precise positioning sensors, artificial intelligence (AI)-based safety features, autonomous driving systems, connected vehicles, high-throughput computing, and edge computing servers. Particularly, deep learning (DL) methods empowered volume video processing to extract safety-related features from massive videos captured by roadside units (RSU). Safety metrics are commonly used measures to investigate crashes and near-conflict events. However, these metrics provide limited insight into the overall network-level traffic management. On the other hand, some safety assessment efforts are devoted to processing crash reports and identifying spatial and temporal patterns of crashes that correlate with road geometry, traffic volume, and weather conditions. This approach relies merely on crash reports and ignores the rich information of traffic videos that can help identify the role of safety violations in crashes. To bridge these two perspectives, we define a new set of network-level safety metrics (NSM) to assess the overall safety profile of traffic flow by processing imagery taken by RSU cameras. Our analysis suggests that NSMs show significant statistical associations with crash rates. This approach is different than simply generalizing the results of individual crash analyses, since all vehicles contribute to calculating NSMs, not only the ones involved in crash incidents. This perspective considers the traffic flow as a complex dynamic system where actions of some nodes can propagate through the network and influence the crash risk for other nodes. The analysis is carried out using six video cameras in the state of Arizona along with a 5-year crash report obtained from the Arizona Department of Transportation (ADOT). The results confirm that NSMs modulate the baseline crash probability. Therefore, online monitoring of NSMs can be used by traffic management teams and AI-based traffic monitoring systems for risk analysis and traffic control.

Analysis of Crash Characteristics of Hydrogen Storage Structure of Hydrogen Powered UAV

In the context of green aviation, as an internationally recognized solution, hydrogen energy is lauded as the “ultimate energy source of the 21st century”, with zero emissions at the source. Developed economies with aviation industries, such as Europe and the United States, have announced hydrogen energy aviation development plans successively. The study and development of high-energy hydrogen fuel cells and hydrogen energy power systems have become some of the future aviation research focal points. As a crucial component of hydrogen energy storage and delivery, the design and development of a safe, lightweight, and efficient hydrogen storage structure have drawn increasing consideration. Using a hydrogen-powered Unmanned Aerial Vehicle (UAV) as the subject of this article, the crash characteristics of the UAV’s hydrogen storage structure are investigated in detail. The main research findings are summarized as follows: (1) A series of crash characteristics analyses of the hydrogen storage structure of a hydrogen-powered UAV were conducted, and the Finite Element Analysis (FEA) response of the structure under different impact angles, internal pressures, and impact speeds was obtained and analyzed. (2) When the deformation of the hydrogen storage structure exceeds 50 mm, and the strain exceeds 0.8, an initial crack will appear at this part of the hydrogen storage structure. The emergency release valve should respond immediately to release the gas inside the tank to avoid further damage. (3) Impact angle and initial internal pressure are the main factors affecting the formation of initial cracks.

Bending collapse of optimal arched thin-walled structures

Arched thin-walled structures are quite promising to significantly improve the bending resistance of beams under transverse loading since they can switch transverse forces to axial ones. The crashworthiness of a type of optimal arched structure with circular section (OASCS) is investigated in this work. Static and dynamic three-point bending tests show that OASCS specimens exhibit nearly 5 times the energy absorption and specific energy absorption (SEA) of corresponding straight tubes. Metallic OASCS must be fabricated by some metal forming process, which has an important influence on crashworthiness. Numerical simulations of the tests are then carried out by using LS-DYNA, and the results agree well with the experiment. The fabrication process of OASCS is simulated by LS-DYNA, and the simulation results are in good agreement with the experiment. The forming effects result in a 20% increase in SEA of OASCS. The effects of boundary conditions and structural parameters are also investigated numerically. It is interesting to find that the response of OASCS is insensitive to the span, which is the most critical factor in three-point bending. Crashworthiness optimization and partial multi-cell enhancement are also employed to further improve the performances of OASCS, which leads to about 40% and 60% increases in SEA, respectively. Finally, the application of OASCS to a bumper system is considered in pole crash analyses.

Crash Analysis and Size Optimization of a Vehicle’s Front Bumper System

Bumper beams and crash boxes are structural vehicle elements that have very critical functions during frontal collisions. This study examined the energy ab-sorbed by the front bumper system and the reaction forces to the cage structure representing the vehicle cabin frame during a full frontal crash against a rigid wall. For this purpose, 25 different crash analyses were conducted using the ANSYS software by changing the sheet thickness values of the bumper beam (t1) and crash boxes-frontal crash rails (t2) that make up the front bumper system. The internal energy (absorbed energy) of the front bumper elements from the analysis results and the reaction forces acting on the representative cage struc-ture in the opposite direction to the movement were examined. When the ob-tained data were interpreted according to variable thickness, it was observed that the variable t2 significantly affected both the absorbed energy and reaction force results, while the variable t1 did not have a significant effect on the results. Based on this, the thickness value t2 was determined as the design variable of the size optimization problem. When the optimization problem established with the reac-tion force and absorbed energy constraint functions was solved, the optimal t2 thickness value with the highest absorbed energy / peak reaction force ratio was found.

A finite element-guided mathematical surrogate modeling approach for assessing occupant injury trends across variations in simplified vehicular impact conditions.

A finite element (FE)-guided mathematical surrogate modeling methodology is presented for evaluating relative injury trends across varied vehicular impact conditions. The prevalence of crash-induced injuries necessitates the quantification of the human body's response to impacts. FE modeling is often used for crash analyses but requires time and computational cost. However, surrogate modeling can predict injury trends between the FE data, requiring fewer FE simulations to evaluate the complete testing range. To determine the viability of this methodology for injury assessment, crash-induced occupant headinjury criterion (HIC15)trends were predicted from Kriging models across varied impact velocities (10-45 mph; 16.1-72.4 km/h), locations (near side, far side, front, and rear), and angles (-45 to 45°) and compared to previously published data. These response trends were analyzed to locate high-risk target regions. Impact velocity and location were the most influential factors, with HIC15 increasing alongside the velocity and proximity to the driver. The impact angle was dependent on the location andwas minimally influential, often producing greater HIC15 under oblique angles. These model-based head injury trends were consistent with previously published data, demonstrating great promise for the proposed methodology, which provides effective and efficient quantification of human response across a wide variety of car crash scenarios, simultaneously. This study presents a finite element-guided mathematical surrogate modeling methodology to evaluate occupant injury response trends for a wide range of impact velocities (10-45 mph), locations, and angles (-45 to 45°). Head injury response trends were predicted and compared to previously published data to assess the efficacy of the methodology for assessing occupant response to variations in impact conditions. Velocity and location were the most influential factors on the head injury response, with the risk increasing alongside greater impact velocity and locational proximity to the driver. Additionally, the angle of impact variable was dependent on the location and, thus, had minimal independent influence on the head injury risk.

Trajectory data based freeway high-risk events prediction and its influencing factors analyses

The frequent crash occurrences have caused massive loss of lives and properties all over the world. In order to improve traffic safety, it is vital to understand the relationships between traffic operation conditions and crash risk, and further implement safety countermeasures. Emerging studies have conducted the crash risk analyses using discrete and aggregated traffic data (e.g., loop detector data, probe vehicle data), where crash events were selected as the prediction target. However, traditional traffic sensing data obtained at segment level cannot describe the detailed operation conditions for the vehicle platoons near crash locations. Thus, more microscopic and high-resolution traffic sensing data are needed. In addition, considering the random occurrence feature of crashes, high-risk events should be paid more attentions given their higher occurrence probability and consistent causations with crashes, which could proactively reduce crash likelihood. In this study, HighD Dataset from German highways was utilized for the empirical analyses. First, high-risk events were obtained using safety surrogate measures with Modified Time to Collision (MTTC) less than 2 s. Traffic operation characteristics within 5 s prior to event occurrence were extracted based on vehicle trajectory data. Then, a total of three different logistic regression models were established, which are standard logistic regression model, random-effects logistic regression (RELR) model, and random-parameter logistic regression (RPLR) model. Among which, the RPLR model was showed to have the best fitness and prediction accuracy. The results showed that the disturbed traffic flows in both longitudinal and lateral directions have positive impacts on high-risk events occurrence. Besides, too close following distance between vehicles would lead to high-risk events. Moreover, RPLR models could provide a high prediction accuracy of 97 % for 2 s ahead of the high-risk events. Finally, potential safety improvement countermeasures and future application scenarios were also discussed.

Exploring the Impacts of Driving Environment on Crashes in Tunnel–Bridge–Tunnel Groups: An Eight-Zone Analytic Approach

Tunnel–bridge–tunnel groups (TBTGs) are emerging roads that often involve simple road alignments, but complex driving environments. Investigating crashes occurred in TBTGs is essential for revealing the driving environment–adaptability relationships for such roads. This study seeks to analyze the crash characteristics of component sections in TBTGs with different driving environments and compare the impact of differences in the key factor on the crashes. After TBTGs were defined through a proposed safety-critical distance metric determined via visual theory and actual crash analyses, an eight-zone analytical method considering road types and lighting was developed to probe into crashes in TBTGs. The results show that the proper safety-critical distances for bridge–tunnel and tunnel–tunnel groups are 150 and 500 m, respectively. In TBTGs, the crash rate in ordinary sections is higher than that in bridges and tunnels, particularly in the access zone. The first passed tunnel witnesses a higher proportion of crashes at the access zone and transition zone than the second tunnel. The influence of bridge and tunnel ratios on crashes is related to the ratio and type of bridges and tunnels. The findings presented herein can provide evidence-based guidance for the safety design and management of TBTGs.

Scalable and Accessible Crash Hot Spot Detection for Traffic Law Enforcement

Law enforcement agencies have begun utilizing traffic and crash data to improve traffic law enforcement delivery. However, many agencies often do not have the resources or expertise to harness fully the benefits this data offers. A free to use, scalable traffic crash hot spot detection tool was developed to aid law enforcement agency decision makers, statewide to the local municipality level. The tool was developed to identify crash hot spots algorithmically with a range of customizable parameters based on location, date and time, and crash factors, enabling quick, dynamic queries. These capabilities provide the ability for law enforcement agencies to conduct “what if” analyses and make data-driven allocation decisions, placing officer resources where they are most needed. The two-step algorithm first identifies potential hot spots based on crash density and then ranks each hot spot using a standardized z-score measure of relative significance. To test the viability of the tool, a pilot was conducted identifying 27 hot spots across Wisconsin where targeted enforcement was then deployed. Despite officer skepticism, results from the pilot found officers at sites targeted for speeding and seatbelt violations were nearly twice as likely to initiate traffic stops compared to non-targeted hot spots. Empirical Bayes before-and-after crash analyses found fatal and injury crashes reduced significantly by nearly 11% during the months with targeted enforcement, while property damage crashes and total crashes were unchanged. Overall, the results show the algorithm can identify hotspots where, coupled with targeted enforcement, traffic safety improvements can be made.

Machine learning applied to road safety modeling: A systematic literature review

Road safety modeling is a valuable strategy for promoting safe mobility, enabling the development of crash prediction models (CPM) and the investigation of factors contributing to crash occurrence. This modeling has traditionally used statistical techniques despite acknowledging the limitations of this kind of approach (specific assumptions and prior definition of the link functions), which provides an opportunity to explore alternatives such as the use of machine learning (ML) techniques. This study reviews papers that used ML techniques for the development of CPM. A systematic literature review protocol was conducted, that resulted in the analysis of papers and their systematization. Three types of models were identified: crash frequency, crash classification by severity, and crash frequency and severity. The first is a regression problem, the second, a classificatory one and the third can be approached either as a combination of the preceding two or as a regression model for the expected number of crashes by severity levels. The main groups of techniques used for these purposes are nearest neighbor classification, decision trees, evolutionary algorithms, support-vector machine, and artificial neural networks. The last one is used in many kinds of approaches given the ability to deal with both regression and classification problems, and also multivariate response models. This paper also presents the main performance metrics used to evaluate the models and compares the results, showing the clear superiority of the ML-based models over the statistical ones. In addition, it identifies the main explanatory variables used in the models, which shows the predominance of road-environmental aspects as the most important factors contributing to crash occurrence. The review fulfilled its objective, identifying the various approaches and the main research characteristics, limitations, and opportunities, and also highlighting the potential of the usage of ML in crash analyses.

Convolutional neural networks with refined loss functions for the real-time crash risk analysis

The real-time crash risk analyses were proposed to establish the relationships between crash occurrence probability and pre-crash traffic operational conditions. Given its great application potentials that link with Active Traffic Management System (ATMS) for proactive safety management, it has become an important research area. Currently, researchers mainly developed the real-time crash risk analysis models with traffic flow descriptive statistics employed as explanatory variables and with re-sampled balanced dataset, which hold the limitations of insufficiently capturing the temporal-spatial traffic flow characteristics and failing to provide classification capabilities when deal with the imbalanced datasets. In this study, a Convolutional Neural Network (CNN) modelling approach with refined loss functions has been first time introduced to the real-time crash risk analyses. The primary objectives of the proposed CNN models are: (1) utilizing the tensor-based data structure to explore the multi-dimensional, temporal-spatial correlated pre-crash operational features; and (2) optimizing the loss functions to overcome the low classification accuracy issue brought by the imbalanced data. Data from the Shanghai urban expressway system were utilized for the empirical analysis. And a total of three types of loss functions, including traditional binary cross entropy, the α-weighted cross entropy and the focal loss, were introduced and being tested with varying ratios of crash and non-crash datasets. The modeling results show that the CNN model has better classification performance compared to the traditional Multi-layer Perceptrons (MLP) model with the tensor-based structure data. Besides, the developed CNN model with focal loss function has substantial classification enhancement under the imbalanced datasets. Finally, thedistributions of predicting probabilities for balanced and imbalanced datasets were plotted to understand the effects of the imbalanced dataset and revealed how the proposed CNN model with focal loss function improves the model performance.

Comparison of exposure in pedestrian crash analyses: A study based on zonal origin-destination survey data

In recent years, the importance of exposure for pedestrian crash analysis has received increasing attention. Unlike the case of motor vehicle crashes, the definition of exposure for pedestrian crash analysis is sometimes vague and the mechanism behind the association between exposure and pedestrian crash is often ambiguously defined. In this study, the number of roads crossed and walking distance is estimated using an integrated trip assignment method at the aggregate level. The number of walking trips are also considered and compared with the distance travelled and road crossing based exposure using joint probability models. Results show that models using the road crossing based exposure approach provides the best model fit. It is found that the number of roads crossed is the most sensitive to vehicle-pedestrian collisions, as it is more strongly correlated to potential vehicle-pedestrian interactions. The results also indicate that the sensitivity of number of road crossed could vary with road types, and road safety measures for pedestrian protection should be widely implemented on low-grade roads.