Merging Taper Lengths for Short Duration Lane Closures

By
LuAnn Theiss, P.E. (Corresponding Author)
Associate Research Engineer
Texas Transportation Institute
3135 TAMU
College Station, Texas 77843-3135
Phone: (979) 845-9949
Fax: (979) 845-6006
l-theiss@tamu.edu
Melisa D. Finley, P.E.
Associate Research Engineer
Texas Transportation Institute
3135 TAMU
College Station, Texas 77843-3135
Phone: (979) 845-7596
Fax: (979) 845-6006
m-finley@tamu.edu
and
Gerald L. Ullman, P.E., Ph.D.
Senior Research Engineer
Texas Transportation Institute
3135 TAMU
College Station, Texas 77843-3135
Phone: (979) 845-9908
Fax: (979) 845-6006
g-ullman@tamu.edu

Prepared for Transportation Research Board, Washington, DC

90th Annual Meeting, January 23-27, 2011
Submitted on November 15, 2010
Word Count: 249 (Abstract), 5981 (Text), 1000 (Figures), and 0 (Tables) = 7230 words

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21

Page 2

ABSTRACT
Merging taper lengths described in the Manual on Uniform Traffic Control Devices are assumed
to be applicable to roadways of all types. But driver expectations and traffic operations vary
greatly between the higher-speed freeway environment and lower-speed signalized urban streets.
This study investigated the operational impacts of reduced taper lengths on lower-speed urban
arterials.
The researchers found that drivers do react differently when merging taper lengths are
modified. This is based on the fact that both the merging taper and the work vehicle in the
closed lane can serve as visual cues to drivers to vacate the closed lane. For longer taper lengths,
the channelizing devices are the primary motivator of driver lane changing. However, occluded
vehicles are more likely to become trapped, creating mobility issues in the traffic stream. For
shorter taper lengths, drivers are reacting to both the merging taper and the work vehicle itself.
Although fewer vehicles become trapped near the merge point, the merge point is much closer to
the work vehicle. For the no taper conditions (i.e., mobile operations), where the work vehicle
was much larger than the trucks used during the merging taper observations, fewer drivers
remain in the closed lane at comparable locations.
Obviously, both motorist and worker safety must be considered when choosing the
appropriate merging taper length for shorter duration work activities on urban arterials. Future
research should investigate the worker safety implications of installing/removing various
merging taper lengths versus the time it takes to complete the work activity.

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

Page 3

INTRODUCTION
When the normal function of a roadway is altered for construction, maintenance, and utility
operations, temporary traffic control provides for the continuity of the movement of traffic. The
2009 Manual on Uniform Traffic Control Devices (MUTCD) (1) defines the minimum
temporary traffic control requirements on streets and highways. The MUTCD also contains
typical applications that depict common uses of temporary traffic control devices, since defining
details that would be adequate to cover all applications is not practical. The temporary traffic
control selected for each situation depends on many variables, including but not limited to the
type of roadway, type of work, duration of operation, and location of work with respect to road
users.
When work is required in the traveled way, lane closures are used to separate road users
from the work activity. A lane closure typically includes a transition area where drivers are
redirected out of their normal path with channelizing devices that form a merging taper. A
merging taper is important because it provides positive guidance for motorists as they merge out
of the closed lane. In addition, a merging taper should be long enough to allow merging drivers
to have adequate advance warning and sufficient length to adjust their speeds and merge into an
adjacent lane. However, urban arterials are typically characterized by relatively low speeds
(45 mph or less) and frequent intersections and driveways. Depending on the location of work,
merging taper lengths can easily extend into upstream intersections or interfere with driveways,
creating deployment issues. When typical applications of work zone traffic control cannot be
deployed as prescribed, engineering judgment is needed. The MUTCD recognizes this and
allows for adjustment to merging taper lengths when they are to be used in close proximity to
crossroads, curves, or other influencing factors. In addition, the MUTCD also suggests that
longer tapers are not necessarily better than shorter tapers (particularly in urban areas) because
extended tapers tend to encourage sluggish operation as well as encourage drivers to delay lane
changes unnecessarily.
In some cases it takes longer to setup and remove a full set of temporary traffic control
devices than to perform the actual work. In addition, it is believed that the risk to workers during
the temporary traffic control installation and removal may be as great, or even greater, than the
risk incurred to actually perform the work. Consequently, the MUTCD provides flexibility and
allows for agency judgment concerning the use of simplified control procedures for short
duration and mobile work activities. However, for short duration operations that close a travel
lane on a multilane road, a merging taper in accordance with MUTCD requirements must still be
used. The time necessary to install and remove a MUTCD merging taper may still viewed as
excessive by many who conduct very short duration work activities (maintenance crews, utility
crews, etc.). Unfortunately, relatively little data regarding driver reaction to merging tapers of
various lengths, particularly on lower speed arterials, are available in the literature.
Consequently, it is difficult to objectively and accurately assess the potential trade-offs
associated with shorter merging taper lengths that may be faster to install and remove, but which
may not provide the same level of guidance and control to motorists as the current taper lengths
provide. Therefore, as part of a recent Florida Department of Transportation (FDOT) sponsored
project (2), Texas Transportation Institute (TTI) researchers designed and conducted field studies
on urban arterials to evaluate the operational impacts of conducting the following:
short duration work activities with a standard merging taper,
short duration work activities with a 160 ft merging taper,

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

1
2
3

short duration work activities with a 100 ft merging taper, and

mobile work activities with no merging taper.

BACKGROUND

Merging Taper Lengths

6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44

Page 4

Until the late 1970s, the MUTCD (3) specified minimum desirable taper lengths based on only
one formula: L=WS, where W is the width of the closed lane in feet and S is the 85th percentile
speed in miles-per-hour. This formula applied only to relatively flat grades and straight
alignments, but was considered valid for all speeds. The necessity of making adjustments to the
taper length were noted, particularly for providing adequate sight distance and/or the close
proximity of interchange ramps, crossroads, etc. However, some transportation professionals felt
that the standard taper lengths for speeds less than 60 mph were excessively long.
In 1977, Graham and Sharp (4) proposed a revised taper length formula that yielded
shorter tapers at speeds less than 60 mph (L=WS2/60, where W is in ft and S is in mph).
Proponents of the revised formula felt that the ability to stop and/or change direction was
inversely proportional to the square of the velocity, and shorter taper lengths would interfere less
with driveways and intersections. Graham and Sharp conducted field studies to directly compare
traffic operations when standard and proposed taper lengths were used in the same work zones.
The data collected included speed, erratic maneuvers, traffic conflicts, and lane encroachments.
The field studies only considered long-term lane closure situations (i.e., no short duration study
sites were included). In addition, none of the work zone sites studied included the use of arrow
panels.
Graham and Sharp found that the use of the proposed taper lengths did not produce a
greater number of erratic maneuvers and slow-moving vehicle conflicts than with the standard
taper lengths. In addition, the proposed taper lengths did not result in a greater number of
passenger vehicle or truck encroachments on adjacent lanes. Thus, Graham and Sharp concluded
that the shorter proposed taper lengths were not more hazardous than those previously used.
However, they also concluded that taper lengths shorter than those studied may show an increase
in conflicts; thus, the new proposed taper lengths were probably the minimum that should be
considered. Based on these results, the proposed taper length formula was included in the 1978
MUTCD (5) for urban, residential, and other streets where the posted speed is 40 mph or less.
Since that time, two formulas have been used to determine the taper length in work zones
(Figure 1).
The MUTCD (1) also contains a table showing stopping sight distance (SSD) as a
function of speed. Although the MUTCD merging taper lengths were not developed based on
this criteria, as shown in Figure 1, merging taper lengths computed from the two formulas
currently used are generally equal to or greater than the stopping sight distances. Thus, if a
driver is unaware of the closure until striking the first channelizing device in the merging taper,
the taper length provides adequate stopping distance between the decelerating vehicle and the
work vehicle.
In the late 1980s, TTI performed research on short duration work operations on freeways
(6). In that study, a no-merging taper condition with an arrow panel in a rural/suburban freeway
travel lane was briefly tried at one site, but was quickly abandoned after observing severe
braking by some drivers to avoid striking the arrow panel. Certainly, driver expectancies

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

Page 5

regarding the need to brake and change lanes are much different on these types of freeway
sections than they are on urban arterial streets, which raises questions about the applicability of
the statement to these lower speed facilities.
Recently, FDOT sponsored driver simulation-based research (7) to examine the
feasibility of using reduced taper lengths to decrease worker exposure while performing work
within the travel way of a multilane facility with a median lane or outside lane closure. The
primary purpose of this study was to investigate whether reducing the standard taper length from
540 ft to 100 ft on roadways with a lane width of 12 ft and a posted speed limit of 45 mph
increases accident likelihood. Researchers also considered the affect of the presence or absence
of a visually occluding lead vehicle and additional traffic that trapped the driver at the beginning
of the taper. In general, those researchers interpreted their results to indicate that the reduced
taper length of 100 ft increased accident likelihood, and that this likelihood was even greater
when a lead vehicle occluded the work zone. However, several limitations in the study
methodology, protocol used, and discussion of results makes the conclusions drawn somewhat
suspect. Most important of these limitations is the lack of a work vehicle with high-intensity,
rotating, flashing, oscillating, or strobe lights operating in the closed lane downstream of the
merging taper (which is recommended by the MUTCD when omitting the advance signing and
channelizing devices) even though the lane closure consisted of only cones (i.e., no advance
signing or arrow panel).
800

Calculated Taper Length (ft)

700
600
500

L=WS

400
SSD
300
200
L=WS2
60

100
0
0

10

20

30

40

50

60

70

Posted Speed Limit (mph)

21
22

FIGURE 1 Comparison of MUTCD taper lengths and stopping sight distances.

23

Work Duration

24
25

Work duration is a major factor in determining the number and types of devices used in work
zones. According to the MUTCD (1) short duration operations include work that occupies a

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44

Page 6

location up to one hour and mobile operations include work that moves intermittently or
continuously. Past research (8,9) has shown that both disparity and overlap exist between the
definitions of short duration and mobile operations among transportation agencies, as well as
among the specific activities associated with each type of operation. For example, work
activities that take 15 minutes or less to complete and move from location to location throughout
the work period could be considered a short duration operation or a mobile operation that moves
intermittently down the road. Intermittently is not defined in the MUTCD, but it does indicate
that mobile operations often involve frequent short stops for activities such as litter cleanup,
pothole patching, and utility operations, and are similar in nature to short duration operations.
The MUTCD definitions are purposely vague in order to allow individual agencies to
further clarify distinctions between work durations, as deemed appropriate. In order to better
classify the type of work activity described in the previous paragraph, some public agencies have
decided to specify the amount of time (e.g., up to 15 minutes, no more than 15 minutes,
approximately 15 minutes) that a mobile operation can stop in their mobile operation definition
(10,11,12,13,14 ). This time period is based on the belief that a well-prepared, efficient crew can
install and remove a full set of traffic control devices for a lane closure in approximately
15 minutes using conventional methods. In essence, the selection of a 15-minute threshold is
implying that anytime the work activity is stopped for longer than the time it would take to
install and remove a merging taper and other appropriate traffic control devices, those devices
should be installed.
Obviously, independent of the exact definitions used for short duration and mobile
operations, these types of activities are inherently different from longer term stationary
operations. At longer term stationary work zones there is ample time to install and realize the
benefits from the full range of temporary traffic control devices (e.g., advance warning signs,
tapers, arrow panels, etc.). However, some maintenance and utility operations only take a few
minutes to complete and thus the time to install and remove temporary traffic control devices can
take much longer than the actual work activity itself. Even the MUTCD recognizes this issue
and indicates that workers face hazards during the installation and removal of traffic control
devices. In addition, there is evidence to suggest that the installation and removal of temporary
traffic control is one of the more dangerous times for highway workers (15,16). The MUTCD
also notes that since the work time is short, delays affecting motorists are significantly increased
when additional devices are installed and removed.
Considering these factors, the MUTCD allows for simplified control procedures for both
short duration and mobile work activities. A reduction in the number of temporary traffic control
devices may be offset by the use of appropriate enhanced colors or markings on the work
vehicles and more dominant devices, such as high-intensity rotating, flashing, oscillating, or
strobe lights on work vehicles. The appropriateness of such adjustments is ultimately based on
positive guidance considerations (17). Generally speaking, these larger and more visible devices
on a vehicle allow it to be seen farther upstream thereby providing some advance information to
drivers about a downstream blockage or lane closure information that normally would have
been provided through the upstream warning signs and arrow panel. However, the safety of
short duration and mobile operations should not be compromised by using fewer devices simply
because the operation will frequently change locations.

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

Page 7

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Summary

16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35

Field studies were conducted in Broward, Orange, and Hillsborough counties in Florida to
evaluate the operational impacts of shorter merging taper lengths, including a no-taper condition
(i.e., mobile operation). The data for this study were collected on several urban arterials under
the following conditions:
the speed limit was 40 or 45 mph;
the duration of the work operation was approximately 15 minutes or less;
the work vehicle had warning lights per MUTCD (1) and FDOT standards (18);
there were no advance warning signs and arrow panel;
there were no sight obstructions;
right lane closures;
one or two lanes remained open to traffic;
daytime lighting conditions existed with dry pavement; and
the volume and complexity of the roadway were considered.

In summary, some maintenance and utility operations only take a few minutes to complete and
thus the installation and removal of temporary traffic control devices may take much longer than
the actual work activity itself. Independent of the whether these types of operations are defined
as short duration or mobile work, simplified control procedures are desired as a way to minimize
overall worker and motorist risk. While simplified control procedures are currently allowed, the
time necessary to install and remove a MUTCD merging taper is still viewed as excessive by
many who conduct work activities that take 15 minutes or less to complete. The use of shorter
taper lengths would further reduce the time that workers are exposed to traffic during the
installation and removal of traffic control devices. However, previous merging taper length
research is limited, so questions still exist as to whether reduced taper lengths would be
acceptable for slower speed roadways.
METHODOLOGY

The following merging taper treatments, shown in Figure 2, were evaluated in the field studies:
100 ft merging taper length with 25 ft device spacing,
160 ft merging taper length with 40 ft device spacing,
MUTCD standard 540 ft merging taper length with 25 ft spacing, and
no-taper condition (i.e., mobile operation).

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31

(a) 100 ft

Page 8

(b) 160 ft

(c) 540 ft

(d) Mobile

FIGURE 2 Merging taper lengths evaluated.

Cone spacing for the 100 ft taper treatment was based on FDOT standards (18) which
require 25 ft spacing of cones in the taper on a facility with posted speeds of 30 to 45 mph. A
160 ft taper treatment, using the same number of cones placed at 40 ft, was also included, since
lane stripes are generally placed at 40 ft intervals on the pavement, this merging taper would be
simpler to install (i.e., field personnel could simply place cones according to the lane stripes).
The standard 540 ft taper length was based on MUTCD criteria. Cone spacing for the standard
taper treatment was also 25 ft. For all treatments, standard 36-inch reflectorized channelizing
cones were used. The work vehicle was an FDOT pickup truck, similar to the one shown in
Figure 2(a-c), and was used for the taper treatments because it represented the minimum size of
vehicle that would likely be used for short duration utility operations. In accordance with FDOT
standards (18), the advance warning signs, arrow panel, and buffer space were omitted for all of
the merging taper treatment observations.
No merging taper was used during the mobile operations. The work vehicle was a utility
company bucket truck, similar to the one shown in Figure 2(d). Researchers hypothesized that
the larger utility truck would likely be more visible to approaching motorists than the standard
FDOT pickup truck. Again, the advance warning signs, arrow panel, and buffer space were
omitted for all of the mobile operations.
A total of 86 operations were observed at 43 different locations. Not all treatments were
observed at all sites. Due to phasing of the data collection, all treatments that used a merging
taper were observed at the same sites, and the mobile operation data were collected in a separate
phase at different sites.
The researchers documented the site characteristics of each location. These
characteristics included: speed limit, number of lanes open, time of day, sight distance,
intersection spacing, surrounding land uses, and weather conditions. Speed profile data were
captured to assess the speed and deceleration rates of free-flowing vehicles in the closed lane.
Video cameras and manual tabulation were used to capture lane distribution and erratic

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

Page 9

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

maneuver data. Data collection locations varied by type of work (short duration or mobile) and
data type. Additional information is provided in the discussion below.
Primary measures of effectiveness (MOEs) selected for this research were lane
distribution, percent remaining in the closed lane, percent occluded, percent trapped, and vehicle
acceleration/deceleration rates. Lane distributions were based on the percent of traffic in each
lane at various points upstream of the lane closure and at the beginning of the taper, allowing the
researchers to determine how far upstream of the lane closure motorists are moving out of the
closed lane. These data include all vehicles in the study area, regardless of their point of entry to
or exit from the study area. The percent remaining in the closed lane is used to more closely
evaluate the behavior of vehicles in the closed lane, and was estimated as the amount of traffic in
the closed lane at various points upstream of the taper and work vehicle divided by the amount of
traffic in the closed lane at 750 ft upstream (or 540 ft for the mobile operations to be able to
compare the beginning of the taper data for the 540 ft merging taper treatment). It includes only
vehicles that entered the study area in the closed lane, perceived and reacted to the work activity,
and merged into the open lane. It does not include vehicles that entered from or exited to side
streets or driveways located within the study area. The percent occluded is based on the percent
of vehicles entering the study area in the closed lane within 4 seconds of the vehicle ahead of
them. The percent trapped is based on the amount of traffic in the closed lane within 250 ft of
the beginning of the taper (or within 250 ft of the work truck in the case of the mobile
operations) that decelerated to a stop, or almost stopped, waiting for a gap in the traffic stream in
the open lane divided by the amount of traffic in the closed lane at 750 ft upstream (or 540 ft for
the mobile operations). Vehicle acceleration/deceleration rates near the taper were also
calculated to quantify driver reactions as they approached the work activity; however, these data
are not discussed herein. For each MOE, the average value by treatment type within each
roadway category (speed limit and number of lanes remaining open) was computed and
compared.

29
30
31
32
33
34
35
36
37
38
39
40
41

The scope of this paper is limited to discussion of the results for the sites with 45 mph posted
speed limit and only one lane remaining open. Data for sites with lower speed limits and more
lanes open followed similar trends and are discussed in the project research report (2).

RESULTS

Effects on Lane Distribution

As drivers approached the work operation, they exited the closed right lane, creating a shift in the
lane distribution. Figure 3 shows the lane distribution as a function of distance to the work
vehicle for all four treatments. Generally, a higher percentage of vehicles remained in the closed
lane at the beginning of the merging taper with the 540 ft tapers (24 percent) than with the other
taper treatments (20 and 14 percent for 160 ft and 100 ft, respectively). Interestingly, the lowest
percentage of traffic (3 percent) was present when no merging taper was used.

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

Page 10

50%

Percent of All Traffic in the Closed Lane

45%
40%
35%
30%
25%
20%
15%
Mobile

10%

100 ft Taper
160 ft Taper
540 ft Taper

5%
0%
1250

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

1000

750

540
500

250

160 100

Distance Upstream from Work Vehicle (ft)

FIGURE 3 Driver response to treatments.

Overall, one sees two different driver response patterns in Figure 3. The graph for the
540 ft taper length is shifted significantly to the left (farther upstream) than for the other three
treatments. Obviously, the 540 ft merging taper encouraged drivers to exit the closed lane
farther upstream than the other treatments, but a larger portion of drivers were still in the closed
lane at the beginning of the taper. However, the slope of the graph for the 540 ft tapers is similar
to the other taper length treatments.
The lane distribution values for the other three treatments all begin at about the same
value far upstream of the work vehicle, but begin to diverge as vehicles get about 300 ft from the
work vehicle. These diverging graphs imply that both the merging taper and the work vehicle
together serve as a warning system, providing cues to approaching drivers about the need to exit
the closed lane. For the 160 ft and 100 ft tapers, the proximity of the start of the merging taper to
the work vehicle decreases, and so more drivers move out of the closed lane prior to reaching the
beginning of the merging taper because many are reacting to the realization that there is a work
vehicle blocking the closed lane. This hypothesis is further confirmed by examining driver
behavior in response to the large bucket truck used for the mobile operations. Even though no
taper is present to provide a visual cue as to where they must begin to vacate the closed lane, the
percentage of vehicles in the closed lane at distances 300 ft and closer to the work vehicle are
less than those for the 160 ft and 100 ft taper treatments that were installed using a smaller work
vehicle (i.e., pickup truck).

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46

Page 11

Effects on Merging Behavior

The beginning of the merging taper defines the point at which drivers must either begin to merge
or stop to wait for an acceptable gap in the traffic stream in the open lane. Researchers were also
interested in the behavior of motorists and traffic flow in the vicinity of the beginning of these
merging tapers. Certainly, some drivers make a deliberate decision to move as far forward in the
closed lane as possible prior to beginning to merge. However, other drivers are forced to stay in
the closed lane because a suitable gap in the open lane may not be available. In either case, a
large number of vehicles trapped in the closed lane at the merge point can lead to turbulent
traffic flow as vehicles attempt to merge into the open lane of traffic from a standstill. This was
considered to be an undesirable outcome by researchers.
The traffic and site characteristics documented at each of the 17 sites used in the study
had considerable variation. Some of the characteristics that may influence traffic operations in
the urban environment include traffic volume, the presence of signalized intersections, sight
distance, turning movements, the presence of bus stops, and the frequency of buses in the traffic
stream. The platooning effect that signalized intersections introduce into a traffic streams impact
a drivers ability to merge out of a closed lane. For example, at higher traffic volumes, a platoon
of vehicles released from an upstream intersection that is located a short distance from the
beginning of the work zone may have fewer and smaller gaps into which the closed lane traffic
can merge than one that is located further upstream. In addition, when a platoon is tightly
grouped, a drivers ability to see beyond the leading vehicle is reduced. As the platoon disperses
further downstream, merging gaps become larger and more abundant. Certainly, some drivers
will intentionally remain in the closed lane to move as far forward as possible before merging.
However, it is very possible that a considerable number of drivers in the closed lane were
unaware of the lane closure as they encountered the work zone treatments because they were
right behind another vehicle and so had the taper and work vehicle occluded from view. The
researchers evaluated this possibility by identifying those vehicles entering the study area that
had occluded views of the channelizing devices and work vehicle, and assessing how many of
those occluded vehicles became trapped in the closed lane at the beginning of the taper.
Occluded vehicles were those entering the study area in the closed lane within 4 seconds of the
vehicle ahead of them. Overall, almost half (48 percent) of the closed lane vehicles observed
during the field studies were occluded. Occluded vehicles are less like to be able to see the
traffic control system (which includes both the merging taper and the work vehicle). Site and
traffic characteristics, such as traffic volume and distance from upstream intersection, contribute
to higher percentages of occluded vehicles.
Because occluded vehicles may not have a clear view of the work zone, they are more
likely to become trapped than vehicles that are not occluded. Trapped vehicles were those
vehicles in the closed lane within 250 ft of the beginning of the taper that decelerated to a stop,
or almost stopped, waiting for a gap in the traffic stream in the open lane. Trapped vehicles
present some concern because they create speed differentials within the traffic stream that can
contribute to traffic flow turbulence. In addition, trapped vehicles may become more impatient
as they wait for a gap to move into the open lane, and could tend to select shorter gaps in which
to merge, creating other potential safety concerns.
Figure 4 shows the percent of closed lane traffic that entered the study area occluded
compared to the percent of vehicles that became trapped within 250 ft of the merging tapers (or
within 250 ft of the work vehicle during mobile operations). Overall, a higher percentage of

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

Page 12

occluded vehicles generally resulted in a higher percentage of vehicles becoming trapped. The
combination of a lack of advance warning signing (a key positive guidance component of work
zone traffic control systems) and a fairly high frequency of vehicle occlusion of the channelizing
devices and work vehicle together resulted in a substantial percentage of closed lane traffic
becoming trapped at the taper merge point. The general trends in Figure 4 show that the
percentage of closed lane traffic becoming trapped at the taper merge point decreased as the
merging taper decreased. Researchers believe this occurred because the platooning may have
prevented many motorists from seeing the channelizing devices forming the merging taper far
enough in advance to prevent becoming trapped near the transition area. With the shorter taper
lengths, drivers had more time to for the platoon to disperse, allowing drivers to view the
merging taper and work vehicle, and move out of the closed lane prior to becoming trapped in
the transition area.
Only one erratic maneuver was observed during the study. In this instance, no merging
taper was present when a driver left the open lane and used the closed lane to pass a slowermoving vehicle in the open lane, re-entering the open lane just upstream of the work vehicle.
Although it is possible that having a merging taper upstream of the work vehicle may have
discouraged this maneuver, researchers believe that is also possible that the driver may have
attempted the pass even when the cones were present.
70%

Percent Trapped Within 250 ft

60%
Mobile
100 ft Taper

50%

160 ft Tapers
540 ft Taper

40%

30%

20%

10%

0%
0%

20
21

10%

20%

30%

40%

50%

60%

70%

80%

Percent Occluded

FIGURE 4 Percent trapped compared to percent of traffic entering occluded.

22

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

Page 13

SUMMARY AND CONCLUSIONS

Overall, the results indicate that there are differences in how drivers react to merging tapers of
different lengths upstream of a work vehicle on urban arterials. These differences reflect the fact
that both the merging taper and the work vehicle in the closed lane can serve as visual cues to
drivers that they need to vacate the closed lane. For longer taper lengths, the channelizing
devices begin farther upstream of the work vehicle, and are the primary motivator of driver lane
changing (in fact, they physically require drivers to vacate the closed lane once they reach the
channelizing devices). For shorter taper lengths, drivers are reacting to both the merging taper
presence and the work vehicle itself. As a result, more drivers have vacated the lane by the time
they reach a shorter taper length than a longer one. Of course, the beginning of the merging
taper is much closer to the work vehicle. For the no taper conditions (i.e., mobile operations),
where the work vehicle was much larger than the trucks used during the merging taper
observations, fewer drivers re in the closed lane upstream of the work vehicle at comparable
locations.
Although longer tapers force drivers out of the closed lane earlier, occluded vehicles are
more likely to become trapped, creating mobility issues in the traffic stream. Conversely, shorter
tapers result in fewer vehicles in the closed lane at the merge point and a smaller percentage of
these vehicles becoming trapped near the merge point, but the merge point is much closer to the
work vehicle. Obviously, both motorist and worker safety must be considered when choosing
the appropriate merging taper length for shorter duration work activities on urban arterials.
Future research should investigate the worker safety implications of installing/removing various
merging taper lengths versus the time it takes to complete the work activity.

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

1
2
3
4
5
6
7
8
9
10
11

Page 14

ACKNOWLEDGEMENTS
The research team would like to thank the Project Director, Jim Mills of the Florida Department
of Transportation (FDOT). Others who provided support during this project include: FDOT
Orlando South Maintenance Office, FDOT Broward Maintenance Office, FDOT Tampa
Maintenance Office, Florida Power & Light Company, and Tampa Electric Company. Their
participation was critical to the success of this research project. The contents of this paper reflect
the views of the authors, who are responsible for the facts and the accuracy of the data presented
herein. The contents do not necessarily reflect the official view or policies of the State of
Florida.

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

Page 15

REFERENCES
(1)
(2)

(3)
(4)
(5)
(6)

(7)

(8)

(9)
(10)

(11)

(12)

(13)

Manual on Uniform Traffic Control Devices, 2009 Edition, FHWA, U.S. Department of
Transportation, December 2009. Available at http://mutcd.fhwa.dot.gov/. Accessed
November 15, 2010.
Theiss, L, M.D. Finley, and G.L. Ullman. Merging Taper Lengths for Short Duration
Lane Closure.,Report No. BDK51, Florida Department of Transportation, Tallahassee,
Florida, December 2009. http://www.dot.state.fl.us/researchcenter/Completed_Proj/Summary_RD/FDOT_BDK51_rpt.pdf Accessed November 15,
2010.
Manual on Uniform Traffic Control Devices, 1971 Edition, FHWA, U.S. Department of
Transportation, November 1970.
Graham, J.L, and M.C. Sharp. Effects of Taper Length on Traffic Operations in
Construction Zones. Report No. FHWA-RD-77-162. Federal Highway Administration,
Washington, D.C., December 1977.
Manual on Uniform Traffic Control Devices, 1978 Edition, FHWA, U.S. Department of
Transportation, December 1979.
Dudek. C.L., G.L. Ullman, K.N. Balke, and R.A. Krammes. Traffic Control for Short
Duration and Stop-and-Go Maintenance Operations on Four-Lane Divided Roadways.
Report No. FHWA/TX-87/377-2F. Texas Transportation Institute, College Station,
Texas, March 1988. http://tti.tamu.edu/documents/377-2F.pdf. Accessed November 15,
2010.
Duley, A.R., J.F. Morgan, G. Conway, J. Wang, J. Abich, and P.A. Hancock. A Human
Factors Examination of Driver Response to a Specific Work Zone Design (Design
Standard #613, Duration Note 2) and Key Moderating Factors. Final Technical Report.
Project No. BD 548-18. University of Central Florida, Orlando, Florida, June 2007.
Ullman, B.R., M.D. Finley, and N.D. Trout. Identification of Hazards Associated with
Mobile and Short Duration Work Zones. Report No. FHWA\TX-04/4174-1. Texas
Transportation Institute, College Station, Texas, September 2003.
http://tti.tamu.edu/documents/0-4174-1.pdf. Accessed November 15, 2010.
Research Results Digest 339: Improving the Safety of Mobile Lane Closures. National
Cooperative Highway Research Program, Transportation Research Board of the National
Academies, Washington, D.C., August 2009.
Book of Standards for Highways and Incidental Structures. Maryland State Highway
Administration, Baltimore, Maryland, 2003. Available at
http://apps.roads.maryland.gov/BusinessWithSHA/bizStdsSpecs/desManualStdPub/publi
cationsonline/ohd/bookstd/index.asp. Accessed on November 15, 2010.
Minnesota Manual on Uniform Traffic Control Devices, 2005 Edition with 2007-09
Revisions. Minnesota Department of Transportation, St. Paul, Minnesota, 2005.
Available at http://www.dot.state.mn.us/trafficeng/publ/mutcd/index.html. Accessed on
July 30, 2010.
Work Zone Safety Set-Up Guide. New Jersey Department of Transportation, July 2005.
Available at
http://www.state.nj.us/transportation/publicat/pdf/WorkZoneSafetySetupGuide.pdf.
Accessed on November 15, 2010.
Texas Manual on Uniform Traffic Control Devices, 2006 Version. Texas Department of
Transportation, Austin, Texas, 2006. Available at

TRB 2011 Annual Meeting

Paper revised from original submittal.

Theiss, Finley and Ullman

(14)

(15)

(16)

(17)
(18)

Page 16

http://www.txdot.gov/txdot_library/publications/tmutcd.htm. Accessed November 15,

2010.
Virginia Work Zone Protection Manual: Standards and Guidelines for Temporary Traffic
Control. Virginia Department of Transportation, Charlottesville, May 2005. Available at
http://www.virginiadot.org/business/resources/WAPM-2005-Revised10_05.pdf.
Accessed on November 15, 2010.
Bryden, J.E., L.B. Andrew, and J.S. Fortuniewicz. Intrusion Crashes on Highway
Construction Projects. In Transportation Research Record: Journal of the
Transportation Research Board, No. 1715, Transportation Research Board of the
National Academies, Washington, D.C., 2000, pp. 30-35.
Schrock, S.D., G.L. Ullman, A.S. Cothron, E.Kraus, and A.P. Voigt. An Analysis of
Fatal Work Zone Crashes in Texas. Report No. FHWA/TX-05/0-4028-1. Texas
Transportation Institute, College Station, Texas, October 2004.
http://tti.tamu.edu/documents/0-4028-1.pdf. Accessed November 15, 2010.
Lunenfeld, H., and G.J. Alexander. A Users Guide to Positive Guidance (3rd Edition).
FHWA, U.S. Department of Transportation, Washington, D.C., September 1990.
2008 Design Standards 600 Series, Traffic Control Through Work Zones. Florida
Department of Transportation, Tallahassee, Florida. Available at
http://www.dot.state.fl.us/rddesign/rd/rtds/08/2008Standards.shtm. Accessed on
November 15, 2010.

TRB 2011 Annual Meeting

Paper revised from original submittal.