The objectives of this project are to identify the geometric design factors associated with bicycle crashes at roundabouts and identify design options that may reduce the incidence or severity of these crashes. A crash analysis was undertaken covering Australia and New Zealand to identify the crash characteristics, which identified that 93% of crashes occurred in speed zones of 60 km/h or less, with 63% occurring on roads with speed zones of 50 km/h or less. The most common type of crash was the adjacent direction crash (entering motor vehicle colliding with a cyclist on the circulating carriageway), accounting for 67% of the crashes followed by vehicles travelling in the same direction crashes comprising 13% of the crashes. From the crash analysis a representative sample of roundabouts in Queensland, New South Wales and Victoria were identified for detailed investigation to identify contributing factors to the crashes occurring at the roundabouts. Twelve of the roundabouts were on urban local roads (the most common crash location), two on urban arterial roads and three on rural arterial roads and 88% of the crashes occurred on the circulating lane at an entry to the roundabout involving a motor vehicle entering the roundabout, undertaking a turning or straight through movement. Geometric information was obtained for each of the roundabouts and analysed to assess the affect the geometry may be having on the crashes. This analysis identified that the local road roundabouts typically had entry curve radii ranging between 40 m and 50 m, which would cater for entry speeds greater than 40 km/h, which is greater than the target speed of [less or equal] 30 km/h for a road space shared by motor vehicles and cyclists. Whilst this target speed has been adopted for this project, further investigation is needed to identify the relationship between motor vehicle speeds and bicycle crash outcomes. Travel path curvature at the entry was identified as a key characteristic of the roundabout that influences possible vehicle speeds at the roundabout. To estimate the vehicle speeds on the travel path curves, the ARNDT model and the horizontal curve equation were used. Both of these methods have limitations for application to the roundabouts analysed and so the speeds obtained were considered to be indicative only. Sight distance was examined and it was found that the requirement to meet the sight distance near the holding line was met (Criterion 2), and on the urban local roads the sight distance for drivers approaching the roundabouts (Criterion 3) was not met. Some research has shown that restricting sight distance (Criterion 3) can reduce the approach speeds of motor vehicle but this needs more detailed investigation and development of criteria for application. Possible treatments were identified, with the entry path curvature being the main component that needed to be increased to slow the entering vehicles. On local roads, with smaller design vehicles, the entry curve radius could be reduced to achieve a speed of [less or equal] 30 km/h, but on the arterial road roundabouts, reducing the entry speeds was not achievable due to the larger design vehicles on these roads. On the local roads, the alignment of the roundabout to achieve the desired entry speed was similar to the alignment of a radial-type roundabout. Suggestions are provided to amend the nominated Austroads Guides, however guidance on designing a roundabout, particularly an urban local road roundabout for a specific speed, is a key task for the future. (Author/publisher)
Samenvatting