A6.6 - Typical accident reduction factors

Introduction

The following section provides average accident reduction factors for treatments or improvements in urban and rural areas. These reductions can be applied to the accident rate calculated using any of the three accident analysis methods.

In rural areas, accident migration should also be considered.

The reduction factors are only a guide to possible reduction rates and the evaluation documentation will need to substantiate all claimed accident reductions, particularly if they are expected to be greater than indicated here.

Typical accident reductions

The following five tables provide a typical range of injury accident reductions for mid-block and intersection treatments:

in urban (speed limits of 70 km/h or less);
in high speed areas (speed limits of 80 km/h or more);
for cyclist and pedestrian treatments in urban areas.

When determining the accident reduction for implementing more than one measure, it is not appropriate to add all of the reduction factors together, particularly if the measures are treating similar accident types. In these cases judgement should be exercised in determining the likely overall effectiveness.

Table A6.18(a) Typical accident reductions for mid-block treatments in urban areas

Measure	Typical effectiveness of measure (% reduction)
Flush medians 50 km/h	10% to 25% of all accidents
Raised medians 50/60 km/h	25% of all accidents
Lighting - installation or upgrade	35% of night time accidents that are due to poor lighting
Ban on street parking on both sides of the street	20% of mid-block accidents. There is little research on banning parking on only one side of a street only, but some research indicates that accidents may increase.
Implementation of area wide traffic calming on local streets	25% of all accidents within the traffic calmed area.

Table A6.18(b) Typical accident reductions for intersection treatments in urban areas

Measure	Typical effectiveness of measure (% reduction)
Lighting - installation or upgrade	35% of night time accidents that are due to poor lighting.
Installation of throat or fishtail islands at priority intersections	20% to 40 % of all accidents.
Establishing a right turn phase at traffic signals	10% of accidents involving right-turn-against accidents.

Table A6.18(c) Typical accident reductions for pedestrian and cyclist treatments in urban areas

Pedestrian measure	Typical effectiveness of measure (% reduction)
Kerb extensions to assist pedestrians to cross	35 % of pedestrian (type N) accidents.
Pedestrian refuges to assist pedestrians to cross	15 % of pedestrian (type N) accidents.
Pedestrian refuges and kerb extensions	30 % of pedestrian (type N) accidents.
Zebra crossings	No reduction in general and if located on a multi lane road or at a site with a speed limit of greater than 50 km/h an increase in accidents is possible.
Elevated pedestrian platforms constructed in conjunction with local traffic management or calming schemes	80 % of pedestrian (type N) accidents.
Mid-block traffic signals	45 % of pedestrian (type N) accidents, however an increase in motor-vehicle only accidents is possible if no crossing facility was previously installed.
Grade separated crossing facilities	30% of all accidents or up to 80% of pedestrian accidents depending on the extent to which pedestrians are prevented from crossing at grade.
Cyclist measure	Typical effectiveness of measure (% reduction)
Cycle lanes	10% of cycle accidents.
Advanced stop lines for cyclists at signalised intersections	10% of cycle accidents at signalised intersections.

Table A6.18(d) Typical accident reductions for mid-block treatments in high speed areas

Measure	Typical effectiveness of measure (% reduction)
Route lighting - installation	30% of night-time accidents that are due to poor lighting.
Passing lanes or crawler lanes (ie, passing lanes on an uphill gradient)	30% of overtaking accidents within passing lane. 40% to 60% of head on accidents within passing lane. 15% of rear-end/obstruction accidents within passing lane. Reduce these percentages linearly to zero for accidents following the passing lane up to 5 km away. Ensure loss of control accidents do not increase due to design
Shoulder widening	0% to 20% of loss of control and overtaking accidents on straights from shoulder widening alone. 0% to 20% of loss of control, overtaking and head-on accidents on bends from shoulder widening alone. 0% to 40% of loss of control, overtaking and head-on accidents on bends if sight-rails and traffic signs are installed at the same time as shoulder widening.
Guardrailing	Accident reduction in terms of changed severity: 40% reduction in fatal accidents. 30% reduction in serious accidents. 10% reduction in minor accidents.
Resurfacing of curves	Compare injury accident rate at site with typical injury accident rate and injury accident rates at other local sites that are considered satisfactory.
Installation of reflective raised pavement markers	6% of all accidents.
Installation of edge marker posts	30% to 40% of off-road on curve and loss-of control on curve accidents.
Edge lines where none previously existed	8% of all accidents.
Marking no-overtaking lines missing from a section of road where they are deemed necessary	50% to 60% reduction in head-on accidents. 40% to 60% reduction in overtaking accidents.
Marking centrelines where none previously existed	13% of all accidents providing that seal width is sufficient.
Installation of audible edge lines (rumble strip/vibraline)	11% of all accidents.
Implementation of clear zones where there were previously hazards within 6 metres of the roadway	35% of loss of control and off the road accidents.

Table A6.18(e) Typical accident reductions for intersection treatments in high speed areas

Measure	Typical effectiveness of measure (% reduction)
Intersection lighting - installation or upgrade	30% to 50% of night-time accidents at intersections that are due to poor lighting.
Right turn bays and associated seal widening at priority intersections	75 % of accidents involving vehicles turning right from the main road and those travelling in the same direction.
Installation of throat or fishtail islands at intersections	35 % of intersection accidents.
Installing advance warning of intersections where it is deemed necessary	7% of all intersection accidents.
Conversion of rural cross-road to two rural T-junctions (100 m plus stagger)	Reduction (or increase) depends on traffic flows. Use accident prediction models for two T-junctions to determine the benefits.

Accident migration

Accident migration downstream of the treated site is normally not an issue in the urban road environment (50 km/h to 70 km/h). Accident migration is more prevalent on rural roads and in close proximity to the site being treated. The migration of accidents from the improved site down to the next curve or substandard road element (eg, narrow bridge) is more likely than migration to a similar element 20 km downstream.

To assess the possibility of accident migration, 1 to 2 kilometres either side of the study area should be considered. If road elements, such as low design speed curves (75 km/h or less), narrow bridges and railway crossings occur within this 1 to 2 kilometres, the analysis should assess whether an increase in travel speeds through the project area will increase accidents at the adjoining road elements. If there is an expected increase in the accident occurrence then either:

the negative benefits need to be included in the economic evaluation
improvements need to be made to downstream road elements to eliminate or reduce the accident migration
a reduced estimate of accident savings should be used in the analysis.

A similar exercise should be undertaken for a longer length, up to 5 km either side of the study area, if the speed change from the site improvements is expected to influence speeds and driver perception over a wider area. This may be the case for major realignments.

References

A6.6 - Typical accident reduction factors