Journal of Civil Engineering and Architecture 9 (2015) 225-231
doi: 10.17265/1934-7359/2015.02.011
Trends and Causes of Traffic Accidents in Dubai
Akmal Abdelfatah
1
, Mohamed Saif Al-Zaffin
2
and Waleed Hijazi
1
1. Civil Engineering Department, American University of Sharjah, Sharjah 26666, United Arab Emirates
2. General Department of Traffic Police, Dubai Police, Dubai 1493, United Arab Emirates
Abstract: This paper investigates the traffic accidents’ trends and causes in Dubai. For this purpose, the traffic accident statistics for
years 2003 to 2012 are used to examine the trends of different types of accidents and the causes of these accidents. Also, the paper
determines the most common types of accidents over the study period and identifies some of the causes for these accidents.
Furthermore, the paper shows the impact of the new traffic law amendments, which were issued in 2008, on the accident rates and
trends. The results showed that the implementation of the traffic law amendments did not have major impact in year 2008. However,
there was a significant effect of these amendments in the following years. Also, lack of awareness of traffic laws and regulations in
Dubai has been proven to be one of the major causes of accidents. Finally, the paper provides some recommendations to reduce the
traffic accident rates, in order to achieve the Dubai Police Department 2020 strategic goal to have zero fatalities.
Key words: Traffic safety, accident causes, fatalities.
1. Introduction
Dubai has witnessed a very high population and
economic growth rate over the past two decades. This
increase was, unfortunately, accompanied by an
increase in the traffic accident rates as well, which were
higher than the rates observed in most of the western
countries. Having a similar situation in most of the
major cities in the UAE (United Arab Emirates), the
government of UAE has recognized this increase and
applied major amendments of the Federal Traffic Law
as of March 2008. The new system implemented much
higher penalties on traffic offences and introduced
registration of traffic points even in absentia offenses
unlike the previous system that registers traffic points
only for spot offenses. In addition to these amendments,
Dubai Police Department has declared that the strategic
goal for Dubai is to have zero fatality per hundred
thousand of population by the year 2020.
2. Literature Review
There are several studies that deal with the rate of
Corresponding author: Akmal Abdelfatah, Ph.D., associate
professor, research fields: transportation engineering and traffic
analysis. E-mail: [email protected].
accidents and fatalities as related to the population or
the number of vehicles registered in a certain city or
country. In 2004, the WHO (World Health
Organization) [1] reported that traffic accidents are the
9th cause of death, with a total number of deaths equal
to 1.3 million. Lately, WHO showed that road
accidents were the cause for 1.24 million fatalities
worldwide. The data in the report were collected from
182 countries around the world and included about
99% of the world’s population [2]. This shows the
seriousness of the situation and the need for major
efforts to improve the current systems which can result
in decreasing the number of fatalities due to road
accidents.
Other research efforts studied the road traffic
accidents in China for the period 2000-2005. The
research concluded that the driver experience, the
classification of the road, and the level of urban
development surrounding the road are the main factors
that characterize the accidents’ fatalities in China. In
addition, there are other reasons that include: cultural
issues such as drunk driving, poor road performance
that is due to limited funding, overloaded heavy trucks
because of expensive toll, resulting in more sever
D
DAVID PUBLISHING
Trends and Causes of Traffic Accidents in Dubai
226
fatalities and lack of safety. It suggested allocating
fund for investments in transportation infrastructure
that is not only beneficial to traffic but also to the
national GDP (gross domestic product). Furthermore,
the road traffic safety system in China needs to be
enhanced [3, 4]. Peng et al. [5] considered the traffic
accidents in urban areas in China, where the population
growth is causing higher frequency of traffic accidents.
They utilized a database that includes historical data
regarding road traffic accidents in Shanghai. The data
were studied in order to determine the relationship
between traffic accidents and the road parameters.
They also provided recommendations to the local
authorities in order to improve the traffic safety on
local roads. Another research evaluated the characters
and causes of accidents migration in south China [6].
Another important reason for road traffic accidents
is related to the driver behaviour. The attitude of the
drivers was examined through surveying 2,614 drivers
in Norway. The questionnaire results indicated that
behaviours such as speeding and rule violations,
careless driving, attitude towards drinking and driving
are among the most important behaviours that impact
the involvement in accidents. It was also concluded
that age and gender are significant factors that impact
the variations of drivers’ attitude [7].
Erdogan [8] evaluated the situation in Turkey using
the accident data for 2001-2006 and concluded that
although there is a decrease in fatalities, the fatalities
ratio is still high when compared to developed
countries. There are many improvements that could be
applied to the transportation system in terms of safety
to decrease fatalities.
As a result of a study on the rural Sweden roads, it
was mentioned that the presented data should consider
many factors such as environmental issues, and the
type of the traffic flow, for instance, to be
homogeneous (light vehicles only) or
nonhomogeneous (light and heavy vehicles). The study
developed empirical models describing the
relationships between accident frequency and traffic
flow. The results concluded that the consideration of
homogeneous or nonhomogeneous traffic has a
significant impact on the accidents’ trends [9]. A
similar study used regression analysis to evaluate the
traffic accident data in Jordan and the UAE from 1990
till 2004, and Qatar till 2006. The study results
indicated that exponential models showed the best fit to
predict the relationship between number of fatalities
and the population or number of vehicles. The models
showed an acceptable average absolute error of 20.9%
for Qatar, 10.9% for Jordan and 5.5% for the UAE.
Also, the authors concluded that the gross domestic
income and the fatality rate showed a strong linear
relationship [10]. The effect of driver’s age on traffic
accidents in Florida US has been studied by Abdel-Aty
et al. [11]. The authors used the 1994 and 1995 traffic
accident data in Florida to develop log-linear and logit
models. These models considered the relationship
between the driver’s age and several factors that impact
traffic accidents, such as traffic volume, severity of the
injury, types of collision, road type, and speed
distribution.
As a way to decrease the number of accidents,
approach traffic calming processes have been studied
by Knapp [12]. The speed-control measures are
controlled by three ways: vertical speed control,
horizontal speed control and narrowing. Vertical speed
control includes humps placed across the roadway. On
the other hand, the horizontal speed control is placed on
the road to force the driver to change direction in order
to pass it. The narrowing includes the curb extension
that reduces space in which a car can pass through.
In conclusion, it is clear that the traffic accidents are
causing major problems in different parts of the world.
Many of the reasons for accidents that have been
reported in the literature are applicable to Dubai, such
as speeding, rule violations, and careless driving.
However, some of the most common reasons for
accidents in China and Europe are not very common in
Dubai, such as drunk driving. As for the road
conditions, it is not a major problem in Dubai as most
Trends and Causes of Traffic Accidents in Dubai
227
of the roads are fairly new and have been designed and
constructed according to the international standards.
3. Data Collection
The data collected in this paper are from different
sources. The number of accidents and population from
2003 till 2012, and the number of registered vehicles
from 2003 till 2005 are provided by the General
Department of Traffic Police, Dubai Police. The
number of registered vehicles in Dubai from 2006 till
2012 is obtained from RTA (Road and Transportation
Authority).
4. Research Methodology
The research is aiming at using regression analyses
to estimate the trend of accidents in Dubai. Two
accident rates will be analysed: one based on
population and the other is based on number of
registered vehicles. Based on the regression analyses,
the research recommends more reliable rate to be used
for accident analysis in Dubai. Furthermore, the
research defines the most common types of accidents in
Dubai and determines the general trend of these
common types of accidents based on regression
analysis.
5. Analysis and Discussion
Table 1 shows the total number of accidents,
population, rate of accidents per 100,000 population,
the number of registered vehicles, and the rate of
accidents per 10,000 registered vehicles for years 2003
to 2012.
Fig. 1 shows the data for the number of accidents per
100,000 population and the regression analysis results
Table 1 Traffic accident data of Dubai (2003-2012).
Year Total number of accidents Population
Accidents per 100,000
population
Number of registered
vehicles
Accidents per 10,000
registered vehicles
2003 2,287 1,141,487 200.4 435,722 52.5
2004 2,413 1,228,178 196.5 500,272 48.2
2005 2,794 1,321,453 211.4 630,000 44.3
2006 3,224 1,421,812 226.8 740,187 43.6
2007 3,335 1,529,792 218.0 868,592 38.4
2008 4,011 1,645,973 243.7 996,997 40.2
2009 3,576 1,770,978 201.9 1,021,880 35.0
2010 2,686 3,300,000 81.4 1,031,961 26.0
2011 2,636 3,361,183 78.4 1,055,412 25.0
2012 2,932 4,089,107 71.7 1,137,750 25.8
Fig. 1 Accidents per 100,000 population in Dubai (2003-2012).
200.4
196.5
211.4
226.8
218.0
243.7
201.9
81.4
78.4
71.7
y = -4.710x
2
+ 35.55x + 158.8
R² = 0.815
0.0
50.0
100.0
150.0
200.0
250.0
300.0
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Accidentsper100,000Pop
多项式 (Accidentsper100,000
Pop)
Accidents per 100,000
population
Polynomial (accidents per
100,000 population)
Accidents per 100,000 population
Year
Trends and Causes of Traffic Accidents in Dubai
228
for the trend of accidents. The most suitable regression
is a polynomial curve (second order) with a coefficient
of determination (R
2
) value of about 0.82. The
regression analysis shows a good fit with the actual
data.
Fig. 2 illustrates the rate of accidents per 10,000
registered vehicles in Dubai and the regression analysis
results. It is clear that the linear regression for this case
presents an excellent fit with an R
2
of about 95%.
Considering the results shown in Figs. 1 and 2, it can
be noticed that there is a general decreasing trend in the
total number of accidents. As in Figs. 1 and 2, there is a
significant reduction in the number of accidents
between 2008 and 2009. This can be attributed to the
application of the new traffic law that applied the
traffic points’ system. This system was announced in
March 2008 and it authorizes police departments in
UAE to withdraw a driving license when the drivers are
involved in serious traffic violations. According to the
new system, a driving license is withdrawn if the driver
receives a total of 24 traffic points with a 12-month
period. Some violations are penalized by 24 traffic
points (such as drunk driving, driving a vehicle without
a license plate, not stopping after causing an accident
with injuries, and dangerous overtaking by trucks).
Also, some other violations are considered serious as
they result in a penalty of 12 points (such as dangerous
driving “racing”, reckless driving, speeding (driving 60
km/h above the speed limit), running away from a
policeman, and falling or leaking load for trucks).
Consequently, the number of accidents is in decrease.
It seems that the use of the rate of accidents per
100,000 of the population might not be as good as the
rate of accidents per 10,000 registered vehicles. This
fact can be attributed to the fact that there is a large
segment of the population in Dubai who are
construction workers that do not travel a lot. This
conclusion is drawn based on the regression analyses
results. According to the equation shown in Fig. 1, the
number of accidents per 100,000 population can reach
0 in year 2013, which is not realistic. However, the
equation shown in Fig. 2 shows that the number of
accidents is expected to be 0 (or practically a very
small value) between 2019 and 2020, which is more
acceptable.
Accidents in Dubai are classified into six categories
as illustrated in Table 2. The accident categories are: (1)
hitting a stationary object; (2) hitting a person; (3) flip
over; (4) hitting an animal; (5) falling; (6) hitting a
moving object (include accidents between two cars).
Headings of Table 2 are given numbers that refer to
previous sequence.
As shown in Table 2, the most common type of
accidents is hitting a moving object that includes
accidents between cars (Type 6). This represents more
than 60%
of the total accidents from 2003 till 2012, as
Fig. 2 Death per 10,000 registered vehicles in Dubai (2003-2012).
52.5
48.2
44.3
43.6
38.4
40.2
35.0
26.0
25.0
25.8
y = -3.143x + 55.19
R² = 0.945
0.0
10.0
20.0
30.0
40.0
50.0
60.0
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Accidents per 10,000 Vehicle
Registered
线性 (Accidents per 10,000
Vehicle Registered)
Accidents per 10,000 vehicles
registered
Linear (accidents per 10,000
vehicle registered)
Accidents per 10,000 vehicle registered
Year
Trends and Causes of Traffic Accidents in Dubai
229
Table 2 Types of accidents in Dubai (2003-2012).
Year Type 1
% of
Type 1
Type 2
% of
Type 2
Type 3
% of
Type 3
Type 4
% of
Type 4
Type 5
% of
Type 5
Type 6
% of
Type 6
2003 312 13.6 408 17.8 200 8.8 19 0.8 5 0.2 1,343 58.7
2004 298 12.4 479 19.9 141 5.8 11 0.5 9 0.4 1,475 61.1
2005 437 15.6 505 18.1 135 4.8 15 0.5 13 0.5 1,689 60.5
2006 486 15.1 603 18.7 129 4.0 7 0.2 8 0.3 1,991 61.8
2007 456 13.7 665 19.9 130 3.9 10 0.3 10 0.3 2,064 61.9
2008 545 13.6 754 18.8 176 4.4 9 0.2 16 0.4 2,511 62.6
2009 493 13.8 537 15.0 177 5.0 8 0.2 11 0.3 2,350 65.7
2010 439 16.3 336 12.5 143 5.3 10 0.4 12 0.5 1,746 65.0
2011 418 15.9 333 12.6 138 5.2 2 0.1 8 0.3 1,737 65.9
2012 511 17.4 359 12.2 146 5.0 5 0.2 6 0.2 1,905 65.0
Type 1: hitting a stationary object; Type 2: hitting a person; Type 3: flip over; Type 4: hitting an animal; Type 5: falling; Type 6: hitting
a moving object (include accidents between two cars).
Fig. 3 Percentage of accidents due to hitting a moving object (2003-2012).
indicated in Fig. 3. The regression analysis for this type
of accidents shows a very good fit with an R
2
of about
0.87. Also, the trend is increasing, which indicates that
the percentage for this type of accidents is increasing
over years. The driving behaviour is the main cause of
such accidents.
The second most common accident is hitting a
person (Type 2) with a percentage of almost 17% of the
total accidents, shown in Fig. 4. The regression
analysis for this type of accidents resulted in a
polynomial equation (second order) with an R
2
of about
0.82. The trend of this type of accidents is decreasing.
This can be attributed to the fact that the authorities in
Dubai are implementing more physical restrictions to
control the location of pedestrian crossings. Also, the
fact that Dubai Traffic Police Department has
conducted several awareness campaigns that targeted
all types of road users including pedestrians has a
significant contribution to the reduction in this type of
accidents.
Finally, the third most common type of accidents is
accident Type 1 (hitting a stationary object), which
represents about 15% of the total number of accidents
in Dubai. The trend for Type 1 accidents is depicted in
Fig. 5. The regression analysis for this type did not
show a strong fit as the R
2
is about 0.54. However, the
trend shows a slight increase in the percentage of these
accidents. This type of accident is hitting a stationary
object, which is caused by the bad drivers’ behaviour.
These three most common types of accidents in
Dubai represent about 95% of the total number of
accidents. In addition to that, the two types of accidents
y = 0.757x + 58.64
R² = 0.867
54.0
56.0
58.0
60.0
62.0
64.0
66.0
68.0
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Year
%ofAccidentType6
线性 (%ofAccidentType6)
% of accident Type 6
Linear (% of accident Type 6)
Percentage of accident Type 2
Trends and Causes of Traffic Accidents in Dubai
230
Fig. 4 Percentage of accidents due to hitting a person (2003-2012).
Fig. 5 Percentage of accidents due to hitting a stationary object (2003-2012).
that depend on the drivers’ behaviour (Types 6 and 1)
represent more than 80% of the total number of
accidents in Dubai. This is an alarming sign that the
Dubai Traffic Police Department has to consider when
applying new strategies to improve the traffic safety on
Dubai’s roads. In contradiction to the trend of the Type
6 accidents, Type 2 accidents show a reducing trend.
6. Future Work
The authors are planning for some future work to
expand this research. Some of the proposed future
studies are:
considering the rate of fatalities as an indicator of
traffic safety on Dubai roads;
investigating the distribution of the characteristics
of drivers involved in the accidents;
considering the number of traffic fines issued and
examining its impact on the traffic safety.
7. Conclusions
The number of accidents is of a major concern to
Dubai Government in general and to the Dubai Traffic
Police Department in specific. The paper showed the
trends of the total number of accidents and the most
common types of accidents in Dubai. The total number
of accidents has an increasing trend from 2003 to 2008,
followed by a decreasing trend from 2009 till 2012,
which is mainly due to the implementation of a new
traffic law in Dubai. The reduction in accident rates as
related to population and number of registered vehicles
y = -0.156x
2
+ 0.865x + 17.81
R² = 0.824
0.0
5.0
10.0
15.0
20.0
25.0
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Year
%ofAccidentType2
多项式 (%ofAccidentType2)
y = 0.057x
2
− 0.281x + 14.06
R² = 0.541
0.0
5.0
10.0
15.0
20.0
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Year
%ofAccidentType1
多项式 (%ofAccidentType
1)
% of accident Type 2
Polynomial (% of accident Type 2)
% of accident Type 1
Polynomial (% of accident Type 1)
Percentage of accident Type 2
Percentage of accident Type 1
Trends and Causes of Traffic Accidents in Dubai
231
is a positive sign of an improved safety on Dubai roads.
However, the rate of accidents per 10,000 registered
vehicles is a more robust measure.
Furthermore, data show that there are different types
of accidents where the most common one is hitting a
moving object and then hitting a person followed by
hitting a stationary object. The first and third common
types mainly depend on driving skills, while the second
type is influenced by both the driving skills and the
awareness of the public. Therefore, Dubai authorities
(RTA and Dubai Police Department) may consider
some of the following recommendations to improve
the traffic safety and help in reducing the accident rates
further:
apply stricter driving tests when issuing a new
driving license. Also, whenever a driving license is
withdrawn because of traffic violations, the driver has
to go through a stricter training program. In addition to
that, keeping track of the violations’ records of these
drivers in the future and applying tougher penalties on
them might help in reducing the number of accidents;
continue to conduct traffic awareness campaigns
using different languages as the population in Dubai
includes significant groups from different parts of the
world;
consider higher penalties for causing accidents,
especially the accidents that are caused by the personal
errors of the driver or pedestrian.
These suggestions may result in some improvements
in the accidents’ trends in Dubai.
References
[1] WHO. 2004. The Global Burden of Disease. Switzerland:
WHO.
[2] WHO. 2013. Global Status Report on Road Safety 2013.
WHO.
[3] Zhao, S. 2007. “Rapid Motorization and Road Traffic
Accidents in China.” In Proceedings of the 11th World
Conference on Transportation Research [CD-ROM].
[4] Zhao, S. 2009. “Road Traffic Accidents in China.”
International Association of Traffic and Safety Sciences 33
(2): 125-7.
[5] Peng, C., Lu, J., and Yin, W. 2010. “An Empirical Study
of the Urban Traffic Accident Causes with
Correspondence Analysis Method.” In Proceedings of the
10th International Conference of Chinese Transportation
Professionals, 1299-306.
[6] Jiao, C., Yang, M., and Hao, Y. 2009. “Analysis of
Characters and Causes of Road Traffic Accident Migration.”
In Proceedings of the 9th International Conference of
Chinese Transportation Professionals, 662-8.
[7] Iversen, H., and Rundmo, T. 2004. “Attitudes towards
Traffic Safety, Driving Behaviour and Accident
Involvement among the Norwegian Public.” Ergonomics
47 (5): 555-72.
[8] Erdogan, S. 2009. “Explorative Spatial Analysis of Traffic
Accident Statistics and Road Mortality among the
Provinces of Turkey.” Journal of Safety Research 40 (5):
341-51.
[9] Hiselius, L. 2004. “Estimating the Relationship between
Accident Frequency and Homogeneous and
Inhomogeneous Traffic Flows.” Accident Analysis and
Prevention 36 (6): 985-92.
[10] Bener, A., Hussain, S. J., Al-Malki, M. A., Shotar, M. M.,
Al-Said, M. F., and Jadaan, K. S. 2010. “Road Traffic
Fatalities in Qatar, Jordan and the UAE: Estimates Using
Regression Analysis and the Relationship with Economic
Growth.” East Mediterranean Health Journal 16 (3):
318-23.
[11] Abdel-Aty, M., Chen, C., and Schott, J. 1998. “An
Assessment of the Effect of Driver Age on Traffic
Accident Involvement Using Log-Linear Models.”
Accident Analysis and Prevention 30 (6): 851-61.
[12] Knapp, K. 2000. “Traffic-Calming Basics.” Civil
Engineering—ASCE (American Society of Civil
Engineers) 70 (1): 46-9.
