All-way stop-control alternative
When cross street traffic volumes are heavy enough to meet the MUTCD warrants for AWSC control, roundabouts become an especially attractive solution because of their higher capacities and lower delays. The selection of a roundabout as an alternative to AWSC should emphasize cost and safety considerations, because roundabouts always offer better performance for vehicles than AWSC, given the same traffic conditions. Roundabouts that are proposed as alternatives to stop control would typically have single-lane approaches.
A substantial part of the benefit of a roundabout compared to an all-way stop intersection is obtained during the off-peak periods, because the restrictive stop control applies for the entire day. The MUTCD does not permit stop control on a part-time basis. The extent of the benefit will depend on the amount of traffic at the intersection and on the proportion of left turns. Left turns degrade the operation of all traffic control modes, but they have a smaller effect on roundabouts than on stop signs or signals.
The planning level analysis that began earlier in this chapter may be extended to estimate the benefits of a roundabout compared to AWSC. Retaining the previous assumptions about the directional and temporal distribution factors for traffic volumes (i.e., K=0.1, D=0.58), it is possible to analyze both control modes throughout an entire 24-hour day. Only one additional set of assumptions is required. It is necessary to construct an assumed hourly distribution of traffic throughout the day that conforms to these two factors.
A reasonably typical sample distribution for this purpose is illustrated in Exhibit 3-9, which would generally represent inbound traffic to employment centers, because of the larger peak in the AM period, accompanied by smaller peaks in the noontime and PM periods. Daytime off-peak periods have 4 percent of the AADT per hour, and late-night off-peak periods (midnight to 6 AM) have 1 percent.
The outbound direction may be added as a mirror image of the inbound direction, keeping the volumes the same as the inbound during the off-peak periods and applying the D factor of 0.58 during the AM and PM peaks. This distribution was used in the estimation of the benefits of a roundabout compared to the AWSC mode. It was also used later for comparison with traffic signal operations. For purposes of estimating annual delay savings, a total of 250 days per year is assumed. This provides a conservative estimate by eliminating weekends and holidays.
The comparisons were performed using traffic operations models that are described in Chapter 4 of this guide. The SIDRA model was used to analyze both the roundabout and AWSC operation, because SIDRA was the only model readily available at the time this guide was developed that treated both of these types of control. SIDRA provides an option to either include or omit the geometric delay experienced within the intersection. The geometric delay was included for purposes of estimating annual benefits. It was excluded in Section 3.4.4.1 that dealt with driver-perceived approach delay.
The results of this comparison are presented in Exhibit 3-10 and Exhibit 3-11 in terms of potential annual savings in delay of a single-lane roundabout over an AWSC intersection with one lane on all approaches, as a function of the proportion of left turning traffic for single-lane approaches for volume distributions of 50 percent and 65 percent on the major street, respectively. Each exhibit has lines representing 10 percent, 25 percent, and 33 percent left turn proportions.
Note that the potential annual benefit is in the range of 5,000 to 50,000 vehicle-hours per year. The benefit increases substantially with increasing AADT and left turn proportions. The comparison terminates in each case when the capacity of the AWSC operation is exceeded. No comparisons were made beyond 18,000 AADT, because AWSC operation is not practical beyond that level.