32

FIGURE 14 Entry capacity for a two lane roundabout with a 60 m inscribed diameter and a 4 m entry lane width(44).

AUSTROADS methodology in that it takes intoconsideration the effects of the origin destination pattern within the roundabout and the queuing characteristics of the approach flows.

SIDRA estimates critical gaps and follow upheadways as a function of roundabout geometry, as well as the circulating and entry flow rate characteristics. However, the user can also specify known critical gap and follow up headways to take into consideration local conditions. SIDRA includes an option describing the roundabout performance using accepted U.S. definitions, such as for delays and levels of service(46,47).

Great Britain

Capacity Formula

The capacity formula used in Great Britain is a statistically derived empirical formula based on a large number of measurements of capacity at saturated roundabouts. It was developed by the Transport Research Laboratory (TRL) and has the following form:

C = k ( F f c Q c )

w h e r e k , F , a n d f c a r e c o n s t a n t s d e r i v e d f r o m t h e g e o m e t r y o f

the roundabout, and Equation, LR 942) (45).

# Q_{c }is the circulating flow (Kimber's

In addition to the geometric parameters used in most other capacity methods (inscribed diameter, entry width, circulating width, as shown in Figure 5), the British method incorporates the following geometric variables, shown in Figure 15:

The Approach Half Width, v, is measured from the right

curbline, along a normal to it, to the centerline or left edgeline at a point upstream of the flare.

Average Effective Flare Length, l', is shown in Figure 8.

The right edge of pavement would follow the line GFD if there were no flare. GFD is the upstream half width v away from the centerline (or, in the case of a raised median, from the

median curb). BA is the normal to the curb along which the entry width e is measured, and its length ise. The length of BD is (e v), and the length of BC is(e v)/2. The average effective

flare is CF, a curve

(e v)/2 away from the right curb. The

length of CF is l', the average effective flare length.

Sharpness of Flare, S, is defined by the relationshipS=

1.6

(e v)/ l'. It is a measure of the rate at which extra width is

developed in the entry flare. Large values ofS correspond to short, severe flares, and small values ofS correspond to long, gradual flares.

Entry Angle, , represents the conflict angle between

entering and circulating streams of traffic.

Entry Radius, r, is measured as the minimum radius of

curvature of the right curb at entry. For some designs, the arc of minimum radius may extend into the following exit, but this is not important if half or more of the arc length is within the entry region.

Vane Island is a painted island that divides lanes entering

a roundabout. A vane island provides entry deflection for the right lanes when the central island is too small to provide this deflection.

The approach used for the Britis h regression equation is based on one of the largest sets of data points collected on roundabout capacities. It provides estimates of the effects of different geometric parameters of British roundabouts. Statistical tests have been performed to confirm thesuitability of the parameters.

Figure 16, extracted from the RODEL software manual (48) shows the relationships between the entry capacity and geometric characteristics.

Performance analysis methods are also given for mini roundabouts (49), with efficiency factors (K) given for single mini roundabouts and for double miniroundabouts, with three, four, and five arms. Total design flows for all approaches are

recommended to be less than 2,500 vehicles for three roundabouts and less than 2,000 vehicles for four roundabouts.

arm arm

ARCADY Software

The computer program ARCADY (Assessment of Roundabout CApacity and DelaY) is the program originally developed by the Transport Research Laboratory to calculate capacities according to the British formula (LR 942) (40). In addition to predicting capacities, queues, and delays, it also predicts crash frequencies as a function of geometry, thus permitting the user to design for safety as well as for capacity. These predictions can be used to test design options for new roundabouts and modifications to existing ones. The program has the ability to predict the variability of queues and delays. VISUAL ARCADY/4 for Windows was released in 1996. Pedestrian crossings can be included in the analysis and queue lengths can be viewed, animated, and printed.

RODEL Software

RODEL (ROundabout DELay) was developed by R.B. Crown in 1987 (48). It is designed to facilitate experimentation