We will begin our work by considering the operations
of the intersection of U.S. 95/Styner-Lauder Avenue under two kinds of
control, the existing stop-sign control and the proposed signal control. In problem 1, we will focus our attention only on this intersection,
without consideration of the effects of the adjacent intersections. This
approach will allow us to develop an initial assessment of whether
stop-sign control is better than signalized control, but we must keep in
mind that other factors (including, for example, the proximity of
adjacent intersections and the manner in which they are controlled) can
also have a very big influence on our final recommendation.
These effects will be considered later in Problem
2.
For this particular analysis,
we must first complete two computations for the
existing traffic volume conditions. Then, we will consider the
projected future volumes. We can pose three questions that will
guide the computations:
What is the level of service at the existing TWSC intersection at U.S. 95/Styner-Lauder?
What would the level of service be if the intersection were signal
controlled?
What would the level of service be at the
intersection under future projected traffic conditions?
Discussion: Take a few minutes to consider these questions. When you are ready, click continue below to proceed.
Problem 1:
U.S. 95/Styner-Lauder Avenue Intersection
In this problem, you will consider the following issues as
you work through the computations for three sub-problems:
Sub-problem 1a: Analysis of
the existing TWSC intersection
Basic TWSC intersection methodology
Peak hour factor
Data requirements for TWSC intersection analysis
Critical gap
Follow-up time
Performance measures for TWSC intersection analysis
Sub-problem 1b: Analysis of
the proposed signalized intersection Basic signalized intersection methodology
Data requirements for signalized intersections
Default values for signalized intersections
Saturation flow rate assumptions
Right turn on red
Lost time
Lane widths
Signal timing issues
Sub-problem
1c: Analysis of future conditions
Appropriate assumptions for future analysis
Forecasting future volumes
Uncertainty analysis
Appropriate level of detail
Accuracy vs. precision
Comparing control types
Sub-problem 1a: Analysis of the existing TWSC intersection
Step 1. Setup
The analysis of transportation facilities can be complex,
requiring both the skilled use of standard computational procedures such as the
HCM as well as good judgment in the use and interpretation of results from the
computational procedures. In the setup of this problem, we will consider
the data, the issues, and the tools that are relevant to this problem. In
particular, we will consider the analysis of the U.S. 95/Styner-Lauder Avenue
intersection under two-way stop-control.
Here are some issues to consider as you proceed with the
analysis of the existing intersection and its performance:
Which tool or tools from the HCM should be used for the
analysis of a TWSC intersection?
What data are required for the analysis?
What default values should be used?
What time periods should be analyzed?
What measure should be used to determine the
performance of the intersection?
What other factors should be considered?
Discussion: Take a few minutes to
consider these questions. Click continue when you are ready to
proceed.
Sub-problem 1a: Analysis of the
existing TWSC intersection
Let's discuss each of these issues and how each affects
the operational analysis that we are about to complete.
Which tool or tools from the HCM should be used for
this analysis? The analysis tools for unsignalized intersections are
contained in chapter 17 of the HCM 2000. Chapter 17 includes analysis tools
for two-way stop-controlled (TWSC) intersections, all-way stop-controlled (AWSC)
intersections, and roundabouts. In this analysis, we will use the procedure
for TWSC intersections. Since we are interested in existing conditions at
the intersection, at a fairly detailed level, we will use the operational
analysis procedure in chapter 17.
What data are required for the analysis? Since we
are interested in determining the average control delay that drivers
experience at the existing intersection, we will use the operational
analysis method that is included in chapter 17 of the HCM. The following
data are needed to conduct an operational analysis:
traffic volumes for each movement
number of lanes and lane configuration for each
approach
proportion of heavy vehicles
peak hour factor
grades on each approach
other geometric features (channelization, medians)
We should note one potential point of confusion in the
aerial photograph
for this intersection. While the eastbound approach striping does not show
this explicitly, this approach does operate as if it is configured with an
exclusive left turn lane and a shared through/right turn lane. These
details are always important to check in the field.
Exhibit 1-6. Aerial Photograph of
the Intersection of U.S. 95 with Styner Ave/Lauder Ave
The intersection of U.S. 95/Styner Avenue/Lauder Avenue is a
four leg TWSC intersection. The westbound approach, Styner Avenue, serves
a growing residential area in the eastern section of the city. Lauder
Avenue serves as a minor entrance to the University of Idaho and access to
student and faculty residences. A gas station/mini-mart is located on the
northwest quadrant of the intersection. Other auto-oriented businesses are
located on the other quadrants.
Note: You can see other views of the intersection approach by
moving your mouse to the approach and clicking on the approach.
Page Break
ID# C101A03
Sub-problem 1a: Analysis of the
existing TWSC intersection
Following are other points to consider regarding the
required data:
Traffic volumes for each movement were obtained
from two sources. The existing traffic volumes (See
Exhibit 1-7) reflect the results of a manual traffic count conducted
during the afternoon peak hour. The future traffic volumes (see
Exhibit 1-8) reflect a horizon year ten years hence and come from the
application of an historical growth trend that has been observed on US 95
over the past ten years. We are assuming, based on our own familiarity
with the area, that there is no significant volume of pedestrian crossings
at the intersection which might impede the flow of vehicles.
The number of lanes and lane configuration
for each approach can be determined from the
aerial photograph.
Heavy vehicles: note that the data provided in
Exhibits 1-7 and 1-8 does not include any breakout of heavy vehicles. Even
though the proportion of heavy vehicles is a required input to our
analysis, we sometimes find in the real world that not all the information
we need has been previously collected. One option would be to return to
the field to collect additional field data, but this can be expensive,
time-consuming, and outside the realm of practicality for some purposes.
In this case, we happen to know from our familiarity with the area that
truck volumes are negligible during the peak hour, and so it will be
reasonable to assume that there are no heavy vehicles (i.e.,
trucks, through buses, and large recreational vehicles) in the traffic stream.
The grade of all approaches is known to be
level, and so a 0% grade is assumed for each approach. The grade of an
approach has an effect on the critical gap for the movement. A downgrade
approach to the intersection reduces the critical gap because vehicles
accelerating from a stop have an easier time entering or passing through
the traffic stream.
The peak hour factor
(PHF) is a measure of the traffic demand fluctuations within
the peak hour. For this problem, we are assuming that a value of
1.0 for the peak hour factor is appropriate. This means that there is no
variation in the volume during the hour. In reality, field studies show
that the peak hour factor for this intersection is less than 1.0,
indicating that there is indeed a variation of traffic volumes across the
hour. We are using a value of 1.0 for the PHF because we are interested in
evaluating average conditions across the hour; in other situations
where the analyst is interested in evaluating conditions during the peak
15-minutes of the hour, then the field-observed value for the PHF should
be used.
The intersection has no median that vehicles from
the minor street can use as a refuge as they cross U.S. 95. The minor street approaches (Styner and Lauder) have no
storage (flared
approaches) at the stop line in which right turning vehicles can
by-pass through vehicles that are already waiting at the stop line.
Sub-problem 1a: Analysis of the
existing TWSC intersection
What default values should be used?
Driver behavior at a TWSC
intersection is described by two parameters, the
critical gap
and the follow up
time. The HCM provides default values that represent average
values from a number of measurements made at sites throughout the United
States for critical gap and follow up time. However, conditions at the site
(unusual geometric features, high volumes causing more aggressive driver
behavior) that you are studying may produce values that are different than
these default values. It is always better to use values that are estimated
from the site that you are studying, if these values can be measured.
However, it should be pointed out that critical gaps cannot be
directly measured in the field. Rather, they are estimated using
statistical procedures based on the distribution of gaps that are accepted and
rejected by drivers in the field. An appropriate but somewhat complex
method for doing this was presented by
Troutbeck
in 1992. A rough method to check the validity of
the critical gap would be to measure the follow-up time, which can be done
easily, and to estimate the critical gap using the approximate relationship, tf/0.6.
What time periods should be analyzed? The data
that have been collected for this site represent the volumes for
the afternoon peak period. If there are other peak times during the day, such as
the morning peak or sometimes a midday peak, these should also be included in an
operational analysis.
What measure should be used to determine the
performance of the intersection? The level of service for a TWSC
intersection is based on control
delay. Control delay is the primary measure of effectiveness for TWSC intersections and will be used as the parameter to compare the
performances of the various alternatives that we consider in this problem.
In fact, control delay is also used as the primary measure of effectiveness
for signalized intersections. In addition, we will consider
two other performance measures, each describing another aspect of the
operation of the intersection. The v/c, or
volume/capacity ratio
is useful for showing how close the intersection is to capacity operation. The queue length provides
a way for the analyst to determine the adequacy of the geometric design of
the facility by examining the projected length of a queue compared with
the length of turning or storage lanes.
Sub-problem 1a: Analysis of the
existing TWSC intersection
What other factors should be considered? One of the
important elements in determining the capacity of each minor movement at a
TWSC intersection is the conflicting flow, which is defined as that flow to
which the minor movement must yield the right of way. For this
problem, we might pay particular attention to the northbound right turn
movement. If you study the
aerial photograph
for this approach (link), you will note that there is a flare, providing a
short lane for right turning vehicles.
We should also study whether or not right turning
vehicles commonly used this lane, and, if so, whether this results in a
lower conflicting flow faced by the minor street movement. At first glance,
you might conclude that the right-turning traffic shouldn't even be included
as one of the conflicting traffic streams since there is no physical
conflict with any of the minor movements. However, field studies have shown
that minor street drivers are reluctant to initiate their movement until
they can be sure that the approaching vehicle is in fact going to make a
right turn. It is for this reason that the HCM includes the major street
right-turn volume as a potential conflict; the 0.5 factor associated in some
instances with this right-turn volume was determined through field studies
to be an appropriate "calibration factor" that accounts for the hesitancy
minor street drivers often exhibit when confronted with an apparent
right-turning vehicle. It should be noted that the HCM suggests that the major street right turn movement from a separate lane be entirely
excluded from the conflicting flows for the minor street movement because,
in such cases, the minor street drivers are usually quite confident that the
approaching vehicle is actually going to make a right turn.
In the case of this particular sub-problem, the flare or
right-turn lane is very short, and so we will assume that the effect of the
right turning vehicles is accounted for with the 0.5 factor assumed in the
HCM procedure. For a complete discussion of the conflicting flow rates, see
Exhibit 17-4 of the HCM.
Review the
traffic data to see
the existing afternoon peak period volumes for this site.
The table shown above provides a substantial amount of information
regarding the forecasted operation of the intersection, including the estimated
capacity, v/c ratio, queue length, and delay. What information from the
table is most
relevant to our analysis?
Discussion: Take a few minutes to review
the data presented in this table. Document what you consider to be some of
the key aspects of these results. When you are ready to continue, click
the continue button below to proceed.
Sub-problem 1a:
Analysis of the existing TWSC intersection
Level of service (LOS) for a TWSC intersection is determined by
the control delay and is defined for each movement, but LOS
alone does not tell the whole story! Besides LOS, other performance measures
that should also be considered include the volume/capacity ratios and the
estimated queue length. Let's investigate each of these measures of
effectiveness in greater detail (see
Exhibit 1-9).
First, note that the left turn movements on the major
street (NBLT and SBLT, movements 1 and 4), experience acceptable
delay. The HCM methodology estimates that both movements will experience
level
of service A, with less than 10 seconds of control delay per vehicle. We
can also see that the movements on the minor street
approaches (westbound on Styner and eastbound on Lauder) experience moderate
to high levels of delay. Both left turn movements (movements 7 and
10), for example, operate at level of service E with delays of 36 and 47
seconds, respectively.
The volume/capacity ratio is also important to consider
because it tells us how close we are to capacity for each
movement. Here, the v/c ratio is less than 0.60 for all movements. Thus, there is ample capacity available for the existing conditions.
Queue length is always an important consideration at an
unsignalized intersection, and especially when it is necessary to
determine the adequacy of turning bays or when there is the possibility of a
queue spilling back into the adjacent upstream intersection. In the
case of the Styner-Lauder/U.S. 95 intersection, the
95th-percentile queue length estimates are less than four vehicles, meaning that there is
sufficient space to store vehicles as they are waiting to enter the
intersection. Even so, the queue length requirements for the right turns
from the minor street (movements 9 and 12) are 3 vehicles. This is a length that some drivers might consider to be
excessive (even though it only has a 5 percent probability of occurring), and especially so for drivers in a smaller community like Moscow.
Sub-problem 1a:
Analysis of the existing TWSC intersection
Note in the
Exhibit
1-9 how delay and level of service
are reported. First, the results are reported for each movement. Then, the results for the minor street approaches are reported. What
is the value in showing both results? The results for each movement
allow the analyst to see the operation of the intersection at the smallest
level. If any one movement is experiencing a high delay, such as
movement 10, attention can be paid to resolving that problem. The
approach data provides a broader look at the intersection performance and
is often useful when comparing the operation of a number of intersections. Finally, you should note that the delay and level of service for the
entire intersection is not reported. This is in part due to the
assumed zero delay for the major street through and right turn movements
and the importance of monitoring the operation of the minor street
movements. In any case, it is important to remember that,
for an unsignalized intersection as
a whole, LOS is undefined.
In summary, a review of the performance characteristics
indicates that the intersection is operating acceptably, but in a marginal
range with respect to both control delay and queue lengths.
Now that we've completed the analysis of the existing TWSC
intersection, we'll proceed to the analysis of the intersection under the
proposed signal control.
Sub-problem 1b: Analysis of the Proposed
Signalized Intersection
Step 1. Setup
We will now consider the operation of the U.S. 95/Styner-Lauder
intersection under signal control. Here are some issues to consider as you proceed with
this analysis of the
existing intersection.
Which tool or tools should be used for this analysis?
What data are required for the analysis?
What default values should be used?
What time periods should be analyzed?
How do we construct a signal timing for a proposed traffic signal using
the HCM?
Discussion: Take
a few minutes to consider these questions. Click continue when you are ready to
proceed.
Sub-problem 1b: Analysis of the Proposed
Signalized Intersection
Let's discuss each of these issues and how each affects
the operational analysis that we are about to complete.
Which tool or tools should be used for this analysis?
There are a number of tools that the analyst might consider to solve
this problem. A hand-calculated critical movement analysis or the HCM
planning analysis might be considered to get a quick assessment of the
intersection performance and operation. These are both discussed at
greater length in Problem 6 in
this case study. Other non-HCM tools such as Synchro, CORSIM, or
aaSIDRA might be used. Finally, the HCM operational analysis method
for signalized intersections, which produces estimates of control delay, v/c
ratio, and queue length, is a potential tool for this sub-problem. We
will use this latter tool for this analysis. Why did we make this
choice? When we are considering conditions in which demand does not
exceed capacity, or the flows from one intersection do not spill back and
affect the adjacent upstream intersection, the HCM operational analysis
method provides a useful and easily applied tool. In addition, we are
conducting a comparative analysis between signalized and unsignalized
intersection operational characteristics, so we are looking for procedures
that allow this kind of comparison to be made easily and consistently. We will explore the use of non-HCM tools in problems 3 and 4 of this case
study. We will address some of the major issues that are often
confronted by analysts using the HCM methodology in this and subsequent
problems in this case study.
The analysis tools for signalized intersections are
contained in chapter 16 of the HCM 2000. The
computational methodologies included in this chapter are complex and cover a
wide range of conditions that are often observed in the field.
Sub-problem 1b: Analysis of the Proposed
Signalized Intersection
What data are required for the analysis? The
following data are needed to conduct an operational analysis:
Volume related information, including traffic volumes
on each intersection approach.
Geometry related information, including the number of
lanes and lane configuration for each approach.
Traffic signal related information, including signal
phasing and timing.
What default values should be used? There are
a number of traffic and geometric conditions for which default values have
been established. Each condition must be considered and a decision
made regarding whether field data should be used to override these default
values:
The default ideal saturation flow rate is 1,900
passenger cars per lane per hour of green. We will use this default
value as a starting point in our analysis. But field data should
always be collected if possible for the particular site that you are
studying. We should revisit this assumption if we find that the
intersection's critical v/c ratio is near a threshold in which mitigation
is required. In this case, it would be worthwhile to collect
additional field data on some of the more critical factors affecting the
outcome of the analysis, including flow rates, green splits, peak hour
factor, and lane utilization.
Are there a significant number of vehicles that turn
right during the red phase? If so, and if a field count of the number of
right-turn-on-red vehicles is available, then these vehicles should be excluded from
the analysis. The analysis procedure contained in the HCM is based on data
collected at a variety of sites, including those where right-turns-on-red
regularly occurred, an so a "normal" number of right-turns-on red is
already implicitly assumed in the procedure. Therefore, if a field count
of the number of right-turns-on-red is not available, then no
special adjustment should be made. The data available to this analysis
does not include a separate count of right-turns-on-red, and so no
additional adjustment will be made.
Are there parking maneuvers
within 250 feet of the intersection
approaches? Parking is not currently allowed on the intersection
approaches for this problem. In situations where parking is allowed,
remember that parking and unparking are both counted as separate parking
maneuvers.
Do buses stop along the intersection approaches? There are no bus routes that
currently serve this area. It is
important to clarify that this factor relates to local buses stopping at
the intersection and not through buses which are included as part of the
heavy vehicle data.
Do bicycles and pedestrians impede vehicular flow? Field observations show that
both bicycle and pedestrian traffic is minimal. If this were not the case,
then both would need to be included through field observation or
estimation.
How will the signal operate and what will be the lost
time at the intersection? Lost time is a factor that is commonly
misunderstood by users of the HCM signalized intersection procedure. This
issue will be discussed in detail in
sub-problem 4a.
Do narrow lane widths impede traffic flow? Field
studies have shown that the existing lane widths do not have an effect on
the final saturation flow rate of this intersection. Each approach
was measured in the field as 12 feet wide, which is the standard width and
therefore no adjustment to saturation flow rate is necessary.
Do heavy vehicles affect the performance of the
intersection? Heavy vehicles are not present in sufficient numbers to
affect the saturation flow rate at the intersection.
Sub-problem 1b: Analysis of the
Proposed Signalized Intersection
What time periods should be analyzed? The
data that have been collected for this site represent the peak 15-minute
flow rates for the afternoon peak period. A 15-minute analysis is used here
(as opposed to the one-hour analysis used for the existing stop sign control
described in sub-problem 1a) because signalization is an expensive
mitigation option that almost always adds to total system delay. Therefore,
we want to evaluate it at a higher standard -- that is, we want to be sure
it is able to perform adequately even during the peak 15 minutes of the peak
hour. If there are other peak times
during the day, such as the morning peak or sometimes a midday peak, these
should also be included in an operational analysis. Since we are
considering a decision that may take several years to implement, we will
also consider traffic conditions that are expected over the next ten years. Review the traffic data
to see the existing and projected afternoon peak period volumes for this
site. This is also discussed in
Problem 3
when we consider an analysis of event traffic following a football game.
How do we construct a signal timing plan for a proposed
traffic signal using the HCM? The construction of a timing plan
for a signalized intersection can be a complex process, though there are
also some simple approaches that give very reasonable first-approximations. In this problem,
we will assume that the proposed new signal will operate in fixed time mode,
and the methods included in Appendix B of Chapter 16 of the HCM can be used
to determine the signal timing plan for this condition. There are
other tools that can be used for developing signal phasing and timing plans
including the Institute of Transportation Engineers (ITE) Traffic
Engineering Handbook, as well as the ITE guidelines for left turn phasing.
The critical movement analysis technique is another good way to quickly
develop a reasonable signal phasing and timing plan.
The decision to signalize the intersection or not does not
depend on whether the traffic controller is fixed time or actuated. To
simplify this analysis, then, we have chosen to assume that it will operate
in fixed time mode. In
Problem 4, we
will illustrate how the HCM procedure considers actuated controller
operational parameters.
The following signal timing was produced for this
sub-problem, using the methods of Appendix B, chapter 16, of the HCM.
Exhibit 1-10. Signal Timing Data for sub-problem 1b
Phase
Movement
Green
Yellow
All red
1
NB/SB
35
4
1
2
EB/WB
15
4
1
Click continue below to proceed with the analysis.
Sub-problem 1b: Analysis of the
Proposed Signalized Intersection
Step 2. Results
When we apply the methods of chapter 16 of the HCM, we
obtain estimates of capacity, v/c ratio, queue length, delay, and level of
service for each lane group. In addition, capacity, delay, and level
of service are produced for each approach. Also, the delay and level
of service are produced for the intersection as a whole.
The results from this calculation are summarized in
Exhibit 1-11 for the existing volumes and signalized intersection control.
Exhibit 1-11. U.S. 95/Styner-Lauder Avenue (Dataset2)
Lane Group Capacity, Control Delay, and LOS Determination - Existing
Volumes (Signal Control)
Discussion: Take a few moments to reflect on the analysis results
presented in this table and consider the performance measures most relevant
to an assessment of the overall operational characteristics of the
signalized intersection. Click on continue when you are ready to proceed.
Sub-problem 1b: Analysis of the
Proposed Signalized Intersection
A review of the information contained in
Exhibit
1-11 provides the following insights
on the current operation of this intersection, if it were to be controlled
by a traffic signal:
The intersection as a whole operates at level of
service B.
The eastbound and westbound approaches (formerly the
minor street approaches, Styner Avenue and Lauder Avenue) operate at level
of service C, with control delay varying between 20 and 25 seconds per
vehicle.
The northbound and southbound approaches (on U.S. 95)
both operate at level of service A, with control delay varying between 6
and 7 seconds per vehicle.
No lane group operates at less than level of service
C.
Every lane group has a volume/capacity ratio under
0.50.
At 60 seconds, the signal cycle length promotes low
delays and short queue lengths, though it might need to be increased
somewhat to accommodate pedestrian crossing time requirements.
It is important to consider each of these three levels of
analysis (intersection as a whole, approach, and lane group). Each
level tells the analyst something different, and important, about the
operation of the intersection. Intersection level of service is
important when comparing a number of design alternatives or when comparing a
number of different intersections in a study area. Approach and lane
group level of service is important when identifying operational problems
with specific traffic movements or signal timing allocation among phases. In
problem 4, we will consider various signal operations strategies and
their implications on the measures of effectiveness (MOEs) at the
intersection.
In addition to assessing the performance of the intersection
with existing traffic conditions as we did in sub-problems 1a and 1b, we also
need to look to the future to determine what conditions will be like at some
future time. Any significant transportation investment, such as the installation of a
new traffic signal, must be considered over time, not just with the existing
conditions. For that reason, we will now analyze the performance of the
intersection of U.S. 95/Styner-Lauder, under both control conditions, using future
traffic volumes.
Here are some issues to consider as you proceed with the
analysis of both TWSC and signal control, assuming future traffic volumes.
What is the appropriate future year for this analysis?
Which default values should be used for this future
analysis?
What other factors should be considered?
Discussion: Take
a few minutes to consider these questions. Click on continue when you are ready
to proceed.
What is the appropriate future year for this analysis? The appropriate horizon for a future year analysis depends on a number
of factors. Often, an agency has a standard target year (based on a
regional travel demand model) on the order of five,
ten, or twenty years. Sometimes the horizon is selected based on the
level and type of the investment under consideration. For a major
freeway investment, twenty years is often used. For a new signal
installation, ten years is often used. We will use a ten year horizon
for this analysis.
Which default values should be used for this future
analysis? While we can take field measurements to determine the
appropriate input data and default values to use for the analysis of
existing conditions, this is not possible for a future analysis. What we
must do is to use the existing data as a guideline and then, based on what we know
about the projected changes in traffic conditions and in the transportation
system itself, make estimates of these future values.
For the analysis of future conditions at the intersection, we must
also consider how to establish reasonable values
for other critical analysis parameters. For an unsignalized intersection
analysis, the additional critical parameters include peak hour factor, critical gap, follow
up time, and heavy vehicle percentage. For a signalized intersection
analysis, these critical parameters include arrival
type, saturation flow rate, peak hour factor, and heavy vehicle percentage.
For the analysis of future conditions, it is often appropriate to
select a peak hour factor of 1.0, considering that the projected future
volumes, as well as other critical analysis parameters, have a relatively
high degree of uncertainty associated with them. In this context, then,
using a peak hour factor of 1.0 is equivalent to evaluating the intersection
on the basis of the estimated average conditions for the hour rather than
the peak 15 minutes. In many cases, the accuracy resulting from this
approach is on par with the cumulative accuracy that one can expect for the
other critical analysis parameters that must also be estimated.
The value of the arrival type for the signalized
intersection option depends on whether the signal will be interconnected
with the adjacent intersections and the quality of progression that will be
achieved. It is usually conservative to assume random arrivals (Arrival Type
3). The values of the saturation flow rate, the critical gap, and the follow
up time depend on local conditions and whether, as volumes increase, driver
behavior may change over time. We will assume for this analysis that they
remain the same as the original analysis. These are all reasonable
assumptions, but most likely the assumptions are less accurate than the
estimates we were able to prepare for the existing conditions analysis.
What other factors should be considered? In
addition to the factors discussed above, we must also have forecasts of
future volume levels. In order that they might be as accurate as
possible, future volume projections should not just consider historic
patterns of growth but also current local policies regarding development. For example, if growth management policies are in effect, future growth may
be lower than historical patterns. And, since we can't know precisely the
composition of the traffic stream with respect to vehicle types, it is again
usually conservative to assume a passenger car equivalence of 1.1.
For this case study, we are unaware of any land use
policies that might change historic growth trends. Since we know that traffic volumes on U.S. 95 are
increasing at the rate of 2 percent per year, it is conservative to assume
that this rate will continue and that, with compounding, the volumes in ten
years will be about 22 percent higher than today's volumes. Click here to see the future traffic
volumes.
Let's continue to see the results of the computation.
The results of the HCM analysis for the two types of
intersection control under future traffic flow conditions for the intersection of
U.S.
95/Styner-Lauder Avenue are shown in the following tables.
Exhibit 1-12. U.S. 95/Styner-Lauder Avenue
(Dataset3)
Delay, Queue Length, and Level of
Service - Projected Ten Year Volumes
(Unsignalized control)
Exhibit 1-13. U.S. 95/Styner-Lauder Avenue (Dataset4)
Lane Group
Capacity, Control Delay, and LOS Determination - Ten Year Volume Forecast
(Signal control)
Discussion: Take a few minutes to review
the results on this page and the previous page, and compare them with the result for the
existing traffic flows. What can we learn from these results? Click on
continue to proceed to the discussion on the next page.
The delays for the major street left turn movements
remain low, even with the increased flows expected in the future. Both movements remain at level of service A.
The movements on the minor street approach experience
high levels of delay. Consider the TH/RT lane on the westbound (Styner)
approach, for example. Today, the average delay is 27.6 seconds per
vehicle, or level of service D. But the delay is projected to more
than double to 80.9 seconds in ten years, or level of service F.
The v/c ratio for the minor movements are near or
exceeding 1.0. When the v/c ratio exceeds 0.8, it is possible to see short term breakdown of the operation at the intersection, which leads
to high delays and growing queues. And, the EB LT movement has a v/c
ratio that exceeds 1.0. This means that all of the demand for this
analysis time period will not be served, and that some will spill
over into the next period. We will consider how to deal with
oversaturated conditions in Problem 4 of this case study.
It is worth commenting here on several aspects of the
HCM model forecasts. First, while computer models will often report delay to the
nearest one-tenth of a second, this more precise than is
warranted by the accuracy of the model itself, especially when planning
year horizon volumes are used that are forecasts. It would probably
be more reasonable to round the delay estimates to the two most
significant digits. Second, the high
levels of delay forecasted by the model for the EB LT movement (movement
10) are probably not realistic. You should use delay estimates in
this range, when the v/c ratio exceeds one, with great care. While
they indicate that the delay will likely be high for these conditions, it
would not make sense to compare this delay estimate with another estimate
in the same high range and make solid conclusions regarding the relative
differences between such estimates.
The intersection as a whole operates at level of
service B, both with the existing volumes and the projected future
volumes. Average delays are low, in the range of 10 to 11 seconds
per vehicle.
The critical v/c ratio for the
intersection is 0.34, indicating the intersection has a significant amount
of excess capacity. The fact that the critical v/c ratio is so low is also
an indication that average control delays can be reduced even more by
using a shorter cycle length (providing, of course, that all minimum green
times are still maintained).
The eastbound and westbound approaches (formerly the
minor street approaches, Styner Avenue and Lauder Avenue) operate at level
of service C, both with the existing volumes and the projected future
volumes, with average control delay varying between 20 and 25 seconds per vehicle.
The northbound and southbound approaches (on U.S. 95)
both operate at level of service A, with average control delay varying between 6
and 7 seconds per vehicle.
The 95th-percentile
back-of-queue is significantly longer under signal control than under stop
sign control. This means that drivers trying to enter or exit nearby
driveways may experience difficulty at times when the queue blocks either
the driveway or their view of oncoming traffic. The predicted queue
lengths are not so long, however, that they are in danger of blocking any
upstream intersections.
No lane group operates at less than level of service C.
No lane group is projected to operate with a
volume/capacity ratio greater than 0.60.
Discussion: Let's consider one additional point before we look at all
of the results for problem 1. When we project into the future, we do
so with some degree of uncertainty. Here, the greatest uncertainty
lies in our forecast of future volumes. What if our forecast is off by
10 percent? By 25 percent? How will this affect our final
results? Click on continue to proceed with this discussion of uncertainty
analysis.
Uncertainty
analysis is often used to consider the effects that uncertainty in the input
values for a problem will have on the outputs, and whether these effects
will be significant. To illustrate this kind of analysis, we will consider
three cases. The first case is the base case,
where we assume that the volumes that we used are correct. The second case
assumes that our volume forecasts are high by 25 percent, while
the third case assumes that the volume forecasts are low by 25 percent.
Exhibit 1-14 focuses exclusively on a single performance measure (average
control delay), though in reality we would also want to explore the impact
on other performance measures such as v/c ratio and queue length.
Exhibit
1-14. Comparison of Average Control delays
(sec/veh) (Datasets)
Movement
TWSC
Signal
control
Base case
Volumes +25%
Volumes
-25%
Base case
Volumes
+25%
Volumes
-25%
NB LT
9.1
9.8
8.5
5.9
6.6
5.6
NB TH/RT
6.3
6.7
6.0
SB LT
8.6
9.1
8.1
6.2
6.9
5.8
SB TH/RT
6.9
7.6
6.4
EB LT
297.5
--
34.1
20.4
24.8
18.6
EB TH/RT
55.7
305.2
20.9
22.0
24.2
20.3
WB LT
134.4
--
27.6
19.6
21.8
18.6
WB TH/RT
80.9
443.9
22.0
25.3
30.6
22.1
Exhibit 1-14 shows two very interesting points. First,
for the minor movements at the TWSC intersection, the volume changes have a
large impact on the final results. This shows that the projected
operation of the intersection is unstable to begin with, and that any
increase in the volumes will further degrade the intersection performance.
Second, we can have some degree of confidence that, under signal control,
the intersection will perform as planned, even if we are somewhat uncertain
about the input volumes that we've projected. These results can help
decision-makers feel more certain about the range of possible ramifications
associated with different control decisions.
In addition to varying the input volumes, we could also
determine the sensitivity of other input or default values on the final results. For example, the critical gap, the follow up time, the saturation flow rate, and
the arrival type are all important parameters in the computation of capacity
and delay. The base case values for these parameters could also be varied
to determine their effect on the final results.
Exhibit 1-15
and Exhibit 1-16 below shows the average control delay per vehicle for the minor street approaches (Lauder
Avenue and Styner Avenue) for both the existing and projected traffic volumes
under both TWSC and signal control.
Preparing graphic representations of delay often helps you to better visualize
the relative performance of the various movements at an intersection. Another
view of this data is shown on the next page.
TWSC
Intersection - Existing and Future Delay Estimates
Page Break
U.S.
95/Styner Avenue/Lauder Avenue
Signalized
Intersection - Existing and Future Delay Estimates
Page Break
ID# C101C09
Sub-problem 1c: Analysis of
Future
Conditions
Exhibit 1-17 shows a plot of the delay and level of service for each
minor stream movement for both the existing conditions and projected ten year
conditions, under TWSC. The existing conditions are shown in blue on the left, while
the projected conditions are shown in red on the right.
Each of the movements on the minor streets (Styner and Lauder) degrade to
LOS F in ten years. This performance is below standards established by both the
City and the Idaho Transportation Department, and clearly show that
consideration should be given to changes in either the intersection geometry or
control.
How can we integrate the
information from the analyses of both intersection control options in a
way that is useful to a decision maker?
TWSC
Intersection - Existing and Future Conditions
Page Break
ID# C1010A1
Problem 1: Analysis
of the U.S. 95/Styner-Lauder
Avenue Intersection
We will now integrate the results from
the three computations, considering the intersection
U.S. 95/Styner/Lauder under TWSC and signal control.
Let's begin with
a comparison of the overall delay experienced by all drivers who would use the
intersection. This intersection delay is calculated explicitly as part of the HCM
method for signalized intersections. For the TWSC intersection, we will
assume that the delay is zero for major street through and right turn vehicles
and compute a weighted average of delay for all vehicles entering the
intersection. But while we can compute the average intersection delay for
a TWSC intersection, we should note that
the HCM explicitly does not define
level of service for the intersection as a whole. This fact should
encourage the user to look carefully at the operation of each minor
movement. We should also be reminded that both v/c ratio and queue
length should be considered when reviewing the overall performance of the TWSC intersection.
Exhibit 1-18. Comparison of Intersection
Control Delay Estimates: TWSC vs. Signal
Control
Intersection
Control Delay (sec/veh)
Existing
Volumes
Future volumes
TWSC intersection
9.7
31.8
Signalized intersection
10.9
11.7
When a signal is added, the delay is shifted from some movements to other
movements. In this case, the northbound and southbound traffic on U.S. 95
experience no delay when the side streets (the eastbound and westbound
movements) are stop sign controlled. But when a signal is added, these
movements will also experience some delay. In fact, the average delay for all
vehicles increases by a small amount when the intersection control is changed
from TWSC to signal control.
The benefits of the signalized control, however, are shown when the future
conditions are considered. The additional delay for all vehicles increases
by a small amount, while for TWSC the delay increases significantly from about
10
seconds per vehicle to about 32 seconds per vehicle.
This difference is more dramatic when considering individual movements or
lane groups. For example, for existing volumes, the WB LT movement would
experience almost a 50 percent decrease in delay (from 36 seconds to about 19 seconds) if
the intersection were signalized.
with Analysis
Page Break
ID# C1010A2
Problem 1: Analysis of the U.S. 95/Styner-Lauder Avenue
Intersection
Exhibit 1-19. Summary comparison of average control delay estimates (sec/veh)
Movement
Existing volumes
Future volumes
TWSC
Signal control
TWSC
Signal control
EB LT
47.0
19.1
297.5
21.2
EB TH/RT
25.3
20.7
55.7
22.0
WB LT
36.0
18.9
134.4
19.7
WB TH/RT
27.6
22.9
80.9
25.3
NB LT
8.6
5.7
9.1
5.9
NB TH/RT
6.1
6.3
SB LT
8.3
5.9
8.6
6.2
SB TH/RT
6.6
6.9
A review of the information contained in Exhibit 1-19 provides important information to
think about as the decision to signalize the intersection (or not) is
considered:
Drivers on U.S. 95 will experience some delay if signal
control is added to the intersection, though the delay is minor (between 6
and 7 seconds per vehicle). However, more costly than this delay is the
fact that some portion of the drivers on these approaches will have to
stop, something that might be viewed as a degradation of service by many
drivers.
This reallocation of delay, however, significantly
improves the operations of the Styner (westbound) and Lauder (eastbound)
approaches. For example, drivers will experience a slight improvement in
their level of service (with average delay reduced from 27.6 seconds to
22.9 seconds) if the intersection is signalized with existing volumes.
Considering operations with future volumes, this change is much more
dramatic, with delay being reduced from 80.9 seconds to 25.3 seconds.
It is also important to point out that we should
consider that the off peak period constitutes 80 to 90 percent of the
total time that an intersection operates. During these times when volumes
are often significantly lower than in the peak period, a traffic
signal can sometimes be less efficient than stop-sign control. Or, a
signal may not even be necessary during these time periods! This is why
other forms of control (such as roundabouts) or flashing signals are
sometimes considered during off-peak periods.
Other options might also be available to avoid
signalizing the intersection. If, for example, the signalized
intersections on either side of the Styner/Lauder intersection are
coordinated and timed in such a way as to create regular gaps in the
traffic stream on U.S. 95, then the TWSC intersection analysis might yield
much more acceptable results when these effects are taken into account.
This may or may not be a practical option, but is something that should be
considered so that decision makers are presented with the fullest-possible
range of viable options.
The information presented here indicates that changing
from TWSC to signal control will improve the operation of this intersection,
particularly over the next ten years. The uncertainty analysis that we
conducted as part of
sub-problem 1c
confirms that this decision is a sound one.
Discussion: But have we considered all
relevant factors in this analysis? Take a few minutes to identify other
factors that you think should be considered in this analysis, and continue
to the next page when you are ready.
to Problem 1 Discussion
Page Break
ID# C1010D1
Problem 1: Discussion of the
U.S.
95/Styner-Lauder Avenue Intersection
The analysis presented thus far using the HCM methodologies for
Chapters 16
and 17, and the conclusions reached using these methodologies, all seem
relatively straightforward. The delay for the minor movements at the TWSC
intersection of U.S. 95/Styner-Lauder Avenue will continue to increase
during the next ten years. Installing a signal at the intersection, while
somewhat inconveniencing traffic on U.S. 95 that currently operates with no delay, will
improve traffic operations. But have we taken a thorough look at the
situation, or is there more to consider?
Yes, there is more to this story! We considered the
U.S. 95/Styner-Lauder Avenue intersection as isolated in our initial analysis,
but in reality, most intersections do not operate in isolation, and the U.S. 95/Styner-Lauder Avenue intersection is no exception.
Let's consider what kind of a system
this intersection is a part of. Review the sketch
and aerial photograph to remind yourself
of the geometric layout of the U.S. 95 corridor.
Sweet Avenue is located 1,070 feet to the north of Styner/Lauder. It is
a signalized intersection and serves as a major entry point to the
University of Idaho.
Palouse River Drive is located 2,410 feet to the south of Styner/Lauder. It is currently a TWSC intersection but the Idaho Transportation Department
is considering installing a signal at the intersection.
If we consider the U.S. 95/Styner-Lauder Avenue intersection to be a
part of this arterial system, there are several new issues that must be
considered in our analysis:
What are the arrival patterns at the U.S. 95/Styner-Lauder Avenue intersection of vehicles traveling southbound from
the Sweet Avenue intersection? If Styner/Lauder is TWSC, and if the
vehicles arrive in other than a random manner, the capacity that we
computed earlier might underestimate the actual capacity.
If Styner/Lauder is signalized,
the arrival rate and degree to which vehicles travel in platoons
from Sweet Avenue and arrive during the green phase or the red
phase has a direct impact on delay.
What quality of progression is possible if signals are installed at Styner/Lauder
and at Palouse River Drive? Will it be possible to provide good
vehicle progression if Styner/Lauder is signalized?
What effect will the signalization of Palouse River Drive have on the
operation of Styner/Lauder (effect of arrival patterns as discussed above)
and on the arterial as a whole?
Discussion: How
would you re-cast the problem that you first considered, based on the issues
listed above? Take a few minutes to summarize how you would revise the
original problem statement.