THE WORLD BANK GROUP A World Free of Poverty

Road Safety: Accident Counter Measures at Hazardous Locations

The contents of this page were taken from the DFID funded project RRMP2: Institutional Development Component - Road Safety Training in Bangladesh. The paper, Accident Countermeasures at Hazardous Locations, was based on Towards Safer Roads in Developing Countries, published by TRL

The potential for accident reduction through low-cost, engineering measures at hazardous sites is particularly high. Simple measures can significantly reduce problems at such sites. For example, the use of road signs and markings to channelise traffic through complex intersections, or to provide safe waiting areas for turning vehicles can often result in substantial reductions in accidents. Yet, because of lack of funds and poor maintenance capability, known hazardous locations are often left untreated and remain causes of accidents. Drivers are often presented with misleading information or no advance warning, sightlines may be inadequate, pedestrians may not be catered for and accidents may occur because of a driver's inability to cope with the particular combination of circumstances and environment. By identifying and eliminating the features that make sites hazardous, engineers can improve road safety. This often means reducing the complexity of an intersection or enabling manoeuvres to be made in stages. Reducing the number of decisions drivers must make at any one time simplifies the driving task and helps drivers to progress in safety and comfort with a minimum of conflict with other traffic and pedestrians.

There are four basic strategies for accident reduction through the use of countermeasures. These are:

  • Single site (blackspot programmes) - the treatment of specific types of accident at a single location;
  • Mass action plans - the application of a known remedy to locations with a common accident problem;
  • Route action plans - tile application of known remedies along a route with a high accident rate;
  • Area with schemes - the applications of various treatments over a wide area of town/city, i.e. including traffic management and traffic claming (speed reducing devices).

Blackspot treatment is likely to be the most effective and straightforward in countries with no prior experience of accident remedial work. Many highway authorities in industrialised countries began in this way and only later moved on to mass and route action plans as experience increased.

All of these strategies rely on the availability of data describing accidents and their locations to identify where accidents occur and what are the common features that contribute to them. Accident data and the use of collision diagrams, a key analysis tool for the traffic engineer, are discussed in the next sections.

Identifying Accident Blackspots

Highway engineers and traffic police generally know of the tendency for road accidents to cluster together at certain locations, commonly termed 'accident blackspots'. The straightforward process of plotting accidents on maps reveals this and this method remains an important means of identifying accident blackspots in many countries. Reasonably accurate and complete records are essential for this. Without precise location data, accidents cannot be plotted with any certainty, although in the total absence of data - or during an interim period while data are being collected - it may be possible to make a start on remedial works at 'known' blackspots, based on local knowledge of sites where accidents occur most frequently. It is preferable, however, to identify blackspots in an objective way using accident records.

Road intersections are often accident blackspots. It is important to distinguish between accidents occurring at intersections or on flicks (the sections of road between intersections) as the factors contributing to the accidents and possible treatments are generally very different for each. While in many cases the location will be clear, there will be accidents near to intersections that might fall into either category. In such cases, depending upon the quality and extent of data, it is desirable to examine the factors contributing to the accident in order to establish whether the features of tile intersection where important, and if so to classify the accident accordingly as an intersection accident. Generally, accidents that occur within 30 metres of an intersection can be regarded as 'intersection' accidents.

Research has shown that the numbers of accidents at a particular site will vary widely from year to year, even if there are no changes in traffic or in the road layout. In statistical terms, road accidents at individual sites are rare, random, multifactor events. This means that comparison between the numbers of accidents at particular sites must be made with respect to a fixed time period, typically one year. Furthermore, a single year's data will be subject to considerable statistical variation. Ideally, several years' data are required, from which a mean, annual accident rate can be calculated. Three years is generally regarded as a practicable minimum period for which a reasonably reliable annual average rate can be calculated.

Whilst the annual number of accidents at an intersection or other location is a straightforward concept, which considering links in the highway network it is helpful to think in terms of accident density. There will generally be specific locations at which accidents occur, for example an unexpected, sharp bend. Elsewhere, accidents may occur along a section of road without any obvious single feature. Here it is necessary to think in terms of accidents per kilometre, that is, the accident density along a particular link.

Given these data, it is then possible to rank sites in terms of their accident history. For intersections, this would be in terms of average numbers of accidents per annum; for links it would be in terms of average accidents per km per annum.

The severity of accidents should also be taken into account, as accidents with fatal and serious injuries are more costly in both social and economic terms. If sufficient research has been carried out to identify the costs of accidents of different types and with different severity, then they can be weighted relative to their cost. Thus, if a fatal sideswipe accident costs a society 20 times more than a similar slight injury accident, then it can be counted as 20 accident units. Using weightings, however, has the disadvantage that a few, random fatal accidents can sometimes dominate the selection. Alternatively, if such cost information is not available, qualitative weighting can be applied. For example, in South Korea and in Trinidad and Tobago, the Equivalent Accident Numbers (EAN) used for initial ranking purposes are 12 for a fatal; 3 for an injury accident and 1 for a damage-only accident. An EAN score can thus be awarded to each site, based on the sum of EAN values. This allows comparison of sites. Care should be taken to choose sites where remedial action will be most effective. For example, treatment of a site having 3 injury and 3 damage-only accidents with similar causes is more likely to be successful than treating a site with the same EAN comprising a single fatal accident.

Where possible, the effects of traffic volume should also be considered. In simple terms, more traffic would be expected to lead to more accidents. If traffic flow data are available it can be helpful to compare sites in terms of accidents per unit of traffic. Such accident rates are often expressed as accidents per million vehicles entering an intersection or accidents per million vehicle km on a link. If such data are available sites can be compared in terms of these rates that gives an indication of their relative safety, given their traffic volumes.

Traffic flow data are rarely available in sufficient quantity or accuracy to justify this approach. It is recommended that effort be concentrated on collecting comprehensive and accurate accident data, and that blackspots be identified initially on the basis of annual accident totals, where possible averaged over a three-year period. If sufficient data are available, these can be weighted to reflect severity.

Ports & Logistics | Railways | Roads & Highways
Rural Transport | Transport Economics & Sector Policy | Urban Transport

Help in Using This Site