WAYS OF REDUCING ACCIDENTS ON SOUTH
AFRICAN ROADS
I.O. JUNGU-OMARA and M.J.W.A. VANDERSCHUREN
University of Cape Town, Department of Civil Engineering
Private Bag X3, Rondebosch, 7701
ABSTRACT
Road accidents claim between 13,000 and 14,000 lives annually in South Africa. Many
victims are jaywalking pedestrians, often under the influence of alcohol or drugs. Motorists
are involved in fatal accidents due to various human, vehicle and road factors. With the
annual increase in vehicle ownership and use, the question is how to lower the fatality rate
in South Africa?
In order to substantially lower the fatality rate on South African roads, stringent speed
enforcement should be implemented and measures adopted to bring amenities closer to
informal settlements prone to jaywalking; coupled with education, enforcement and
engineering.
Fine-tuning the annual reduction of road fatalities involves enforcement to reduce vehicles
prone to tyre bursts and engineering techniques to minimise road sections that are wet,
slippery or have sharp bends.
1. INTRODUCTION
Road carnage in South Africa is caused by human, vehicle and road factors. In order to
reduce the number and severity of accidents on South African roads, contributing factors
need to be identified. Ways of reducing fatalities and the severity of road accidents can be
adopted from international practices. Education of the public about the importance of
adhering to traffic rules and regulations in conjunction with enforcement and engineering
should form the basis of ways of reducing the fatality rate.
International experiences and several South African sources provide information to
estimate the potential decrease in annual road fatalities
2. FACTORS CAUSING ROAD ACCIDENTS
There were 9 918 fatal road accidents in South Africa during 2002 [1]. The main
contributory factors to fatal accidents in December 2002 were categorised as follows:
human factors 78%, road factors 12% and vehicle factors 10% [6].
2.1 Human Factors
Human factors are the main causes of road fatalities in South Africa. An individual’s ability
to hear, see, evaluate and react to information influences traffic safety [3]. Information that
a driver is subjected to includes: on-coming traffic, adjacent vehicles and pedestrians or
bystanders, and vehicles behind or in front of the driver. Human factors that reduce a
driver’s ability to safely interact with this information include: speeding, alcohol or drug
Proceedings of the 25th Southern African Transport Conference (SATC 2006) 10 – 13 July 2006
ISBN Number: 1-920-01706-2 Pretoria, South Africa
Produced by: Document Transformation Technologies cc 454 Conference organised by: Conference Planners
abuse and fatigue. Other aspects of human behaviour that influence the road safety risk
are: jaywalking, age, gender, and the violation of traffic rules and regulations.
In order to lower the fatality rate most substantially, the major contributory factors need to
be identified. Between 2001 and 2003, the highest number of fatalities occurred in
December. The festive season in December is characterised by excessive consumption of
alcohol, speeding and long distance travel. Table 1 shows human, vehicle and road factors
with their respective percentage contribution to the national fatality rate in December 2002.
These statistics allude to the fact that the South African fatality rate is not purely influenced
by one contributory factor (see Table 1).
Lowering the national fatality rate entails identifying the top tier of contributory factors and
determining ways of reducing these factors. In December 2002, the main contributory
human factors were: jaywalking 47% and speeding 30% (see Table 1). As stated earlier, in
December 2002, human factors constituted 78% overall. National road safety campaigns
should therefore, primarily focus on lowering the occurrence of jaywalking, alcohol or drug
abuse and speeding.
Jaywalking is when pedestrians cross roads at undesignated crossings with motorists not
anticipating their presence, collisions result out of inevitability. Jaywalking also involves
pedestrians under the influence of alcohol or drugs that cross or walk along side high
mobility routes. Informal settlements next to highways are characterised by high
pedestrian travel. Residents of these settlements often need to access amenities (e.g.
shops, access to public transport, etc) on the other side of highways [5]. The direct desire
lines are across highways as detours via pedestrian bridges are very long. This land use is
inappropriate, as high mobility routes should not have direct vehicular or pedestrian
access.
Speeding occurs when motorists do not adhere to stipulated speed limits. Inertia keeps the
driver and fellow passengers moving forward at the same speed as prior to the collision,
after the vehicle is stationary due to an accident. Therefore, higher travel speeds increase
the impact and severity of the accident. Head-on collisions result in all the occupants of
both vehicles being subjected to the impact of the combined speeds.
2.2 Vehicle Factors
Aspects of vehicles that increase the likelihood of accidents when travelling are vehicle
factors. These factors include: unroadworthy vehicles, worn-out tyres, wrong air pressure,
overloading of vehicles, faulty brakes and vehicle lights. In December 2002, the main
contributory vehicle factors were: tyre bursts 56% and brakes 19% (see Table 1). In these
cases, drivers have little control over vehicle factors and accidents are almost inevitable.
Tyres are designed to carry certain vehicle loads under specific tyre pressures. Worn-out
tyres have reduced wall thicknesses, increasing the likelihood of tyre bursts under vehicle
loads. Long distance travel heats up tyres increasing internal air pressure but this increase
is magnified depending on the vehicle load. Overloading of vehicles greatly increases the
likelihood of tyre bursts, especially on long distance trips without regular stops. When the
tyre bursts and the vehicle is in motion, the weight of the vehicle, passengers and goods
are transferred onto the burst tyre, will often result in overturning, especially if the brakes
are applied simultaneously.
455
Table 1. Adjusted contributory factors to fatal accidents for December 2002.
Human Factors Urban Rural Total Group % Group %
Unknown 0 0
Pedestrian: Jaywalking 76 272 348 47
Speed 30 191 221 30
Overtaking 1 34 35 5
Turn in front 9 25 34 5
Disregard of traffic signals/
stop signs 7 10 17 2
Following too close 2 6 8 1
Driver-alcohol suspected 3 24 27 4
Pedestrian: alcohol suspected 4 16 20 3
Fatigue 1 27 28 4
Subtotal 133 605 738 100 78
Vehicle Factors Urban Rural Total Group % Group %
Unknown 0 0
Overload 0 9 9 10
Brakes 1 17 18 19
Tyre burst 1 51 52 56
Tyres smooth 1 2 3 3
Lights 0 11 11 12
Subtotal 3 90 93 100 10
Road Factors Urban Rural Total Group % Group %
Unknown 0 0
Poor visibility 34 6 40 34
Poor street lighting 9 1 10 8
Sharp bend 18 2 20 17
Blind rise/corner 6 0 6 5
Poor road surface 12 3 15 13
Road wet/slippery 18 2 20 17
Poor/defective road signs 1 0 1 1
Narrow road 2 0 2 2
Road works 3 1 4 3
Subtotal 103 15 118 100 12
Total 239 710 949 100
Source: Festive season report on road accidents, NDOT, 2003
Malfunctioning brakes prevent drivers from avoiding colliding with adjacent vehicles,
bystanders, objects or pedestrians. When brakes are not functioning, the use of hand
brakes will often result in overturning of the vehicle due to inertia. Accidents are almost
inevitable due to malfunctioning brakes, as hooting may not move objects out of the path
of collision in time.
456
2.3 Road Factors
Finally, road factors increase the likelihood of accidents when travelling. Road factors
include: road surface type, slippery roads, poor visibility or inadequate street lighting,
sharp curves, inadequate road fencing and sight distance. In December 2002, the main
contributory road factors were: poor visibility 34%, sharp bends 17% and slippery roads
17% (see Table 1).
Poor visibility is due to sections of road having inadequate street lighting or dense fog.
Inter visibility is reduced between vehicles making accidents almost inevitable as
perception-reaction time is limited.
Sharp bends require low travel speeds in order to avoid driving off the road at a tangent.
When faced with a sharp bend at certain travel speeds, motorists struggle to keep vehicles
on the road as the vehicle tends to travel straight. This tangential vehicular movement
results in accidents with adjacent surroundings or overturning.
Slippery road sections prevent motorists from avoiding collisions with adjacent vehicles,
bystanders or objects. At a certain speed, drivers feel in control of the vehicle but slippery
roads prevent the vehicle from stopping in time.
3. STATISTICAL MODEL TO ESTIMATE ACCIDENT REDUCTION
In order to substantially lower the fatality rate, measures need to be determined that will
reduce the top tier of human, vehicle and road factors.
As stated earlier, in December 2002, human factors comprised 78% of the fatality rate.
Nationally efforts should focus on lowering the main contributory human factors i.e.
jaywalking, alcohol or drug abuse and speeding.
Lowering the occurrence of jaywalking in South Africa involves preventing human
settlements close to freeways. Theoretically, these high mobility routes should not allow
direct pedestrian or vehicular access. Building high concrete walls does not remove
surrounding trip attractions that make residents walk alongside or cross freeways. In the
event when informal settlements crop up next to freeways, a study should be done to
determine [5]:
• How pedestrians get involved in accidents adjacent to freeways,
• What trip attractions give residents incentive to walk alongside or cross freeways,
• What measures would bring these trip attractions closer to informal settlements,
removing the need to walk alongside or cross these high mobility routes and
• The effectiveness of implementing these measures.
The author strongly feels that this approach would yield a 90% reduction in pedestrian
fatalities. This reduction rate was assumed as the average pedestrian, sober or not, would
have no reason to cross these high mobility routes or walk alongside but allows for a one
in ten chance of the odd jaywalker. Nationally, this would mean reducing pedestrian
fatalities from 348 to 17 in December 2002 (see Table 1).
Speeding is the second main contributory factor to fatalities in South Africa. Educating the
public about the dangers of speeding should be done using signboards and road safety
campaigns via the media.
Currently 40% of South African drivers exceed the 120km/h speed limit, 80% the 100km/h
limit and 90% the 60km/h limit [1]. It can be stated that 40% of motorists on rural roads
457
exceed the speed limit and 80-90% of motorists on urban roads. Speed is a contributory
factor in 75% of accidents on South African roads and the higher the speed, the greater
the impact [1]. Higher travel speeds increase the number and severity of accidents. This
explains why, although fewer drivers exceed the speed limit on rural roads, there are more
fatal accidents on rural roads. This highlights the need for more traffic police to be
assigned to urban roads and for very costly fines to speed limit violators on rural roads.
Due to shortage of traffic police, 90% of urban roads can be fitted with speed cameras.
By law, the speed limit is exceeded when one travels 10% over the stipulated limit i.e.
66km/h in urban areas and 132km/h on roads in rural areas. Speed enforcement by traffic
police would keep motorists within the speed limit i.e. 6km/h speed reduction in urban
areas and 12km/h in rural areas (in research done by the CSIR (1980’s) it was found that a
reduction of 1km/h could lead to a reduction of 9 fatal and 120 total crashes per month).
A unit change in urban speed limits resulted in a change in fatalities in South Africa from
0.6% to 4.0% [4]. These South African fatality rates reflect a varying range. It was decided
to use a conservative rate of 3.5%.
The decrease in fatalities in determined as follows:
• Urban 60km/h: 6km/h x 0.9 x 3.5/100 = 0.19
• Urban 100km/h: 10km/h x 0.8 x 3.5/100 = 0.28
• Rural 120km/h: 12km/h x 0.4 x 3.5/100 = 0.17
Total = 0.64
Table 1 indicates that 221 fatalities were due to speed in December 2002 with 30 in urban
areas and 191 in rural areas. These decreases in fatality rates were applied to the number
of fatalities shown in Table 1 as shown in Table 2.
Enforcement should involve the use of speed cameras and violators paying spot fines or
have their vehicles impounded until the fine is settled. Speed enforcement should also
involve the use of Intelligent Speed Adaptation (ISA) in public transport vehicles and
motorists apprehended travelling at 160 km/h. These motorists should have their drivers’
licences being tagged electronically; indicating the number of speed limit violations and
above a certain number it is nullified. This approach should be adopted throughout the
year, and not merely over the festive season.
Table 2. Estimated reduction in fatalities nationally during December 2002.
Urban roads Urban roads Rural roads
(100 km/h)
(60 km/h)
Reduction in average travel 6 km/h 10 km/h 12 km/h
speeds
Decrease in fatality rate 0.19 0.28 0.17
Number of fatal accidents 30 30 191
Number of fatal accidents reduced 6 8 32
In order to primarily lower the fatality rate nationally, these measures should be adopted to
reduce jaywalking and speeding. Secondary measures should seek to lower vehicle and
road factors.
458
Preventative measures should be taken to avoid vehicles factors e.g. routine checks on
the functionality of brakes, tyres, lights, indicators and overall vehicle roadworthiness.
Tyre bursts often result in vehicles overturning, head-on collisions, sideswipes and head-
rear accidents. The public should be educated via the media about how load rating causes
tyre bursts. Motorists should also be educated about the importance of regular stops on
long-distance trips. Drivers (especially long distance public transport drivers) should be
made aware that regular stops prevent overheating of tyres, thus minimising the chance of
tyre bursts. It can be assumed that public education could reduce one out of five possible
tyre bursts i.e. a 20% reduction of fatal accidents. This reduction rate in fatalities is
assumed because four out of every five drivers simply want to get to their destination as
soon as possible.
Traffic police should also ensure that overloaded vehicles are not permitted to undertake
long-distance trips. It can also be assumed that stringent traffic-law enforcement reduces
one out of every five possible tyre bursts i.e. a 20% reduction of fatal accidents. This
reduction rate in fatalities is assumed because the shortage of traffic police would only
stop one out of five overloaded vehicles along high mobility routes during the peak hours
only.
Education and enforcement collectively adds up to a 40% reduction rate of fatalities. Under
this assumption, this would translate to 21 tyre bursts out of 52 being prevented due to
public education and stringent traffic-law enforcement.
Every driver is aware of the safety risk of travelling with faulty brakes, but not all ensure
that they are in proper working condition. It is hard for traffic police to check whether
brakes are in a proper working condition because it requires heavy, specialised equipment
that is fixed into roadworthy testing stations. Corruption in roadworthy testing stations put
vehicles with faulty brakes on South African roads. Given the various problems associated
with trying to curb faulty brakes, it can be assumed that public education, traffic-law
enforcement and testing of vehicle roadworthiness will reduce one out of 10 accidents due
to faulty brakes i.e. a 10% reduction of fatal accidents. This reduction rate in fatalities is
assumed because traffic police never check for faulty brakes and probably one out of ten
motorists would immediately fix their faulty brakes once detected. This translates into two
accidents due to faulty brakes being avoided nationally in December 2002.
Educating the public and engineering solutions are required to mitigate road factors i.e.
poor visibility, sharp bends and slippery roads.
Poor visibility hinders motorists from making the necessary manoeuvres preventing road
accidents. Accidents caused due to misty weather or inadequate street lighting can be
avoided by improving visibility i.e. increasing light intensity. If the level of light is raised
from 1 to 2cd/m2, the incidence of nighttime accidents can be reduced by up to 65% [3].
This assumes that all fatal accidents after dark are due to motorists not having adequate
lighting. It does not take into account other factors that cause accidents at night (e.g. driver
fatigue, alcohol or drug abuse, speed, etc). The author assumes fatal accidents that occur
after dark are purely because of inadequate lighting due to objects suddenly appearing in
front of a moving vehicle. This usually occurs where roads with animals in the vicinity are
not fenced off or when pedestrians jaywalk at night. Fencing off the roads will eliminate
wild and domestic animals being hit by vehicles. The author assumes that one out of 20
fatal accidents after dark are due to animals suddenly appearing in front of vehicles and
pedestrians jaywalking. In December 2002, 40 accidents were caused by poor visibility.
Increasing light intensity from 1 to 2cd/m2 would thus prevent two fatal accidents due to
459
poor visibility. However, more research is required to determine the actual reduction in
fatality rates due to purely poor visibility be improved as other factors could be at play e.g.
fatigue.
Traffic engineers can identify sharp bends but these road sections cannot be re-aligned,
as it is not cost effective. Travel speeds on sharp bends should be reduced to safe speeds
in relation to available stopping sight distance. Traffic engineers can apply various speed
reduction measures to ensure motorists reduce their travel speeds. In addition, adequate
road signage would make the sharp bends more visible. It can be assumed that the right
speed reduction measures can lead to three out of five motorists lowering their speeds.
This would ensure a 60% reduction of fatal accidents i.e. 12 fatal accidents would have
been prevented nationally in December 2002. This reduction rate in fatalities is assumed
because two out of every five drivers could be under the influence of alcohol or drugs,
extremely tired or simply risk takers.
Traffic engineers can also identify slippery road sections. These road sections can have
a layer of single or double seal applied to increase skid resistance. It can be assumed that
increased skid resistance would reduce fatal accidents caused by slippery roads by 80%
ie.16 fatal accidents would have been avoided nationally in December 2002. This
reduction rate in fatalities is assumed because there is a one out of five chance that
accidents at these road sections are due other human or vehicle factor.
Table 3 is a summary of the estimated reductions in fatalities nationally in December 2002.
This table shows that in order to substantially lower the national fatality rate, measures to
reduce jaywalking and speeding need to be implemented. Fine-tuning the reduction in
fatalities involves applying measures to lower the occurrence of tyre bursts, malfunctioning
of vehicle brakes, poor visibility, sharp bends and slippery road sections.
Table 3. Reduction in number of fatal accidents nationally in December 2002.
Contributory Fatal Reduction in Measure
factors accidents fatalities
Jaywalking 348 313 Study to reduce crossing of freeways
Speeding 221 46 Education and enforcement
Tyre bursts 52 21 Education and enforcement
Vehicle brakes 19 2 Education and enforcement
Poor visibility 40 2 Fencing off roads
Sharp bends 20 12 Speed reduction measures
Slippery roads 20 16 Apply a layer of single or double seal
Total 720 412
Between 16h00-16h59 on Fridays is when the highest number of accidents (i.e. fatal,
serious, slight injury and damage-only) occurs in the Western Cape [2]. It is assumed that
alcohol and/or drug abuse is one of the major causes as the weekend drinking spree
starts and mental exhaustion is highest. This could explain the high accident rates. The
author assumes that if traffic police set up roadblocks on Fridays from 12h00 until 18h00,
two out of five accidents could be prevented. This denotes that 5 281 accidents would be
avoided annually. This reduction rate in fatalities is assumed because three out of every
five drunken or extremely tired drivers will not get apprehended due to the various travel
routes used. This approach in the Western Cape should be used in all other provinces
after it is established when most accidents occur in other provinces.
460
Studies suggest that truck-driver fatigue may be a contributing factor in at least 30 to
40% of all heavy-truck accidents [1]. Conversely, educating the public (especially heavy-
duty truck drivers) about the importance of resting periodically during long-distance trips
and the provision of shower facilities at filling stations along national routes can reduce
heavy-truck accidents by 30-40%. 5 274 vehicles with Gross Vehicle Mass (GVM)> 3
500kg were involved in accidents during 2001 [2]. This constitutes 7% of all accidents in
Cape Town. Public education in terms of driver fatigue could prevent 1 582 - 2 110
accidents. This notion should be applied nationally to reduce the incidences of heavy-truck
accidents in all the other provinces.
Western Cape statistics (derived from www.capetown.gov.za/reports) do not indicate which
vehicle factors cause accidents.
The type of vehicle involved in the majority of accidents simply reflects that the likelihood
of a particular vehicle type having an accident increases, as it comprises a larger
percentage of vehicle types.
According to Table 4, mini-bus taxis have more accidents than heavy trucks
(GVM>3500kg) due to more frequent public transport trips made, even if heavy trucks
make up a larger percentage of vehicle types. Motorcars make up the largest percentage
of vehicle types. The probability of this vehicle type being involved in accidents would
therefore be very high as shown in the accident statistics in Table 4. Generally, lighter
vehicle types are involved in more accidents due to higher travel speeds and more vehicle-
kilometres travelled. Motorcycles have the same accident risk as motorcars although they
constitute a lower percentage (see Table 4). This could be due to higher acceleration
capabilities of motorcycles and lack of dedicated lanes for this vehicle type.
Table 4. Types and number of registered vehicles involved in accidents
(Cape Town; 2001).
Vehicle type No. of vehicles % No. of registered Risk per
involved in accidents vehicles registered
vehicle type
Motor car/ station wagon 85 825 66.0 551 892 0.16
Light delivery vehicle 18 227 14.0 147 306 0.12
Combi/ Mini bus 8 428 6.5 10 254 0.82
GVM >3500 kg 5 274 4.1 18 090 0.29
Unknown 5 254 4.0 3 240 1.62
Motor cycle 3 073 2.4 19 717 0.16
Bus 1 394 1.1 5 202 0.27
Other 1 361 1.0 14 515 0.09
Articulated trucks 1 212 0.9 2 601 0.47
Total 130 048 100.0 772 817
Source: Cape Town metropolitan council website- www.capetown.gov.za, 2003
All vehicles are registered, which does not necessarily mean that all are roadworthy at the
time of the accident. This is because vehicle roadworthiness is only tested when vehicles
change owners. It can be assumed that buses and trucks are in a better roadworthy
condition than taxis. This assumption was made as buses and trucks are run by
commercial firms unlike taxis owned by individuals that perceive upkeep and vehicle
maintenance as unrecoverable costs. The risk per registered vehicle type is a reflection of
driver behaviour among vehicle types. Generally, drivers of mini-bus taxis have very little
regard for traffic rules and regulations. This is reflected in Table 4, with combi/ mini-bus
taxis having the highest accident rate or risk per vehicle type.
461
Furthermore, government should legislate to fit semi-open ISA in all mini-bus taxis and
buses. This is vital as speed is a factor in 75% of all accidents. As seen earlier, stringent
traffic law enforcement is estimated to reduce the fatality rate by 28% in urban areas and
17% in rural areas. The author assumes that the reduction in fatality rates is applicable in
urban areas of the Western Cape and to all the accident rates. These measures would
prevent 2 360 mini-bus taxis, 1 477 vehicles with a gross mass greater than 3 500kg, 390
buses and 339 articulated trucks from being involved in accidents in 2002. This approach
should be adopted in all the other provinces as part of the national model.
The fatality rate is highest after dark [2]. If the level of light is raised from 1 to 2cd/m2, the
incidence of night accidents can be reduced by up to 65% [4]. As mentioned earlier, other
factors contribute to accidents after dark. The author assumes that one out of 20 fatal
accidents after dark are due to animals suddenly appearing in front of vehicles and
jaywalking. Therefore, increasing the light intensity by 1cd/m2 would prevent six fatal
accidents, 18 serious accidents and 522 of all accidents after dark. It is assumed that one
out of every 20 accidents after dark is due to animals because most wild animals are deep
in the bush, far away from the nearest road with the odd domestic animal not in its kraal.
More accidents occur on dry rather than wet roads [2]. It can be said that most of the time
it is not raining (i.e. roads are dry), explaining the high frequency of accidents on dry
roads. Reducing the risk of accidents on any road surface type involves addressing
various human, vehicle and road factors.
The Western Cape accident statistics could not be used to develop a national model to
estimate decreases in the fatality rate, but were used to complement the national model.
Road safety audits should be carried out at the worst known road sections and
intersections in accordance with the South African Road Safety Manual.
Proper data collection by the SAPS and traffic police in conjunction with further research
by national and provincial authorities would accurately quantify the decrease in the fatality
rate due to these measures.
4. ANNUAL NATIONAL MODEL AND COST ESTIMATES
Table 5 shows the unit cost of fatal and serious accidents in the Western Cape. It is
assumed that the unit cost of accidents in the Western Cape does not vary substantially
with national unit costs.
Table 5. Unit costs (R) of accidents in South Africa (2001) [2].
Severity Drivers/Passengers Pedestrian All
Fatal 694 584 225 457 471 424
Serious 148 549 57 387 108 098
Table 1 shows the various contributory factors of road fatalities in December 2002. It is
assumed that this worst-case scenario gives a representative percentage split between the
various contributory factors annually. This analogy has been used to estimate the
percentage split between the various contributory factors annually, as shown in Table 6.
This table only takes into account the top tier of contributory factors. The statistical model
used in Table 3 was derived from accident statistics in December 2002. The author
strongly feels that the established worst-case scenario is representative of annual trends
related to causing and preventing accidents. In light of this, it was decided to use reduction
rates estimated for December 2002 annually, as shown in Table 6.
462
Table 5 is incorporated into Table 6 to determine the annual cost savings due to fatalities
reduced. These cost savings are considered theoretical because the likelihood of the
relevant transport authorities or government annually implementing the various measures
is not definite.
The author strongly believes that the likelihood of the relevant transport or government
authorities to implement the various measures to reduce pedestrian fatalities is 25%,
speeding 35%, faulty brakes 5%, tyre bursts 10%, poor visibility 5%, 10% for sharp bends
and wet/ slippery roads. These percentages were chosen because of the current
enforcement of speed, disuse of worn-out tyres and engineering measures continually
adopted to address sharp bends, poor visibility and wet/ slippery roads. It is envisaged that
likelihood of the relevant government or transport authorities to adopt measures to bring
amenities closer to informal settlements prone to pedestrian fatalities is 25%.
Table 6 indicates the viability in implementing the various measures and cost savings
thereof.
Table 6. Annual Cost savings due to estimated accident reductions in
South Africa [2].
Theoretical % Estimated Theoretical Estimated
Fatal Likelihood Fatal Costs Costs
Fatal % reduced to reduced saved saved
2004 Reduction 2004 implement 2004 2004 2004
Human Factors % % % R Millions R Millions
Pedestrian: Jaywalking 47 3 861 90 3 475 35 1 216 R 784 R 196
Speed 30 2 452 33 809 25 202 R 562 R 197
Subtotal 100 8 189 52 4 284 33 1 419 R 1 346 R 393
Vehicle Factors
Brakes 19 200 10 20 5 1 R 14 R1
Tyre burst 56 577 45 260 10 26 R 180 R 18
Subtotal 100 1 032 27 280 10 27 R 194 R 19
Road Factors
Poor visibility 34 444 5 22 5 1 R 15 R1
Sharp bend 17 222 60 133 10 13 R 92 R9
Road wet/slippery 17 222 80 178 10 18 R 123 R 12
Subtotal 100 1 309 25 333 10 32 R 231 R 22
Total 10 530 47 4 897 30 1 478 R 1 771 R 434
Theoretical cost savings are derived from implementing the various proposed measures
while estimated costs take into account the likelihood of various government and transport
authorities actually applying these measures.
463
5. CONCLUSIONS
In South Africa, pedestrians are more prone to road fatalities than motorists. In order to
substantially lower the fatality rate on South African roads, measures need to be
implemented to reduce pedestrian fatalities, speeding and alcohol abuse among
pedestrians and motorists. Table 6 justifies the annual implementation of measures to
reduce speeding and jaywalking in order to lower the national road fatality rate.
December was the month with the highest number of fatal accidents from 2001 to 2003.
Applying measures that reduce pedestrian fatalities, speeding and alcohol abuse in
December, should also be practiced throughout the year to lower the national fatality rate.
Fine-tuning the reduction of fatalities in South Africa involves adopting measures to reduce
the occurrence of:
• Vehicles prone to tyre bursts or have faulty brakes
• Road sections with poor visibility, sharp bends or are slippery
Table 6 indicates that fine-tuning the annual reduction of road fatalities involves
enforcement to reduce vehicles prone to tyre bursts and engineering techniques to
minimise road sections that are wet/ slippery roads or have sharp bends.
More research is required to validate some of the assumptions made in reducing fatality
rates caused by human, vehicle and road factors. Improved data collection among traffic
police can be a source of valid accident statistics needed to quantify fatality rate reductions
due to various measures and techniques proposed.
6. REFERENCES
[1] Arrive Alive website, (2004). http://www.arrivealive.co.za/pages.asp?, Arrive Alive.
Accessed 24th February 2004.
[2] Cape Town metropolitan council website, (2003). http://www.capetown.gov.za/
reports/pdf/traffic.pdf, Cape Town metropolitan council, Cape Town, 2003. Accessed
29th October 2003.
[3] Department of Transport, (2003). Festive season report on road accidents,
Department of Transport, Pretoria, 2003.
[4] Fieldwick R., (1981). The relationship between rural speed limit and accident rate,
National Institute for Transport and Road Research, CSIR, Pretoria, 1981.
[5] Gauteng Department of Public Transport, Roads and Works, (2001). Diepsloot land
use and transportation assessment, Department of Transport, Johannesburg, 2001.
[6] UNIARC Symposium, (2003). Un-roadworthy vehicles and road traffic accidents,
National Department of Transport, U.N.I.A.R.C., Durban, 2003.
464
