Crash Patterns Research Articles

Prevalence and Pattern of Road Traffic Accidents among Commercial Motorcyclists in the Central Tongu District, Ghana.

Background The World Health Organisation estimates that 1.35 million people die as a result of road traffic crashes. Motorcycles as a means of transport are increasingly becoming the preferred and easiest means of transportation for most people in developing countries despite the associated risk. This study determined the prevalence and pattern of motorcycle crashes in Adidome among commercial motorcyclists. Methods A descriptive, cross-sectional study design was used as 114 commercial motorcyclists were recruited to respond to a pretested research questionnaire in the Adidome district of the Volta Region. Data were analyzed using SPSS, version 22.0. Data were presented as simple descriptive statistics. A chi-square relationship was determined using the demographic variables, and the history of accident at a 95% confidence interval with 0.05 was considered as statistically significant. Results The prevalence of road traffic crashes at Adidome was 64.0%. Motorcyclists (74.0%) were reported to have been involved in crashes in the past one year prior to the study. Motorcyclists attributed the last accident to excessive speeding (31.5%) and bad roads (23.3%), this accident as a result of colliding with another motorcycle (50.7%), and slippery surfaces (24.7%). The majority (63.0%) of the respondents had an accident once. The consumption of alcohol was associated with the occurrence of an accident as 34.2% occurred among cyclists who drank alcohol, compared with 29.8% who did not (p < 0.05). Conclusion There should be strict implementation of current road traffic regulations of Ghana by the MTTD of the Ghana Police Service, and penalties should be awarded against anybody caught riding a motorcycle under the influence of alcohol. Helmet and other protective devices must be made compulsory for motorcycle riders to prevent injuries, especially head injuries, if an accident occurs.

Safety of micro-mobility: Analysis of E-Scooter crashes by mining news reports

Dockless electric scooters (E-Scooters) have emerged as a popular micro-mobility mode for urban transportation. This new form of mobility offers riders a flexible option for massive first-/last-mile trips. Despite the popularity, the limited regulations of E-Scooters raise numerous safety concerns among the public and agencies. Due to the unavailability of well-archived crash data, it is difficult to understand and characterize current state quo of E-Scooter-involved crashes. This paper aims to shorten the gap by analyzing a set of reported crash data to describe the patterns of crashes related to E-Scooter use. Specifically, massive media reports were searched and investigated for constructing the crash dataset. Key crash elements such as rider demographics, crash type, and location were organized in an information table for analysis. From 2017 to 2019, there were 169 E-Scooter-involved crashes identified from the news reports across the country. Through the descriptive analysis and cross tabulation analysis, the distinct characteristics of these reported crashes were highlighted. Overall, there was a growing trend for the reported E-Scooter-involved crashes unevenly distributed among the States. The distribution of the crashes across different groups of users, facilities, time periods, and severity levels also showed skewed patterns toward a subset of categories. The quantitative analyses also provide some supportive evidences for warranting the discussion on key issues, including helmet use, riding under influence (RUI), vulnerable riders, and data deficiency. This study highlights the importance of public awareness and timely developing safety countermeasures to mitigate crashes involving E-Scooters.

The impact of a new casino on the motor vehicle crash patterns in suburban Maryland

BackgroundMany states have legalized casino gambling, and casinos create increased vehicle traffic, but the strength of the association between casino construction and vehicle crashes is unknown. MethodsRetrospective analyses of motor vehicle crashes (MVCs) occurring within Anne Arundel County, Maryland (2010–2014) were conducted. The ratio of crashes within one mile of the casino’s location after it was opened were compared to the ratio occurring in the same area before it was opened to determine how the incidence of MVCs near the casino changed with time. Logistic regression was used to determine how crash characteristics may have influenced the incidence of MVCs near the casino after it opened. Results101,860 persons were involved in 43,328 MVCs in Anne Arundel County during the study period; 29,476 (68.0 %) had an at-fault driver ≥21 years of age and complete data. MVCs proximal to the casino occurred most commonly during the day (N = 421, 76.6 %) and involved drivers <40 years of age (N = 366, 66.6 %) and male (N = 316, 57.4 %). After adjustment for impairment and day of the week, there was a significant association with crashes close to the casino after it opened (ORAdjusted = 1.23, 95 % CI: 1.04–1.46, p = 0.02). Crashes occurring close to the casino, after it opened, involved drivers <40 years of age (OR = 1.74, 95 % CI:1.45–2.08) and occurred on weekends (OR = 1.39, 95 %CI:1.15–1.67). ConclusionsIn this single-site study the opening of a casino was associated with an increase in crashes nearby. The generalizability of this finding should be confirmed with analysis of MVC data near other gambling venues.

Spatio-temporal patterns of air crashes in Nigeria between 1960 and 2012

Air transport is considered the safest among the modal transport options. Over the years the passenger travel by air has double necessitating stricter national and international regulatory framework. In spite of the concerted efforts toward ensuring safety on the global scale, incidents of air crashes are still prevalent especially in the developing countries like Nigeria. The Nigerian aviation sector has suffered series of air mishaps since independence in 1960. This study seeks to examine the spatio- temporal pattern of air crashes in Nigeria. The methodology involves longitudinal research design covering over 30 years, from 1960 to 2012. These cut across many segments such as location, factor responsible for the crash, number of survivor, number of death, number of accident per year and number of carrier involved. The data is analysed with descriptive analytical techniques. The study reveals the classification of causal air crashes in Nigeria as human, mechanical, environmental and mixed themes. The study also reveals that the pattern of air crashes in Nigeria follows a cyclic pattern of occurrence over the years. The implication is that stability of the sector against accident is still far-fetched. Therefore more investment is still needed in the sector to achieve a healthier and danger free airspace.

A Closer Look at Highway Safety Diagnostics and Crash Analysis

Diagnostic methods in the context of the Highway Safety Manual (HSM) aim to identify abnormalities or patterns in crash occurrence which may provide an important clue to an effective remedy. These methods are used to assess the nature of the safety problem and on its basis select a countermeasure for which there is a known crash modification factor. Two approaches are presently used to perform diagnostic examinations: the test of proportions and crash type-specific safety performance functions (SPFs). The test of proportions uses empirical proportions (diagnostic norms) of crash types and crash attributes within congestion strata, the total number of crashes, and the number of observed crashes of a specific type to compute the cumulative probability of the observed outcome. Crash type-specific SPFs are used to identify sites at which the observed frequency or severity of specific crash types is higher than expected. This paper examines the strengths of both methods by applying them to the same datasets and comparing the results. The findings suggest that the test of proportions works well in identifying locations with a single disproportionally frequent crash type. Crash type-specific SPFs can identify locations having multiple crash types with an elevated frequency, but will not identify locations having crash patterns susceptible to correction but not having elevated counts. Findings also suggest that the degree of stratification in diagnostic norms influences the number of overlapping sites detected by both methods.

In-depth approach for identifying crash causation patterns and its implications for pedestrian crash prevention

IntroductionA pedestrian crash occurs due to a series of contributing factors taking effect in an antecedent-consequent order. One specific type of antecedent-consequent order is called a crash causation pattern. Understanding crash causation patterns is important for clarifying the complicated growth of a pedestrian crash, which ultimately helps recommend corresponding countermeasures. However, previous studies lack an in-depth investigation of pedestrian crash cases, and are insufficient to propose a representative picture of causation patterns. Method: In this study, pedestrian crash causation patterns were discerned by using the Driving Reliability and Error Analysis Method (DREAM). One hundred and forty-two pedestrian crashes were investigated, and five pedestrian pre-crash scenarios were extracted. Then, the crash causation patterns in each pre-crash scenario were analyzed; and finally, six distinct patterns were identified. Accordingly, 17 typical situations corresponding to these causation patterns were specified as well. Results: Among these patterns, the pattern related to distracted driving and the pattern related to an unexpected change of pedestrian trajectory contributed to a large portion of the total crashes (i.e., 27% and 24%, respectively). Other patterns also played an important role in inducing a pedestrian crash; these patterns include the pattern related to an obstructed line of sight caused by outside objects (9%), the pattern that involves reduced visibility (13%), and the pattern related to an improper estimation of the gap distance between the vehicle and the pedestrian (10%). The results further demonstrated the inter-heterogeneity of a crash causation pattern, as well as the intra-heterogeneity of pattern features between different pedestrian pre-crash scenarios. Conclusions and practical applications: Essentially, a crash causation pattern might involve different contributing factors by nature or dependent on specific scenarios. Finally, this study proposed suggestions for roadway facility design, roadway safety education and pedestrian crash prevention system development.

Factor Identification and Prediction for Teen Driver Crash Severity Using Machine Learning: A Case Study

Crashes among young and inexperienced drives are a major safety problem in the United States, especially in an area with large rural road networks, such as West Texas. Rural roads present many unique safety concerns that are not fully explored. This study presents a complete machine leaning pipeline to find the patterns of crashes involved with teen drivers no older than 20 on rural roads in West Texas, identify factors that affect injury levels, and build four machine learning predictive models on crash severity. The analysis indicates that the major causes of teen driver crashes in West Texas are teen drivers who failed to control speed or travel at an unsafe speed when they merged from rural roads to highways or approached intersections. They also failed to yield on the undivided roads with four or more lanes, leading to serious injuries. Road class, speed limit, and the first harmful event are the top three factors affecting crash severity. The predictive machine learning model, based on Label Encoder and XGBoost, seems the best option when considering both accuracy and computational cost. The results of this work should be useful to improve rural teen driver traffic safety in West Texas and other rural areas with similar issues.

Examining the Environmental, Vehicle, and Driver Factors Associated with Crossing Crashes of Elderly Drivers Using Association Rules Mining

In the aging society, reducing vehicle crashes caused by elderly drivers has become a crucial issue. To find effective methods to reduce these vehicle crashes, it is necessary to give some insights into the characteristics of vehicle crashes and those of traffic violations caused by elderly drivers. However, multiple significant factors associated with crossing crashes due to elderly drivers were not extensively observed in previous studies. To fill this research gap, this study identifies the crash pattern and examines the environmental, vehicle, and driver factors associated with crossing crashes due to elderly drivers. The 5-year crash data in Toyota City, Japan, are used for empirical analysis. The emerging data mining method called association rules mining is applied to discover various factors associated with crossing crashes of elderly and nonelderly drivers, respectively. The significant findings indicate that (1) elderly drivers are more likely to lead to crossing or right-turn crashes, compared with nonelderly drivers; (2) there are more factors including crash location (intersection without signal), lighting (daylight), road condition (dry and other), weather condition (clear and raining), vehicle type (light motor truck), and traffic violation (fail to confirm safety) associated with the large proportion of crossing crashes due to elderly drivers. The findings of this study can be used by traffic safety professionals to implement some countermeasures to reduce the crossing crashes due to elderly drivers.

Spatial analysis of road crashes involving vulnerable road users in support of road safety management strategies

On urban road networks, approximately 2 out of 3 fatalities involve pedestrians, cyclists and motorcyclists, collectively referred to as “vulnerable road users” (VRU) due to their insufficient physical protection in the event of a collision. For a safer and more sustainable road transportation system, adequate protective countermeasures need to be introduced for this user category. However, related (and limited) resources restrict any safety improvements to certain high-risk sites with elevated rates of road traffic collisions. This study reports the results of a spatial distribution analysis of traffic collisions involving VRU in Turin over the period 2006-2016. The traffic road collisions database from the Italian National Institute of Statistics (ISTAT) was used for this purpose. Crash data were firstly geo-localized, and then analyzed using Geographic Information System technologies. A cluster analysis and a Kernel Density estimation were used to build spatial patterns of crashes involving VRU. Hazardous sites were identified on a metropolitan scale. Incorrect estimates of the actual collision frequency, which are typical of studies conducted over short periods, were avoided by considering only those sites where collision rates remained significantly high throughout the entire observation period (eleven years). The results show that clusters occur at intersections, many of which are located along corridors affected by heavy traffic flows and wide cross-sections. A further analysis was conducted to explain the role played by the geometric configuration (layout) of most hazardous sections and intersections in the level and severity of injuries and fatalities.

Identifying Patterns of Pedestrian Crashes in Urban Metropolitan Roads in India using Association Rule Mining

Pedestrian safety is an important component of efforts to prevent road traffic injuries. Pedestrians constitute to around 22% of the total deaths occurring on the world roads. According to the recent report by the Ministry of Road Transport and Highways (MoRTH), Government of India, the number of pedestrian-related deaths was 15,746 (10.5%) of total persons killed in the country during the year 2016. This high proportion of mortality and severity injury among pedestrians necessitates more investigation to identify determinants to reduce crashes in the future. The present research used the Apriori algorithm of supervised association rule mining to identify the patterns of pedestrian severity injury in urban Indian metropolitan city, Chennai. Using the RADMS database of the government of Tamilnadu, vehicle-pedestrian crashes were analyzed for recent two years between 2015 and 2016. The results highlight the fact that middle-aged pedestrians are more vulnerable to road traffic crashes. Exceeding speed limits than the posted speed, especially in the highways, results in fatal crashes among pedestrians. Vehicle-pedestrian crashes are frequent at sites where there are no median separators when drivers do not respect the right of way rules. The findings of the present study will help the traffic safety professionals to understand patterns of crashes and take necessary countermeasures to decrease pedestrian injury-related crashes potentially.

Traffic crash pattern modification as a result of a 30 km/h zone implementation. A case study in Turin (Italy)

The strategy of 30 km/h zones, referred to in the international context as "traffic calming" measures, serves to safeguard pedestrians, cyclists and motorcyclists, collectively referred to as “vulnerable road users” (VRU). Its main function is to compel drivers to observe a maximum speed limit of 30 km/h. However, urban infrastructure transformations modify traffic collision patterns and the involvement of road users, with a spatial temporal redistribution of events. This work seeks to study the effects on collision distribution resulting from the introduction of a 30 km/h zone to the Mirafiori Nord area in the city of Turin in late 2008. Collision frequencies, based on data provided by the Italian National Institute of Statistics (ISTAT), were evaluated over the period 2006-2016. Road traffic collisions involving both VRU and motorized users (“noVRU”) were taken into account. Decreases in collision frequency were found for noVRU related crashes, while the VRU crash rate remained essentially unchanged with only minor fluctuations consistent with the regression to the mean phenomenon. The countermeasures, which sought to protect VRU, were however very effective for noVRU. As the effects of each structural modification spill over into neighboring areas, the analysis of collision frequency was extended to a study area greater than the one in which the 30 km/h zone was realized. In fact, due to the migration of events, the reduction in the collision frequency in the speed restricted zone was accompanied by an increase in the same frequency in the immediate surrounding area.

Dynamic monitoring of software use with recurrent neural networks

User interaction with a software may be formalized as a sequence of actions. In this paper we propose two methods – based on different representations of input actions – to address two distinct industrial issues: next action prediction and software crash risk detection. Both methods take advantage of the recurrent structure of Long Short Term Memory neural networks to capture dependencies among our sequential data as well as their capacity to potentially handle different types of input representations for the same data. Given the history of user actions in the interface, our first method aims at predicting the next action. The proposed recurrent neural network outperforms state-of-the-art proactive user interface algorithms with standard one-hot vectors as inputs. Besides, we propose to feed the LSTM with actions embeddings. This continuous representation performs better than one-hot encoded vector LSTM and its lower dimension reduces at the same time the computational cost. Using the same data set, the second method aims at crash risk detection. To address this task, we propose to use feature vectors composed of actions with above average crash probabilities as inputs of the LSTM – with the idea to take advantage of its ability to learn relevant past information to detect crash patterns. The method outperforms state-of-the-art sequence classification methods. Our approaches are demonstrated on medical imaging software logs from ten different hospitals worldwide, though they might be applied to various user interfaces in a wide range of applications.

Effects of fixed speed cameras on spatio-temporal pattern of traffic crashes: Ankara case

Considering that speeding is the predominant factor in severe crashes, this study aims to investigate the effectiveness of fixed speed camera enforcement on spatio-temporal pattern of traffic crashes in Ankara, Turkey. We investigated 1448 crashes involving fatality and/or injury on the sections with and without fixed speed camera in Ankara for the years of 2009 and 2011. Network-based Kernel Density Estimation Method was used to examine the spatial distribution of crash density pattern for pre and post camera instillation years. Comparison of crash statistics between 2009 and 2011 revealed that the fixed speed cameras installed in 2010 resulted in 23% decrease in the number of crashes involving fatalities and/or injuries, 50% in fatalities, and 27% injuries in the examined regions. Moreover, the fixed speed cameras were found to be significantly effective in their catchment area bands in 7 out of 18 indicators including total crashes, injuries, single vehicle crashes, weekdays crashes, noon crashes, evening crashes, and summer crashes. Consistent with the previous studies, findings showed that the fixed speed cameras were effective, especially in reducing traffic injuries and crashes with fatalities or injuries.

Identify Risk Pattern of E-Bike Riders in China Based on Machine Learning Framework

In this paper, the risk pattern of e-bike riders in China was examined, based on tree-structured machine learning techniques. Three-year crash/violation data were acquired from the Kunshan traffic police department, China. Firstly, high-risk (HR) electric bicycle (e-bike) riders were defined as those with at-fault crash involvement, while others (i.e., non-at-fault or without crash involvement) were considered as non-high-risk (NHR) riders, based on quasi-induced exposure theory. Then, for e-bike riders, their demographics and previous violation-related features were developed based on the crash/violation records. After that, a systematic machine learning (ML) framework was proposed so as to capture the complex risk patterns of those e-bike riders. An ensemble sampling method was selected to deal with the imbalanced datasets. Four tree-structured machine learning methods were compared, and a gradient boost decision tree (GBDT) appeared to be the best. The feature importance and partial dependence were further examined. Interesting findings include the following: (1) tree-structured ML models are able to capture complex risk patterns and interpret them properly; (2) spatial-temporal violation features were found as important indicators of high-risk e-bike riders; and (3) violation behavior features appeared to be more effective than violation punishment-related features, in terms of identifying high-risk e-bike riders. In general, the proposed ML framework is able to identify the complex crash risk pattern of e-bike riders. This paper provides useful insights for policy-makers and traffic practitioners regarding e-bike safety improvement in China.

Crash patterns in two Chinese secondary cities, with comparisons to crash patterns in the United States

Objective: Evaluations of drink-driving crash interventions and crash burden in middle-income countries often rely on assumptions supported by data from the United States. We described crashes using pooled survey data across 2 Chinese agricultural cities in the 600,000 population range and then assessed comparability of selected crash characteristics to those in the United States.Methods: We conducted household interviews on drinking and related harms with representative samples of 1,500 people aged 18 and over in both Jiangshan and Lanxi, China. Near the end of the survey, 2,962 people responded to questions about motor vehicle crashes during the past year. We report survey data weighted to match the population demographically.Results: Across the 2 cities, 28% of adults were licensed drivers and 24% drove a motorized vehicle. Fifty-three respondents (unweighted) reported that they personally were involved in at least one crash in the past year. Among these, 32% (weighted) were driving a car, truck, or bus; 7.5% were driving a motorcycle; 9.5% were pedalcyclists; 17% were pedestrians; and 34% were passengers. Of the crashes, 42% involved just one motorized vehicle. Most crashes (62%) occurred in broad daylight, followed by at dawn or dusk (19%) and at night (19%). Only 4% were single-vehicle nighttime crashes. Someone was injured or killed in 22% of crashes. Respondents thought it likely that at least one driver was drinking alcohol in 21% of the crashes and thought it unlikely in 51%; the remaining 28% were unsure whether anyone was drinking or refused this question. Alcohol involvement was similar in injury and no-injury crashes. Respondents thought a driver had been drinking in 24% of daytime crashes, 34% of crashes at dawn or dusk, and 36% of nighttime crashes. All 3 crashes involving an alcohol-involved pedestrian or pedalcyclist also included an alcohol-involved driver. Respondents said that 40% of the crashes were reported to the police and 40% were not reported; the remaining 20% of respondents did not know whether anyone reported their crash. Among crashes where reporting status was known, all crashes with injuries were reported.Conclusions: Compared to published data, crash rates are similar among licensed drivers in the United States and these Chinese cities. The percentage of crashes that involved injury did not differ significantly between the 2 countries. Injury crashes were well reported in both. Crashes involving property damage only were significantly less likely to be reported to the police in the Chinese cities. Alcohol involvement rates in crashes were similar. Although our crash sample was small, some crash parameters appear to be transferable between these 2 countries.

Examination of vehicle impact against stationary roadside objects

In this paper, a model of a car impact against a fixed roadside objects/safety barrier is presented and discussed to solve a mixed task: to determine the pre-impact velocity of the vehicle center of mass taking into consideration the recorded post-impact final rest position. The principle of conservation of mechanical energy was applied, whereas the energy of the rigid barrier deformation was estimated by the finite element method/FEM for an elasto-plastic body. FEM analysis was performed by integrating SolidWorks - SW Simulation software, the output data being the deformation of the safety barrier in the three dimensional space subjected to limited boundary conditions. The post-impact car motion was modelled by a multiscale spatial model, well established in modern accident reconstruction. The findings of the study point out the precision and accuracy of the crash pattern revealed in the accident reconstruction and essential in clarifying the situation with appropriate real facts. Two cases involving real-world head-on accident collisions against a fixed flexible barrier and a tank body are considered, which exemplify the adopted method.

A clusterwise regression approach for the estimation of crash frequencies

In the current literature, data is aggregated for the estimation of functions to explain or predict crash patterns using either clustering analysis, regression analysis, or stage-wise models. Typically, analysis sites are grouped into site subtypes based on predefined characteristics. The assumption is that sites within each subtype experience similar crash patterns as a function of prespecified explanatory characteristics. To develop functions to estimate crashes, all data points are clustered only as a function of associated site characteristics. As a consequence, estimated parameters may be based on different crash patterns that represents various trends that could be better captured by using multiple functions. To address this limitation, this study proposes a mathematical program utilizing clusterwise regression to assign sites to clusters and simultaneously seek sets of parameter values for the corresponding estimation functions, so as to maximize the probability of observing the available data. A simulated annealing, coupled with maximum likelihood estimation, was used to solve the mathematical program. Results were analyzed for two site subtypes with fatal and all injury crashes: (1) roadway segments for urban multilane divided segments and (2) urban four-leg signalized intersections. Clusterwise regression improved the predicted number of crashes with multiple estimation functions within the same site subtype.

Exploratory spatial analysis of traffic crashes, road mortality and morbidity in Nigeria

AbstractThis study examined the spatial distribution of road traffic crashes, mortality and morbidity in Nigeria for the period 2007–2017. The influence of exposure, road network and socio‐demographic characteristics on fatalities and injuries was analysed. Data were obtained from Federal Road Safety Corps and Annual Abstract of Statistics. Spatial statistical techniques were employed to determine the pattern of crashes, mortality and morbidity. Ordinary least squares and geographically weighted regression models were used to identify significant variables. Crashes and mortality declined by 47.9 per cent and 46.3 per cent respectively, while morbidity increased by 78.6 per cent. Crashes, mortality and morbidity were significantly clustered and high in Niger, Kaduna, FCT, Kogi and Nassarawa States. FCT recorded the highest crashes and injured persons, while Kaduna recorded the highest fatalities. Variables that significantly influenced spatial variation of mortality and morbidity were length of asphaltic concrete, surfaced dressed and earth roads, unemployment and population density. This study shows that type of road is a major factor that contributes to road crash mortality and morbidity. Crash countermeasures such as implementation of safety standards during road construction, speed reduction strategies and enforcement of seat belt use and child restraint should be implemented.

A detailed spatiotemporal analysis of traffic crash hotspots

A number of traffic crash databases at present contain the precise positions and dates of these events. This feature allows for detailed spatiotemporal analysis of traffic crash patterns.We applied a clustering method for identification of traffic crash hotspots to the rural parts of primary roads in the Czech road network (3,933 km) where 55,296 traffic crashes occurred over 2010 – 2018. The data were analyzed using a 3-year time window which moved forward with a one-day step as an elementary temporal resolution. The spatiotemporal behavior of hotspots could therefore be analyzed in great detail.All the identified hotspots, during the monitored nine-year period, covered between 6.8% and 8.2% of the entire road network length in question. The percentage of traffic crashes within the hotspots remained stable over time at approximately 50%. Three elementary types of hotspots were identified when analyzing spatiotemporal crash patterns: hotspot emergence, stability and disappearance. Only 100 hotspots were stable (remained in approximately the same position) over the entire nine-year period. This approach can be applied to any traffic-crash time series when the precise positions and date of crashes are available.

Pattern of road traffic crash hot zones versus probable hot zones in Tunisia: A geospatial analysis

Focusing on how hot zones mapping can predict spatial patterns of crashes and how different mapping approaches compare can help to better inform their application in practice. This study examines the stability of the performance of two spatial autocorrelation measures on the basis of a Road Safety Risk Index (RSRI) through the comparison of the results for three regions (North-West, Center-East, and Center-West) and for three time periods (2002–2005, 2006–2009 and 2010–2013) in Tunisia. Our study differs from others in that it discusses the identification of probable hot zones and enhances the capability to examine a given highway by determining “dangerous probable lengths”, which aims to anticipate the traffic crashes in the future. The identified hot zones and probable hot zones exhibit different regional and temporal characteristics. There are clearly some outstanding spatial clusters of crashes covering specific locations. In both Northwest and Center-West regions, the majority of the identified hot zones and probable hot zones predominantly occur along mainly highways characterized by a dominant rural character. In the Center-East region, both hot zones and probable hot zones are mostly spread northeast and south-west more precisely in NH1 and NH2 where many urban activities are taking place. Spatial autocorrelation indices per region address the diversity within the regions and provide us with useful insights that can be translated into safety policies in Tunisia.