TRANSPORTATION RESEARCH BOARD
82nd Annual Meeting
12-16 January 2003
Session 801
Roundabouts and Low-Vision Pedestrians
ROUNDABOUTS AND PEDESTRIANS WITH VISUAL DISABILITIES:
How Can We Make Them Safer?
Lal C. Wadhwa
Head, Civil and Environmental Engineering
JAMES COOK UNIVERSITY
Townsville Queensland Australia
P03-6042
Thursday 16 January 2003
8:00 am - 9:45 am - MarriottROUNDABOUTS AND PEDESTRIANS WITH VISUAL DISABILITIES:
How Can We Make Them Safer?
Lal C. Wadhwa
James Cook University
Abstract
Roundabouts are a common feature in Europe and Australia. They are becoming increasingly popular in the United States. Their wide-spread use is credited to increased safety, higher vehicle flow capacity and reduced delay to motorists at roundabouts compared to traditional signalised intersections.
However, certain physical and operating characteristics of roundabouts such as continuous flow, curvilinear layout, longer paths for pedestrians and no stopped phase at roundabouts give rise to problems for pedestrians with vision impairment. Possible solutions include improved design procedures relating to the location of cross walks, use of traffic signals, provision of detectable warnings, etc. as well as the development of mobility and orientation aids designed to help find and pursue a safe path through the roundabouts.
Roundabouts
The era of modern roundabouts began in the United Kingdom in 1956 with the construction of the first "yield-at-entry" roundabout. In 1966, a nationwide yield-at-entry rule launched the modern roundabout revolution. Australia and most other British-influenced countries soon built modern roundabouts. Countries such as the United States, where people drive on the right side of the road, were slower to follow, but many of these countries including France, Netherlands, Norway and Switzerland have been rapidly catching up since the eighties. For example, roundabouts have greatly increased in number in France since the adoption of the yield-at-entry rule on national routes in 1983.
There are an estimated 40,000 modern roundabouts worldwide and over 250 in the U.S. (with many more planned or proposed). The greatest roundabout ever built in the US. is in Clearwater Beach, Florida. It carries up to 58,000 vehicles and 8,000 pedestrians per day. Each roundabout has been an outstanding success. Roundabouts have been described as “the safest, most efficient and attractive form of traffic control in the world”. (http://www.roundabouts.net/newpage3.htm) Roundabouts slow all vehicles, provide refuges for pedestrians, and are the only traffic control device in which trees can be planted, fountains can bubble or spray, statutes can sparkle into the next century. When constructed as part of new road construction they are cheaper to build than signalised intersections. If used instead of traffic signals they save money. Their maintenance cost is almost zero. They require no electricity, no regular tune-ups, no annual replacements of parts, suffer no blackouts and cannot be blown away.
Modern American roundabouts have produced remarkable safety records. Since this experience is similar to the roundabout experience reported in other parts of the world, the safety of roundabouts compared to other forms of control has been well established. As a result, the number of roundabouts in the United States is expected to increase geometrically in the next decade.
Physical characteristics of roundabouts
· Roundabouts typically feature a circulatory roadway around a central island.
· The centre island is raised and landscaped
· Roundabouts have raised or painted splitter islands at each approach that separate the entry and exit lanes of a street.
· These splitter islands are designed to deflect traffic and thus reduce vehicle speed.
· Splitter islands also provide a pedestrian refuge between the inbound and outbound traffic lanes
· A roundabouts is a substitute for signalisation, as well as a traffic calming device
· Entering traffic yields to vehicles already in the circle.
Operational characteristics of roundabouts
A roundabout is a form of intersection design control that
¨ accommodates traffic flow in one direction around a central island
¨ operates with ‘give way’ or ‘yield’ control at the entry points, and
¨ gives priority to vehicles within the roundabout
Other operating characteristics include
- continuously moving traffic in clockwise direction (anticlockwise in US, Canada, and other right-side driving countries).
- vehicles yield at the edge of the circulating roadway until a gap in the circulating traffic flow becomes available.
- design speeds at the entry throat range from 25-35 km/h while the circulating flow speed are typically below 50 km/h
Capacity and safety of roundabouts
1. Increase in vehicle flow capacity
The primary objective of using roundabouts against other forms of intersection control is increasing the capacity of an intersection. The reasons for increase in capacity are the following:
¨ Since traffic yields rather than stops, smaller gaps can be accepted.
¨ Omission of lost time (yellow and red) results in higher capacity compared to signalised intersections, especially at isolated locations.
The maximum daily service volume of a single-lane roundabout varies between 20,000 and 26,000 vehicles depending on the % of right turns and the distribution of traffic between minor and major roads. A double-lane roundabout may service 40,000 to 50,000 vehicles per day.
2. Reduction in incidence of serious vehicle crashes
Roundabouts achieve crash reductions of 50 to 90 percent when compared to two and four-way stop control and signalised intersections and greatly reduced severity on those few crashes that do occur. Factors contributing to higher level of safety at roundabouts include
¨ single lane roundabouts result in increased safety because crossing distances are shortened
¨ lower overall speeds result in lesser severity
¨ pedestrian have to cross one direction of traffic at a time at each approach
¨ overall number of conflict points is reduced when compared to a traditional intersection treatment. At a four-way intersection there are 32 possible conflict points between vehicles and only eight at roundabouts. Pedestrians face six conflicts when crossing only one leg of the road whereas at a roundabout they only have two.
¨ more defined path for potentially conflicting vehicles
3. Reduction in delay and concomitant emissions
Signalised intersections with no retiming during off-peak period produce unnecessary delays to stopped traffic when gaps on the other flow are available. Roundabouts have lower delays than for equivalent volume signalised intersection. Factors which contribute to increase in flow capacity also contribute to shorter delays.
Analysis of Safety at Roundabouts
a. Factors contributing to increased safety
Vehicle safety
Modern roundabouts deflect and slow entering traffic.
There are a reduced number of conflict points compared to uncontrolled intersection.
Lower operational speeds yield fewer and less severe accidents.
Slower speeds because of intersection geometry reduce accidents
Netherlands achieved 95% reduction in injuries to vehicle occupants with roundabouts.
Pedestrian safety
A splitter island provides a refuge for pedestrians that will increase safety.
Low speeds reduce frequency and severity of pedestrian-vehicle accidents.
b. Factors contributing to reduced safety for pedestrians and cyclists
Difficult for visually impared pedestrians to interpret vehicle-pedestrian priority.
No stopped phase for pedestrians who want security of a signal.
Tight dimensions of roundabouts may create an uncomfortable feeling to bicyclists.
Longer paths increase travel distances for both pedestrians and bicyclists.
Roundabouts may increase delay for pedestrians seeking acceptable gaps to cross.
Safety record for bicycles and motorcycles has been mixed. According to one study in the United Kingdom, the percentage of accidents involving cyclists was higher in roundabout accidents compared to other forms of intersection (Layfield and Maycock, 1986). Another survey in 1989 of mini-roundabouts in England, Scotland, and Wales found that the crash-involvement rates of motorcycles and bicycles in 50-km/h speed zones was about the same for four-leg mini-roundabouts as for four-leg signalised intersections. However, the rate for cars at the mini-roundabouts was much lower than at the intersections.
International Comparison
Causes of accidents at roundabouts
The relative frequency of the different causes of accidents at roundabouts is shown in Table 1. In 1990, 202 accidents were investigated at 179 urban roundabouts in France. In Australia, the analysis is based on 492 roundabout accidents in Queensland. The U.K. data is from four-way roundabouts and single vehicle accidents not distinguished.
Type of accident / QueenslandAustralia / France / U.K.*
Single vehicle - entry / 5.2 / 10.0
Single vehicle - circulating / 10.4 / 16.3 / 8.2
Single vehicle - exit / 2.6 / 2.5
Rear end - entry / 16.9 / 7.4 / 7.0
Rear end - circulating / 1.2 / 0.5
Rear end - exit / 0.2 / 1.0
Multiple vehicles – entry, circulating / 50.8 / 36.6 / 71.1
Multiple vehicles – circulating, exiting / 6.5 / 5.9
Side-swap (multi-lanes) – entry, exit / 2.0 / 5.9
Side-swap - weaving (circulatory) / 1.7 / 2.5
Pedestrian on circulatory roadway / -- / 5.9 / 3.5
Others / 2.4 / 10.2
Source: B. Guichet (1992), Arndt (1998), Maycock and Hall (1984).
The single largest cause of roundabout accidents is the failure to yield at entry to circulating vehicles. The other major types are single vehicle crashes including vehicle running-off the circulatory roadway.
As shown in Table 1, three major crash types have been observed at roundabouts
1. entering-circulating crashes as a result of drivers failing to give way at entry.
2. rear-end collisions
3. single vehicle crashes
Together these three types constitute 75-95% of all roundabout crashes and are shown as the first seven categories of accidents at roundabouts.
Table 2: Proportion of major crash types at roundabouts
Country / Entering-circulating / Rear-end / Single vehicle / Sub-total / Crash types / Type of RoundaboutAustralia / 51 / 22 / 18 / 91 / All / Single & multilane
France / 37 / 13 / 28 / 78 / injury / Single & multilane
Germany / 30 / 28 / 17 / 75 / all / Single lane
Switzerland / 46 / 13 / 35 / 94 / all / Single & multilane
UK / 20-71 / 7-25 / 8-30 / 80 / injury / Single & multilane
Source: Brilon and Bondzio, 1998 in Roundabouts: An Informational Guide, FHWA, 2000
Crash rates
Crash rates are expressed as average annual number of injury crashes per roundabout and per million entering vehicles.
Table 3: Average annual injury crashes per roundabout and per million entering vehicles
Country / Crash frequency per / Remarksroundabout / million entering vehicles
Australia / 0.6
France / 0.15 / 0.045
UK / 3.31 / 0.275 / Many high volume, multi-lane roundabouts
US / 1.5 / 0.08 / Maryland, Florida, single-lane roundabouts
Source: Various
Reduction in crashes: Relative safety of roundabouts
Studies have conclusively shown that there is significant reduction in the frequency of crashes on roundabouts compared to other types of intersections. In other words, roundabouts are safer as shown in Table 4.
Table 4: Crash reduction: mean % reduction
Country / All crashes / Injury crashesAustralia / 41-61 / 45-87
France / 57-78
Germany / 36
Netherlands / 47
UK / 25-39
US** / 37 / 51
** limited data
Source: Guichet (1997), Garder (1998)
Since roundabouts may have been constructed to replace problem intersections with bad accident history, the rate of reduction may be overstated.
It may be possible to improve the design of some roundabouts and further reduce the crash frequency.
Pedestrian crashes at roundabouts
The risk of pedestrians being involved in a severe collision is low because of reduced speeds at roundabouts. U.K. studies have shown a 50% reduction in crashes at roundabouts compared to other forms of intersections (Maycock and Hall, 1984, Crown, 1998).
Analysis of Crash Data at Roundabouts in Queensland Australia, 1997-2002
There may be variations in the accident reporting processes and/or definition of an accident in different countries. In Queensland Australia, to qualify as valid, crashes must meet the following criteria:
¨ the crash occurs on a public road, and
¨ a person is injured, or
¨ the value of the property damage is:
(a) $2500 to property other than vehicles (after 1 December 1999)
(b) $2500 damage to vehicle and property (after 1 Dec. 1991- 30 Nov. 1999)
or
¨ at least one vehicle was towed away.
Analysis of Queensland crash data for 1997-2002 has been analysed with particular attention to pedestrians at roundabouts. However, information on visual disability is not available. It was thought that contributing circumstances might provide some insight into pedestrian disability but this has not been very revealing.
Table 5 shows the number of annual pedestrian crashes by severity levels at roundabouts in Queensland for the period 1997-2002. The number of roundabouts in the State road system is shown in Table 6. Number of crashes per roundabout is shown in Table 7
Table 5: Number of Pedestrian Crashes at Roundabouts by Severity, Queensland, 1997-2002
Severity
/1997
/1998
/1999
/2000
/2001
/2002
Fatal
/1
/0
/0
/0
/1
/0
Hospitalisation
/0
/6
/5
/3
/8
/4
Medical treatment
/7
/2
/6
/8
/10
/5
Minor injury
/2
/6
/3
/2
/2
/6
Total
/10
/14
/14
/13
/21
/15
Table 6: Roundabouts in Queensland, Australia
Roundabouts / 1997 / 1998 / 1999 / 2000 / 2001 / 20023-way / 69 / 72 / 74 / 87 / 88 / 92
4-way / 90 / 115 / 119 / 123 / 126 / 131
5- & 6- way / 11 / 14 / 15 / 12 / 11 / 10
Total / 170 / 201 / 208 / 222 / 225 / 233
Table 7 : Number of Crashes by Severity per Roundabout in Queensland
Severity
/1997
/1998
/1999
/2000
/2001
/2002
Fatal
/ 0.006 / 0.000 / 0.000 / 0.000 / 0.004 / 0.000Hospitalisation
/ 0.000 / 0.030 / 0.024 / 0.014 / 0.036 / 0.017Medical treatment
/ 0.041 / 0.010 / 0.029 / 0.036 / 0.044 / 0.021Minor injury
/ 0.012 / 0.030 / 0.014 / 0.009 / 0.009 / 0.026Total
/ 0.059 / 0.070 / 0.067 / 0.059 / 0.093 / 0.064Total number of crashes by severity level for the period 1997-2002 at roundabouts and intersections are shown in Table 8. The data in this Table shows the relative proportion of fatal and severe injury (hospitalisation) accidents based on the type of intersection.
Table 8: Queensland Data Analysis (1997-2001)