Mapping Road Traffic Crashes in Sfax Governorate (Tunisia)

Sfax is one of the Tunisian governorates with a large number of road accidents, injuries and fatalities every year. This study aimed to analyze and map traffic accidents in this governorate. We analyzed the spatial distribution of accidents, their distribution by cause, by type of road, by size of traffic, by months of the year and days of the week. Accidents were correlated with several variables such as population numbers and densities, motorization rate, length and structure of the road network, and the amount of traffic. On the cartographic level, we have built a database, through which we have produced a series of thematic maps to argue this analysis. Through cartographic pro-duction, we also aimed to help road users, decision-makers and researchers in this area and in the field of transport. This work showed that Sfax occupies, among the other Tunisian governorates, an advanced position in gravity. Various human, climatic and technical factors explained this situation, of which human factors were the most important, and contributed to almost 90% of accidents. The current situation of accidents in Sfax requires a series of measures and actions to alleviate and mitigate the gravity of this phenomenon.

2030, traffic accidents will become the fifth leading cause of death worldwide, with nearly 2.5 million people dying each year on the roads [3].
In Tunisia, as in the other Third World countries, the question of accidentology remains little studied. However, in developed countries, it is well treated.
The work of Vandersmissen and al [4] [5] [6] and Cynthia Boucher [7] in Canada, Abdulhafedh, A. [8] in the USA, Moellering Baker [9], Susan et al [10] in the United States and those of Whitelegg in Great Britain [11], although the angles of analysis are diverse, identify good examples of study in this area.
Like the whole of Tunisia, road traffic accidents, in the governorate of Sfax, are a very serious phenomenon, but have specific characteristics linked to several factors such as urban structure, density, structure and length of the road network, modal distribution of transport flows, traffic density, motorization rate, population weight, human factor, weather and technical factors… In this paper, we discuss the spatial and temporal distribution of accidents in Sfax governorate and the factors determining this phenomenon. We build a database related to the accident, to produce a series of thematic maps and graphs that facilitate and argue the analysis of this phenomenon. In the first part, we present the study area, the tools and the working methodology. In the second part, we analyze the spatial and temporal distribution of accidents by presenting the various explanatory factors. Finally, and through the conclusion, we present a series of findings and recommendations that help decision-makers, stakeholders and researchers in the fields of accidentology, management and development of road networks and road traffic in the governorate of Sfax and in Tunisia.

Study area and Methodology
The study area is the governorate of Sfax which is located in the Central East of Tunisia ( Figure 1). This coastal governorate comprises 16 [17].
To study the issue of accidents, we used statistics, relating to the Sfax governorate of the year 2020, from the National Observatory of Information, Training, Documentation and Studies on Road Safety (NORS). The data relating to accidents in the agglomeration of Sfax comes from the Civil Protection. These statistics are very precise by date (day and time) and by distance from the center of city. The number of accidents, injuries and deceased, covering the entire Tunisian territory, is provided by the NORS in 2020. Data relating to the motorization, traffic and structure of the road network are extracted from the Sfax Atlas, the Sfax Transport Master Plan (1996), the Sfax City Traffic Study (1998), the Transport Master Plan Update Study (2005) and the National Statistical Institute (NSI). Finally, all demographic data are provided by the NSI. All data are entered into a database of a Geographical Information System (GIS). We used Open Journal of Applied Sciences Phildigit for the drawing of base map backgrounds, Philcarto and ArcMap for the creation of thematic maps and Adobe Illustrator for the layout of maps. We also used Microsoft Excel to generate some charts

Results and Discussion
We deal with the issue of accidents in the governorate of Sfax, through the distribution of traffic accidents, injuries and deceased by delegation, type of road, causes, month, day and by means of transport. The variable traffic accidents will be correlated with other variables such as population (number and density), urbanization rate, roads (length and density), traffic intensity and motorization rate.

Spatial Distribution of Traffic Accidents, Injuries and Deceased
"Over the past 30 years, Tunisian roads have recorded varying results in terms of accidents, injuries and deceased. As a result, the number of car accidents increased from 10,209 in 1996 to 10,980 in 2006" to 5972 accidents with 8574 injuries and 1150 deceased in 2020 [18] [19]. The impact of the Covid-19 crisis on all road safety indicators was very significant compared to 2019 [20], with fewer accidents (−20.06%), fewer deceased (−19.05%) and fewer injuries (−21.13%) ( Table 1).
At the national level, the governorate of Sfax ranks fourth in number of accidents, after the governorates of Tunis, Ben Arous and Nabeul, with 258 accidents. This represents 5.4% of accidents in the country as a whole in 2020 ( Figure 2 and Figure 3). The density of accidents in this governorate is relatively low compared to the governorates of Greater Tunis (Tunis, Mannouba, Ariana and Ben Arous). As for the wounded, Sfax also ranks fourth after the capital Tunis, Nabeul and Ben Arous, with 370 wounded (Figure 4). Figure 4 represents almost 6% of all wounded in the country. The rank of Sfax fits perfectly with that relative to the number of accidents, with an average of 1.5 injured for each accident occurred in the governorate. Admittedly, in 2020, Sfax ranks second in number of deaths due to traffic accidents, after the capital Tunis. The damage of the accidents in Sfax is very catastrophic compared to those in most other governorates, that is one death every three days. The number of deaths in Sfax was 88. It represents nearly 10% of all deaths in the national territory ( Figure 5). This can be explained, even in part, by the high number of accidents and by the     A review of Figure 6 and Figure 8 shows that, at the national level, there is a relationship between the distribution of traffic accidents and the distribution of the number of populations. Indeed, coastal governorates, which are characterized by high population numbers and high urbanization rates [13] [22], have high accident numbers. On the other hand, most of the Western and Southern governorates, which have low rates of urbanization and population numbers [23], have low accident numbers. The Central East and North East governorates also have dense road networks and very heavy traffic from which they experience more traffic accidents ( Figure 7) [21].
This relationship between traffic accidents and population, is also expressed in the governorate of Sfax. The correlation coefficient R is 0.87. This relationship is relatively strong in that the coefficient of determination R 2 is 0.78. The number of accidents is also strongly correlated with the rate of urbanization with a correlation coefficient R equal to 0.76, that of the determination R 2 of the order of 0.56.
According to statistics from the NORS, 80% of accidents in the governorate of Sfax are in the Great Sfax and only 20% occurred outside this agglomeration.
The number of accidents is also proportional to the population density. The    The number of accidents is also proportional to the rate of motorization ( Figure 11) with a coefficient of correlation of 0.69, but this relationship is low.
Generally, the delegations of the Greater Sfax (Sfax Madina, Sakiet Ezzit, Sakiet Eddair, Sfax Gharbia, Sfax Sud and Thyna) have high motorization rates. Peripheral delegations have relatively low accident numbers with low rates of motorization. Figure 12 shows that accidents are more related to rates of motorization in urban areas than in rural areas [12] [24].

Distribution of Accidents, Injuries and Deceased by Type of Road
To get a sense of the road effect on accidents, injuries and deceased, it is useful to understand the structure of the road network, which has generally been in decline over the past few years throughout Tunisia and especially in the governorate of Sfax ( Figure 14) [24].
In Sfax, "five categories of roads make up the road network within the governorate of Sfax" [26] [27] (Figure 13 and Figure 15): • Highway: a single highway (A-1) crosses the Sfax governorate and connects Figure 9. Distribution of traffic accidents in the Sfax governorate and the proportion of each delegation of all accidents. Data source [18]. Open Journal of Applied Sciences     [27]. In 2014, the delegations of the Greater Sfax accumulated 345 km of asphalted roads including 65 km of agricultural tracks while the 10 peripheral delegations concern 1765 km (or 83.6% including 751 km of agricultural tracks). The delegation of Menzel Chaker occupies the first place in terms of asphalted tracks (418 km), then, in second position arrives Bir Ali Ben Khlifa (283 km) and then Skhira (190 km). The length of the road network is therefore closely related to the size of the area covered by the peripheral delegations [26] [28]. Road density reached its highest level in the delegations of the Greater Sfax (88 m/km 2 ) in 2009. However, and on the same date, the lowest road density is recorded in the delegation of Menzel Chaker with 150 m/km 2 . The national roads are the busiest, especially the sections that connect Sfax to the other cities of the Northern and Central governorates of Tunisia. The closer one gets to the Greater Sfax and the focal center of the governorate, the regional road density intensifies and the congestion becomes more noticeable [26]. The road density is not balanced between the Greater Sfax region and the rest of the governorate. Generally speaking, the roads that provide mobility in the governorate suffer from under-sizing and a lack of maintenance leading to its degradation. In addition, there is a significant proportion of agricultural runways that are not yet paved. "Mobility problems are concentrated in urban areas, particularly in the Greater Sfax. Transit failure largely explains congestion problems in the governorate" [27]. The number of accidents, injuries and deceased is positively related to road traffic. The greater the traffic, the greater the number of accidents. "The road traffic map in the Sfax Governorate indicates that the maximum daily traffic is typical of the national roads in the Sfax city center (Figure 16). Road traffic reached 22,581 vehicles/day on the section of the NR-1 connecting the city center to Ring Road 11 in the direction of Gabès" [26]. On the same line, between Ring Road 11 and Mahras, traffic drops to 14,925 vehicles/day, then to about 8000 vehicles/day on the rest of the NR-1 line. "Northbound traffic on the NR-1 is greater than southbound traffic and reaches 31,839 vehicles/day at Ring Road 4 and drops sharply to 15230 vehicles/day, just north of the same ring road (4), as NR-1 branches into several regional and local routes to Mahdia, Sousse and Kairouan, reducing the burden" [26].
The distribution of traffic accidents by type of road shows that agglomeration's roads occupy the first place with nearly 65 accidents. This type of road also has the highest number of injured (24 injured). This is largely due to the high density of the road network and the fairly high traffic in the agglomerations. Figure 16. Road traffic in the Sfax governorate. Source [26]. Open Journal of Applied Sciences National roads have the second highest number of accidents (63) and injuries (20), but the highest number of deceased ( Figure 17). They are therefore more dangerous. The importance of traffic on national roads, the rural aspect of this type of road, and the very excessive or inappropriate speeds explain this gravity.
Local roads are the third largest number of accidents, injuries and deceased. This place, relatively advanced, is explained by the length of the network, the importance of traffic and also by the rural aspect of these roads.

Distribution of Accidents, Injuries and Deceased by Cause
A "road accident can have several causes: human, meteorological or technical. Nevertheless, the human factor is paramount and affects more than 9 out of 10 accidents. Human causes come far ahead of other types of causes. The most common human causes are alcohol, speed, fatigue or drowsiness in nearly 8% of fatal accidents" [29]. Distraction is also a human factor in road accidents. The telephone behind the wheel increases the risk of having a bodily accident several times. Non-compliance with safety distances, failure to wear a seat belt, and failure to wear a helmet are also important causes of road accidents, in addition to other factors such as the age, occupation, and wisdom of the driver [17] [30] [31]. Other factors contribute either directly or indirectly to road safety. An example is "urban development that has an influence on economic and social activities and implicitly on traffic and associated risks" [32] [31]. Weather causes can increase the risk of a road accident. However, the weather is almost never the only cause of a road accident. The human factor is almost always present. Weather causes can be: the presence of fog, the lack of visibility caused by a climatic phenomenon such as rain and sand wind. Technical causes can increase the risk of road accidents. But, like weather factors, they are almost never the only causes of a road accident. The technical factor increases the risk of road accidents. These "causes are of two kinds: those relating to the condition of the vehicle and those relating to the condition of the infrastructure (poor road conditions in particular). The purpose of the roadside inspection obligation is to limit road accidents caused by vehicle technical failure" [32] [31].
In Sfax, and like all the other governorates of Tunisia, human factors are the most important. Lack of vigilance ranks first with almost 28% of accidents. Figure 17. Distribution of accidents, deceased and injuries in Sfax Governorate by type of road in 2020. Data source [18]. Open Journal of Applied Sciences Speed is the second factor with 15%, non-compliance with the priority contributes with 8.5% of accidents, collision from the rear, the not driven to the right, crossing the road contribute with 5% for each factor. All accident factors are almost in the same range in terms of injuries and deceased ( Figure 18).

Distribution of Accidents, Injuries and Deceased by Means of Transport
The numbers and proportions of accidents, injuries and deceased differ depending on the mode of transport and the amount of traffic on each utility vehicle.
The car is the most used means to move by road from one point to another in all countries, especially in Tunisia where other modes of transport (rail, maritime, air...) have low contributions. This means is considered the most dangerous means [27].    Light-duty vehicles contribute the largest share of accidents. Four elements contribute together with 80% of the number of accidents (the car 34%, the motorcycle 24%, the van 14% and the pedestrian 10%). The other means and elements contribute together with 20% of the number of accidents (rollover, taxi, heavy truck, obstacle, bike, bus, train, tractor, fall of a passenger, engineering machine) ( Figure 19). As regards the number of injuries and deceased, all these Open Journal of Applied Sciences means of transport keep almost the same ranks as those of accidents with varying proportions. These contributions of the means of transport in the number of accidents, injuries and deceased can even be explained in part, with the importance of the traffic of these means of transport.

Distribution of Accidents, Injuries and Deceased by Month and Day
In 2020 in Sfax, it is in April that accidents are the least frequent (6 accidents) and in October that they are the most numerous (34 accidents). Autumn and summer periods are more marked by more accidents with 73 and 72 respectively. In April, there were only two wounded, but in August and December, there were 50 for each of the two. The highest proportion of injuries (30%) is recorded in the summer, followed by the fall period with 28%, whereas in the spring the lowest proportion is recorded (15%) (Figure 21 and Figure 22). In general, the importance of accidents, in summer and autumn, can be explained by the growth in the number of trips, especially between regions and towards coastal and tourist areas, which reaches its maximum in summer and by the excessive or unsuitable speed.
Thursdays and Wednesdays in 2020 are the days with the most accidents (45 and 46 respectively). Mondays are the first in number of wounded with 70 followed by Thursdays with 66 wounded. Sunday has the highest number of deaths (22). The weekend accounts for 30% of accidents, 26% of injuries and 43% of those killed ( Figure 23).

Conclusion and Recommendations
According to the analysis of the accident phenomenon in the governorate of Sfax, we can put forward the following remarks: • The Greater Sfax alone accounts for more than 80% of the number of accidents Figure 21. Distribution of numbers of accidents, deceased and injuries in Sfax Governorate by month in 2020. Data source [18]. Open Journal of Applied Sciences Figure 22. Distribution of percentage of accidents, deceased and injuries in Sfax Governorate by month in 2020. Data source [18]. Figure 23. Distribution of accidents, deceased and injuries in Sfax Governorate by day in 2020. Data source [18].
in the governorate. • The closer we get to the urban environment, the more the number of accidents increases, and on the roads of the agglomerations, the more the number of accidents increases.
• On national roads, the proportions of accidents and their severity are greater Open Journal of Applied Sciences than those on local roads and agricultural tracks.
• The length of the road network is not a strong explanation for the number of accidents.
• The highest accident densities are in the most densely populated areas.
• The delegations, which are the most equipped with means of transport, have a larger number of accidents.
• The car is the most serious means of transportation in the number of accidents, injuries and deaths. After the car, the motorcycle is second and the truck third.

Conflicts of Interest
The author declares no conflicts of interest regarding the publication of this paper.