It is a well known fact that road accidents in South Africa are serious causes of concern. According to the latest accident statistics there has been an increase in the number of fatalities.
According to the article on the Arrive Alive website (Road death toll 'unacceptable')  dated 3 January 2010, Transport Minister Sibusiso Ndebele is disappointed with the road death toll, saying that one fatality on the country's roads is one too many. Preliminary statistics released by the department on 2 January 2010 revealed that 1050 people had died in road related incidents during December 2009.
Road safety is not only a major concern for South Africa but indeed a concern that requires the immediate attention of other countries worldwide. The Commission for Global Road Safety has during June 2006 presented a report titled "Make Roads Safe"  that could have an effect on the way we approach road safety. The report aims to focus on political and public attention and relates to the global road traffic injury epidemic that claims the lives of 1.2 million people and injuries around 50 million annually.
Table : Festive period figures - December 2008
Source : Traffic Focus March 2008, p38, "Fatal Road Accident Statistics" by Cielie Karow (RTMC)
The report also indicates that dangerous roads have a significant impact on developmental objectives, especially because of the immense economic and social cost of road crashes to low and middle income countries. South Africa is in this category. It is important that the knowledge gained by high income countries be transferred and implemented in South Africa.
The N3 route is a popular route during holidays and with high volumes of passenger and freight traffic on the roads, there is prone to be serious and fatal accidents. Road rage, speed and driver behaviour are not the only cause of these accidents. Holiday periods are generally during the rainy season and road elements, especially drainage, contribute to accidents.
There are varying viewpoints and opinions among civil engineering professionals on the usage of superelevation values given in the TRH 17  and SANRAL's Geometric Design Guidelines .
The geometric design considerations need to be reviewed especially when it relates to road surface drainage at points where the road has no crossfall during development of superelevation. Currently the SANRAL Drainage Manual, 5th Edition Final Revision, p5-2  recommends that the flow depth during a 1:5 year storm should not exceed 6mm. This is in contrast to the Highway Drainage Manual (FHWA-TS-79-225)  recommendation of 4mm. According to Highway Surface Design (Transit New Zealand) Manual , the critical depth for aquaplaning ranges from 4mm to 10mm depending on tyre and pavement surfacing. The surface water depth therefore, should be restricted to 4mm for all but special situations where superelevation produces long, curved flow paths. There are three common special situations where surface water depth may become critical, namely, horizontal alignment curvature, intersections and ramps and superelevation development.
2.2 Why is the research being undertaken?
Criteria for superelevation where steep vertical grades have a direct effect on drainage on horizontal curves have not been developed according to the TRB . Drainage problems associated with superelevation has resulted in an increase (especially at road widenings) in vehicular accidents and a thorough study of the surface water flow paths on road surfaces due to superelevation requirements need to be further investigated, researched and analysed. The proposed analysis and context of the research proposal will identify the factors such as speed affecting the vehicle dynamics at sharp or reduced horizontal curves where flow paths of surface water are problematic and can lead to aquaplaning.
Driver behaviour during aquaplaning in conjunction with the geometric design of the roadway are factors considered in the research. This study will be done under heavy vehicular traffic conditions during adverse weather conditions. 'Context Sensitive Design' (CSD*)  for the geometric design of safer roads will be addressed in the research study.
2.3 How will the results add to the body of knowledge?
The TRH 17 document was compiled to be a design guideline for the geometric design of rural roads in preference to standards. An objective of the TRH 17 was to continuously amend the draft document, through consultation between the authorities, discussion with practitioners and ongoing research. This has not happened. It is common practice by geometric designers to use the TRH 17 document for the design of rural roads without questioning the values obtained in the tables and graphs. The validity of the guidelines should be judged by the geometric designer in each specific situation as well as the consequences of departing from the values suggested.
Similarly SANRAL's Geometric Design Guidelines have been developed to assist design consultants. The originations of the guidelines stem from AASHTO . Currently there is no standards developed for SA and these guidelines are authority specific.
The literature review will describe the current state of knowledge on the element and will be researched. Applicable literature listed hereunder have been identified.
* "Context sensitive design (CSD) asks questions first about the need and purpose of the transportation project, and then equally addresses safety, mobility, and the preservation of scenic, aesthetic, historic, environmental, and other community values. CSD involves a collaborative, interdisciplinary approach in which citizens are part of the design team."
3.1 Reference documentation
Refer to list of references (11.3 List of References).
How will the literature review link with the problem statement and research objectives?
The literature review gives a background and base knowledge to develop in this research study.
As part of the ongoing commitment towards road safety, convenience and mobility, N3TC conducted a comprehensive road safety audit in 2006  which highlighted speed as the main cause of accidents at various locations along the N3 between Cedara and Heidelberg. Sections where speed limits are posted necessitates patience, appropriate reductions in speed and full concentration on the part of drivers. Accidents in these areas often occur because drivers lose control due to speeding, as well as the high speed differentials between heavy and light vehicles. Mist and adverse weather conditions have a significant impact of driver conditions.
Van Reenen's Pass is one of the most beautiful sections of the N3 transcending the escarpment between the Free State and Kwazulu-Natal renowned for its slippery and precarious roads (particularly as a result of the frequent misty conditions).The road is steep, very steep in places, and twists and turns as it follows the mountainous terrain. The weather is hazardous at times with high winds capable of blowing caravans, light vehicles and light trucks onto their sides. Snow has on a number of occasions completely closed the Van Reenen's Pass, whilst mist and rain regularly reduce visibility and render the road treacherously dangerous. It is of particular interest the visibility of a number of skid marks on the road surfacing and guardrail replacement.
Traffic, in particular heavy traffic, increases annually as the economy grows. Trucks relentlessly 'grind' up and down Van Reenen's Pass day and night, while during peak traffic periods as many as 3000 vehicles per hour use the pass. Using September 2004 to August 2006 statistics , the daily average traffic was 9100 vehicles, of which 2600 were large trucks (5 axles or more), almost a third of all vehicles. During the month of December 2007 recorded at the Van Reenen's Pass, the average daily traffic reached 11000. The speed differentials between trucks in low gear and the powerful new generation of light vehicles is extremely frightening, both up and down the pass. High speed differentials (76% of drivers exceed the speed limit on the pass) and failure to adapt speed in relation to circumstances, have contributed to 70% of all accidents on the pass over this period. The results do not portray a pretty picture and from a road safety point of view, Van Reenen's Pass is the single most dangerous section of the N3.
N3TC has identified and addressed road safety on Van Reenen's Pass from three points of view, namely engineering, education and enforcement . N3TC has implemented interventions to reduce the number of accidents on the pass but currently the greatest causes of accidents is by vehicle unfitness (inoperative brakes in particular), rear-end collisions due to high speed differentials, dangerous inter-lane manoeuvres and driver disorientation in extreme weather conditions.
Elements affecting safety at superelevation
The drainage conditions of the roadway in relation to the vehicle dynamics need to be further investigated as superelevation along sharp horizontal curves with reduced sight distances presents a significant level of driver concentration. The driver tends to correct the vehicular path. The surface stormwater run-off flow path along the horizontal curve effects natural braking forces which in turn can lead to hydroplaning (or aquaplaning).
Aquaplaning occurs when water pressures build up in front of a moving tyre resulting in an uplift force sufficient to separate the tyre from the pavement. During high intensity rainfall events, a water film builds up on the surface on the road. The risk of vehicle aquaplaning increases as the depth of this film increases. The loss of steering and drag force produced during aquaplaning may then cause the vehicle to lose control, especially when a steering tyre is involved. Rainfall intensity is the most important environmental factor in hydroplaning.
The risk of dynamic aquaplaning is directly proportional to the depth of water in the road surface. This depth is affected by a wide range of factors that are contributed to by the environment such as the geometric design, pavement design, drainage design and maintenance and by the condition of the vehicle.
The geometry of the road has a large effect on the water depth and is the factor over which the geometric designer has the most control. The length of time water is able to stay on the road will influence the depth it achieves. Longer flow paths mean more time to accumulate rainfall and result in higher film depths. Changes in superelevation, reduced horizontal alignment and sag curves are some of the problem areas where the slope is low or where water has to flow a long way over the pavement before being intercepted by a drainage system or dissipating into the adjacent terrain. Superelevation changes can result in long curving flow paths which may be problematic. Steeper longitudinal slopes can also increase the flow path length and resulting depth.
The pavement texture depth effects the water depth by allowing some of this water to flow between the aggregate or in grooves providing flow paths to allow water in front of the tyre to be forced out under pressure. Porosity can also be considered as some pavements such as open-graded porous asphalt allow water to drain through them, taking it away from the surface. Wheel track depressions have a significant effect on the drainage patterns increasing water depth and concentrating flow. The N3 has varying pavement surfacing of which an analysis will be done considering the flow path on these surfaces.
Pavement drainage solutions is essential to ensure that no water is able to pond on the trafficable road surface, particularly in sag areas. This is vital in order to reduce the aquaplaning risk.
Vehicle characteristics and behaviour are also important factors in aquaplaning. The speed at which a vehicle needs to travel to begin aquaplaning is determined by water depth but also by the vehicle's weight and tyre characteristics. The vehicle weight determines how much uplift force is needed to induce separation and it follows that a lighter vehicle will aquaplane at a lower speed. Higher tyre pressures increase the aquaplaning speed by reducing the contact area between tyre and road, increasing the vehicle's weight to area ratio. Tyre tread depth also affects aquaplaning the same way as pavement texture, with deeper tread moving the water away from the area of contact more effectively. While minimum tyre tread depth and maximum speed are both specified by law, minimum weight and tyre pressures are not. These are vehicle manufacturer recommendations specific.
In summary, drainage requirements versus vehicle dynamics are the key factors to be considered in the development of superelevation criteria at reduced horizontal curves to minimize aquaplaning on the N3. Practical considerations such as tyre depth, pavement characteristics and drainage solutions will be evaluated along the route.
The various literature reviewed amplifies the need for safer roads and the N3 is considered to be the most travelled route in South Africa.
Significant roadway degradation such as polishing of aggregates, bleeding of bitumen and rutting depletes the friction supply available for cornering. This depletion results from the use of a portion of the friction supply to provide the necessary braking force required to maintain speed on the downgrade. The speed of the vehicles on the roadway and the vehicle dynamics will require to be analysed as differing vehicles have different friction forces exerted on the roadway. It cannot be assumed that the relevant design criteria for a car is similar to that of a truck or vice versa. As the Independent Engineer, my previous audit reports  on the N3 has indicated the various surfacing failures. Skid marks are significantly prevalent and N3TC/SANRAL/DOT accident reports will need to be investigated as part of the research study.
It is noted from the TRH 17 that the design vehicle is a single unit truck. This undesirable combination results in a significant decrease in the margin of safety resulting from roadway grade, especially for heavy vehicles.
On long or fairly steep grades, drivers tend to travel faster in the downgrade than in the upgrade direction. Additionally, research has shown that the side friction demand is greater on both downgrades (due to braking forces) and steep upgrades (due to the traction forces). Downgrades on horizontal curves may be problematic, and that adjustment for it may be desirable in some cases. There are no guidelines as to how this adjustment should be made for two-lane or multilane divided or undivided roadways.
Some adjustment in superelevation rates should be considered for grades steeper than 5%. This adjustment is particularly important on roadways with high truck volumes and on low-speed roadways with intermediate curves using high levels of side friction demand. The superelevation revision proposition highlights that this adjustment be made by using higher design speeds criteria for the geometric design of the roadway.
More definitive guidance on this adjustment, as well as adjustment for other elements of the horizontal curve, is needed. The design speed versus minimum curve radii of horizontal curvature needs further investigation to ensure safety on sharp horizontal curves taking the other related factors like superelevation, etc into consideration.
The article published in the Pretoria News, "Wet weather causes a spike in road accidents"  refers to the significant increase in roadway accidents during rainy weather. The drainage requirements and vehicle dynamics in relation to superelevation design of the roadway will be investigated and researched.
I have consulted various professional engineers, technologists  and independent consultants  in the transportation industry. The response received was favorable in terms of the need for further investigations of superelevation in relation to drainage requirements and vehicle dynamics for roadway geometric design.
Research problems and aims
I am currently the IE (independent engineer) and have audited the routine road maintenance items of the N3 for the past 3 years. As such, I have access to some data to analyse as part of the study.
Various sections of the N3 are currently being upgraded or rehabilitated to improve the quality and lifespan of the road due to vehicular traffic increases. There is significant freight movement. This has resulted in an increase in vehicular accidents. My viewpoint on the design considerations adopted on the N3 is subjective as driver safety factors and other mitigating risks of vehicular accidents need to be further investigated.
From a geometric design analysis viewpoint, road-widening and general roadway rehabilitation are designed according to the existing roadway conditions and drainage problems associated with superelevation is prevalent. Ponding has been recorded and visual evidence is prevalent in certain areas. The general stormwater design criteria should be viewed and analysed differently from the Kwazulu-Natal conditions as opposed to the Free-State and Gauteng conditions as the runoff rainfall intensity and time of concentration varies significantly.
This study will outline and identify the following conditions with specific superelevation criteria development on the N3 route :
Superelevation criteria at steep gradients (rolling to mountainous terrain) with reduced/sharp horizontal curves;
Drainage problems associated with superelevation;
Flow paths on road surfaces due to superelevation; and
Speed and Vehicle dynamics at sharp horizontal curves.
Steep grades at sharp horizontal curves presents a dangerous situation for traffic.
The two scenarios where this condition is prevalent is at broken-back curves on mountainous terrain (Van Reenen's Pass specifically and other identified areas) with multi-lane, 2-way roads (whether it is divided or undivided) and/or high speed downgrade at/before vertical sag curves.
At these locations, the complicating factors of vehicle "off-tracking", pavement slope (crossfall), and pavement friction tests the drivers ability to provide correct vehicle positioning without compromising control of the vehicle. It has also been recorded that wind has been a cause of accidents as the vehicles cannot 'grip' onto the roadway (especially at 'Windy Corner' on the Van Reenen's Pass.
From previous design considerations, accident-related problems have arisen where, as a result of reconstruction, existing highways have been rebuilt using the 8%-10% superelevation rates in accordance with current guidelines. The rate of superelevation development is however not reviewed or adjusted.
The research approach has a both qualitative and quantitative approach.
The theoretical research will comprise of mathematical analysis with modeling and simulation. The current geometric design guidelines for rural roads (TRH 17 and SANRAL's Geometric Guidelines) will be used as base documentation for values in the research input.
Case study sections of the N3 will be decided upon and N3TC, SANRAL and the DoT (both national and provincial) will be informed of the research study. Data collection and statistical information will be sourced from the relevant authorities with prior consent.
This research would require :
the review of current design guidelines;
the development of an action plan to achieve the research objectives;
the collection of statistical data (from SANRAL, N3TC, etc) and other relevant information;
detailed visual assessment and evaluation of the roadway section;
Falling Weight Deflection (FWD) measurements will be performed at 50m intervals alternatively on the left and right outside wheel tracks along the section identified;
rut and riding quality measurements will be measured in both wheel paths as well as texture depth along the outside wheel path as part of the FWD measurements;
Measurements of the stormwater sheet flow runoff depths will be measured using conventional means;
vehicle dynamics will be physically and theoretically analysed;
the evaluation of the effects of various alternatives contextualized in geometric design guidelines and candidate criteria taking into consideration the CSD approach; and
the preparation of mitigating risks, road safety measures and final geometric design criteria for superelevation, drainage requirements taking into consideration the speed and vehicle dynamics at sharp/reduced horizontal curves.
The accident reports sourced by SANRAL/N3TC/DoT's archives will be important for the research study . The possible limitations could be the delay in retrieving these accident reports and visual assessments will be carried out as an alternative
The objective of this research is to :
analyse the speed and vehicle dynamics at sharp horizontal curves;
develop drainage criterias for flow paths or depths on road surfaces due to superelevation;
develop superelevation criteria for steep grades on sharp horizontal curves by identifying and analyzing drainage problems associated;
design safe roads from a geometric design viewpoint by taking factors such as time, cost, quality, CSD into consideration; and
develop an independent software tool to assist geometric designers and authorities in the civil engineering industry.
It is noted that other criteria associated with the design of horizontal curves such as tangent-to-curve transitions, the need for pavement widening, and minimum curve radii would also be considered in the development of the criteria.
The criteria will be based on quantitative data obtained from theoretic considerations and simulations and verified by actual field observation.
The identified areas for the actual field observations will be done by travelling the route and monitoring of the N3. Accident statistics will be essential as investigative reference for the background of the research study.
Plan of research activities
The activities to undertake the research proposal will be to :
expand literature study;
collate statistical data;
acquire permission for design data from consulting engineering firms and SANRAL/N3TC/DoT archives;
co-ordinate with SANRAL/N3TC with respect to experimental areas identified;
analyse the design guidelines (geometric and drainage) for rural roads;
prepare the roadway modeling and simulation for the various superelevation criteria taking into consideration the drainage requirements and vehicle dynamics;
Analyse the sheet flow path (hydraulic analysis) for the drainage requirements (vertical grade versus road width - at sharp/reduced horizontal curves) - the Rational Method will be used;
Investigate the CSD of road rehabilitation projects in relation to geometric design considerations for future road rehabilitation and major construction projects (It is noted that the De Beer's Pass will be constructed as an alternative to the Van Reenen's Pass);
Analyse research findings and observations using relevant software;
Synthesise the research findings, observations and results; and
Write the report.
This research would take 18 months to complete.
The potential outputs for the research study will be to :
Reduce or increase the superelevation values or rates, dependant on research outputs, with respect to vehicle dynamics and roadway widths;
Analyse the drainage flow paths as superelevation rates or values in relation to the drainage requirements and vehicle dynamics will determine the criteria to be adopted for safe driver conditions;
The speed versus vehicle dynamics analysis at sharp or reduced horizontal curves will provide suitable guidelines for future rehabilitation and road-widening projects;
CSD findings and observations to be adopted in future road rehabilitation and major reconstruction projects; and
Produce an independent software program which will be thoroughly researched, tested and developed as a tool for geometric designers and authorities in the civil engineering industry. This software can be integrated using current engineering software utilised by consulting engineering service providers, government institutions and various organizations will be investigated.
The outcome of this research will assist design consultants in determining a much easier and safer design approach to design rehabilitation, road-widening and major construction projects, by identifying problem areas and providing appropriate design values.
The recommended criteria would be documented in the final report and also presented in a form that could be used by various authorities.
Presentations will be made at national and international conferences, seminars or symposiums relating to geometric design of roads. Workshops and lectures will be conducted or presented through educational institutes, CESA and other authorities in the civil engineering industry. The research study will be published as an article in transportation journals and the research summary will be published in several magazines. I am of the opinion that I envisage national acclaim for the research and contribution of the research outputs to building the knowledge base in South Africa.
Key references and documentation
11.1 Authorities, Institutes and other source of reference
Department of Transport (National and Provincial) (DoT)
Road Traffic Management Corporation (Pty) Ltd (RTMC)
South African National Roads Agency Limited (SANRAL)
N3 Toll Concession (Pty) Ltd (N3TC)
Council for Scientific and Industrial Research (CSIR)
Transport Research Board (TRB)
American Society for Civil Engineers (ASCE)
Consulting Engineers South Africa (CESA)
Durban University of Technology (DUT)
University of Stellenbosch (SUN)
Aurecon SA (Pty Ltd (AURECON)
WSP SA Civil and Structural Engineers (WSP)
3D Compu-Systems (3DCS)
11.2 Design Guidelines and Standards
Technical Recommendations for Highways (TRH 17 - Geometric Design of Rural Roads - Draft 1988)
SANRAL Geometric Design Guidelines
SANRAL Drainage Manual (5th Edition - fully revised)
Design of Highway Drainage Manual (FHWA-TS-79-225)
AASHTO - A Policy on the Geometric Design of Highways and Streets 5th Edition (2004)