Austroads report released - Bicycle Safety at Roundabouts

Go here to download the report, and to sign up for notification of new reports on cycling

It's a good report in that it scientifically analyses the causes of bicycle crashes at a host of sites, and provides detailed recommendations for these problem NSW roundabouts:

  • Eastern Avenue – Tresidder Avenue, Kingsford
  • Barnstaple Road – Ingham Avenue, Five Dock
  • Heffron Road – Banks Avenue, Pagewood
  • Anzac Parade – Rainbow Street, Kingsford
  • Phillip Street – Young Street, Redfern

Interesting fact: 23% of crashes at (Victorian) roundabouts involved at least one cyclist (p39)

Call to action:  Ask your local council traffic committee what they are doing about improving cyclist safety at roundabouts


Published: 01 May 2017 
This report investigates how the geometric design components of a roundabout may contribute to bicycle crashes.

An Australian and New Zealand crash analysis found that most of the crashes occurred at urban local road roundabouts, in 50 km/h speed limit zones. The crashes predominantly occurred on the circulating lane near the entry for an approach road and were right-adjacent type crashes.

The study included an in-depth investigation of 17 roundabouts across Queensland, New South Wales and Victoria. A geometric analysis identified that the entry geometry of the roundabouts investigated would permit relatively high entry speeds, in excess of the target speed of less than 30 km/h. This target speed was adopted for analysis purposes, however, further investigation to determine an appropriate speed to prevent or minimise fatal and serious injury outcomes for crashes involving motor vehicle and cyclists is needed.

The motor vehicle speeds on the entry and circulating lanes were estimated using the ARNDT crash prediction model, however the model was developed on rural roads and so the application of this model to urban local roads requires verification. For the purposes of this investigation, the ARNDT model was used to assess geometric alignments to achieve lower approach speeds and it was found that a roundabout with a radial-type of alignment, used in countries in Europe, achieved approach and circulating speeds of less than 30 km/h.

Sight distances were examined and it was found that the available sight distance to vehicles approaching from the right did not meet the design requirements. There is some research which indicates that restricting the sight distance on the approach to a roundabout reduces the approach speeds of vehicles, however, this requires further investigation to develop design criteria.

The report recommends further investigation into motor vehicle/cyclist crash outcomes and the effect of restricting sight distance on the approaches to a roundabout, and the development of design guidance for urban local road roundabouts.

Views: 906

Reply to This

Replies to This Discussion

Hmmm interesting stuff, given that:

a) roundabouts are generally crap for cyclists


b) their principal purpose is to slow and organise motor traffic

putting it another way: if you slow cars to 30km/h or less you can remove roundabouts

My experience is that roundabouts don't slow traffic, they operate on a "first come first served basis", so best to get in there as quickly as possible...

Sometimes getting there first doesn't seem to matter much.

But at least in this instance, if I hadn't stopped for them it would have been a quick pick-up for the hospital...

Drivers maybe don't expect anyone to be coming out of Kelly St that way in the contra flow lane. Takes em a while to compute, esp drivers coming down Bay St. Good fun if you are aware.

And only look to the right, don't worry about anyone on your left.

" What is in motorists’ minds as they approach roundabouts? It is theorised that motorists approach roundabouts with the goal of slowing as little as possible and with assumptions that: (1) they probably won’t stop, (2) only conflicts from the right need be considered, and (3) potential conflicts can be assessed with a brief glance to the right. Bicycle lanes without cyclists can be used to reduce deflection to help maintain higher speeds"

From Bob Cumming, 2012  Paper to ACRS. He reckons too many cyclists take a straight line through roundabouts too, and it would be safer if we took the centre of the lane and travel in curve. Also take the lane well before entering, so you are where a motorist is most likely to be checking for other motorists.

The straightened option often beats the curved because the surface is slippery.

MGIF  is often applied at roundabouts

Indeed. I do tend to make a point if I can by placing myself in front of MGIF while they wait. Easier at signals, where you know they are blocked until the signal changes.

restricting sight distances would give motorists a real excuse for SMIDSY.

Seems remiss not to mention Dutch style designs of intersections as alternatives to roundabouts.

The Dutch also put crossings in for non vehicle users because the main legal implication of a roundabout in Australia is that a driver doesn't have to give way to pedestrians when turning which is a horrible legal outcome and makes just crossing a side road tiresome because you have to give way to through traffic not crossing your path as well. 

Its not uncommon for a pedestrian to be fence diverted in Australia 20m+ down the side roads just to walk along a road with roundabouts.


Update: I called Canada Bay Council a little while ago, and they said that Barnstaple – Ingham roundabout improvement is before the traffic committee. 


© 2020   Created by DamianM.   Powered by

Badges  |  Report an Issue  |  Terms of Service