vV

PERATIVE HIGHWAY RESEARCH PROGRAM it' REPORT

1967 FREEWAY SURVEI LLANCE STUDY ­MARK TWAIN EXPRESSWA~ ST. LOUIS, MISSOURI J

MoDOT "TRANSPORTATION LIBRARY

MISSOURI STATE HIGHWAY DEPARTMENT BUREAU O F PUBLIC ROADS

So. • __ . _ -- ---

I I~

5 Q 1J . (I1Sill

fl v. 6 j·1J

1967 FREEWAY SURVEILLANCE STUDIES

MARK TWAIN EXPRESSWAY-SAINT LOUIS, MISSOURI

Prepared by

MISSOURI STATE HIGHWAY DEPARTMENT

Division of Highway Planning

November, 1968

in cooperation with

u.S. DEPARTMENT OF TRANSPORTATION FEDERAL HIGHWAY ADMINISTRATION

Bureau of Public Roads

The oplnlons , findings , and conclusions expressed in this publication are not necessarily those of the Bureau of Public Roads.

CONTENTS

Introduction .

Study Area .

Traffic Volumes

Densities

Summary

Figure

1

2

3

4

5

6

7

ILLUSTRATIONS

Mark Twain Expressway, St. Louis , Missouri.

Average October Weekday Traffic Volume .

Average October Peak Hour Traffic Volume .

Speed Vs. Density at selected Locations

Typical Urban Freeway Operation Under Various Traffic Density Levels .

Traffic Density Contours and Levels of Service During the Morning Peak Period on the Eastbound Lanes

Traffic Density Contours and Levels of Service During The Afternoon Peak Period on the Westbound Lanes .

Page

1

1

3

5

14

Page

2

4

6

8

9

10

11

Introduction

Since the movement of traffic is the purpose of freeways,

maintenance of satisfactory operating conditions is as impor­

tant as maintenance of the physical structure .

This report is based on the freeway congestion studies

conducted in the summer of 1967 which were part of a continuing

surveillance of freeway operations in Missouri. This is the

second report on studies of this type on the Mark Twain Express­

way. The first study was conducted during the summer of 1964

and the report published in October 1964. Should the need

develop , these studies should provide much of the information

necessary for the planning of control systems or measures designed

to improve traffic flow by pointing out areas where operations

have reached a critical level.

The primary purpose of this report is to provide an inven­

tory of present operating conditions on the Mark Twain Expressway

in St . Louis rather than a detailed analysis of operations at any

location.

Study Area

The name Mark Twain Expressway applies to the portion of

Interstate Route 70 in the St. Louis Area. The area under study

begins at Carson Road and runs to Shreve Avenue, a distance of

5.29 miles. This is the same area covered in the study made in

1964.

The geometrics of the s tudy section are shown in Figure 1.

The sketch at the top of the page has been distorted slightly so

1

f-

\ f- Z ~ Z

=:) Sw IO <to l<t ~ZO

.-I=:) ~ (f)<t <t (f)0 WZ

'!20 00 <to: Z Q <t U.W ~O 0:0: g~

\ 0

=:)<t u Z~O ~~ .-I ~O =:) Zf-OO: 0 W 150: .-I W(f) Wee u. \.:) > (f) =:) -, ee a: W

~ Z

Z W <t ci ~ ~ .-I

~ I . \.:)0

> CD I ~ w W > 0: Z (f)CD W

0 \.:) 0: Z Z I

0

~ ~ ~ r ~ Z 3£ (f)

<t 0 =:)

0 W 0: ee

Z 0 (f) Q:: <t U

•

VERTICAL PROFILE

100

~ 600

~

~ III ~ III 500

400

NUMBER OF LANES

WESTBOUND

~---2 I 3 i 3--------~

( 2 LANES REVERSIBLE )------4

EASTBOUND

~---2 I 3 i 3 -------4

SPEED LIMITS

(MPH)

55 55--------------------------------------------~

FIGURE 1. MARK TWAIN EXPRESSWAY (1-70) ST. 1.00ns MTSSnJ l RT

that each interchange corresponds with its correct milepost on

the line diagrams. This is the same base sketch used on the

density charts in this study. The number of through lanes

varies from two to five lanes (including the two reversible

lanes) as indicated in Figure 1. Grades over 3 percent are

indicated on the profile sketch. Speed limits are shown as

they existed January 1, 19680

Traffic Volumes

The traffic counts used in this study were made during

September 1967. For this study a complete inventory of ramp

traffic was not taken. Counts made on selected ramps were

used to update existing count data.

Counts from the permanent traffic counter located east

of Lucas-Hunt Road indicate that the September average traffic

should be within 3 percent of the annual average traffic ; there ­

fore, no attempt was made to factor this traffic into an annual

average. All traffic figures, both daily and peak hour , are

average weekday traffic.

Through lane volumes were obtained by taking the volume

from the permanent counter and adding or subtracting ramp vol­

umes. This produces through lane volumes which should be with­

in the same accuracy range that counting between each interchange

would provide and has the advantage of providing balanced volumes .

Average September weekday volumes are shown in Figure 2 . As

expected, the lowest volumes occur on the west end with increases

along the way to a maximum volume just east of Lucas-Hunt Road

3

\ ~ \ r 0 ::IJ (J) (J)

1> Z ~

::IJ 0

('\

~

~

'-('T1

Z G'l 0 Z 0 z 0 G'l ~ (J) rn r (J) r

~ 0 1> ~ -t

(5 1> < z

49\90 Orn (Q

(o~ \ <I§.80

FI GURE 2. AVERAGE SE PT EMBER WEEK DAY TRAFFI C VOLUME.

AVERAGE SEPT. WEEKDAY VOLUME 1967

Q

> ...J (I)

~ w > a:: w > a::

tP

~ t o

v

~

-=:i :-- 49290

% ~ 't-

~ L..

44930

~ ~

,:i

48890

w > <l

~J'~ o~

~o

Q ~

I­~ :::> oX

«l

ci > ...J (I)

> : I (!)

I en (!) z ~

50/20

50240

w > <l

Interchange . At this location the two- way weekday volume appears

to exceed 100 , 000 vehicles . This is a sizable increase over the

counts taken in 1964 . A direct comparison cannot be made , however,

since the 1964 counts were daily rather than weekday .

The volumes shown in Figure 3 are for the time period during

which the peak hour most often occurred on the ramps . This pro­

vides approximate peak hour volumes with a balanced flow . The

actual peak volume at any specific location may be slightly

higher than shown .

The maximum volume during the A.M . peak hour (E . B. lanes)

occurs just east of Riverview Boulevard . The volume in either

direction generally decreases with movement we s tward along the

freeway .

The maximum P.M. peak hour volume (W . B. lanes) occurs just

west of Goodfellow Avenue. At this location the freeway is

three lanes , with an additional lane serving as an acceleration­

deceleration ramp .

During the P. M. peak hour, the west end of the reversible

lanes , operating in the westbound direction , has a relatively

low volume (less than one-half the A.M. eastbound peak hour) .

This is apparently due to the difficulty in merging with traffic

in the through lanes .

Densities

The use of traffic dens ity, expres s ed in terms of vehicle s

per lane per mile, has been e stabli s hed a s a means of measuring

over-all operating conditions on freeways. Density , for this

study, was determined by counting the number of vehicles on each

5

\ ." I \ 0 AI (fl (fl

t:> Z ~

AI 0

(">

~~

'-[T1

Z Gl Z 0 Z 0 Gl 0 (fl ."

[T1

(fl I

~ I

t:> 0 ~ ~

is t:> Z <

[T1

FIGURE 3 . AVERAGE SEPTF:\lBER PEAK HOUR TRAFFIC \ ·OLU\lE.

AVERAGE PEAK HOUR VOLUME SEPT., 1967

a > --l II)

~ W :; Q: W > Q:

E B AM 6:45 -7:45 W B PM 4 : 1 5 - 5 : 15

\ v ,:l o

4544

5696

0) fT1 Gl

Z

5064

5/36

~

w > q

3634 ~8

880.

1970:1

""3256

a Q:

i--2 :::> x

co

a > --i II)

>­q ~ X l?

X en l? z lC

5/24

5466

w > q

section of road as they appeared on aerial photographs. In order

to use density as a measure of operation, it is necessary to re­

late it to some recognizable quantity, such as speed, which was

used in this case.

Speed and volume measurements were made at various loca­

tions on the three freeways being studied. These measurements

were then used to compute spot density at these locations.

Density vs. speed charts were plotted, and a least squares fit

was made for each location as shown in Figure 4. Data gathered

from all the locations were used to develop an equation to re­

late speed and density (line No.7 on Figure 4). This equation

has a standard error of 4.4 m.p.h. Densities were calculated

which correspond to the speeds associated with Levels of Service,

as defined by the 1965 Highway Capacity Manual.

At each location where data were collected , the speed limit

was 55 m.p.h. Therefore , it was unlikely that many speeds above

55 m.p.h. would have been recorded, regardless of the density.

Because of this, the speed-density relationships above 55 m.p.h.

are considered unreliable and the Levels of Service above Level

C (50-55 m.p.h.) cannot be distinguished.

Figure 5 was prepared to give the reader an idea of how

the various densi ties look from the air . The observer will

note that the cars do not appear as closely spaced as one would

think . This is due to the fact that most people are used to ob­

serving traffic at ground level where the gap between cars tends

to disappear.

In Figures 6 and 7 the densities are presented as contours

7

J: a.. ~

I o w w a.. (/)

J;'Tr.rmJ;' A

80~------~------~-------r------.-------~------.-------.-------r------'------~

70 1--" =

60 ...... k~ ... ' ... I

50 10... ~ '<'?k::::

40 J't< ',""'<'\:"k:

30

20

10

o o

CD 1-70 S.L., EB. AM, W. OF BEG. OF REV. LNS . 7-26- 67

® 1-70 S.L., WB. PM, AT COLLEGE AVE . PED. O'PASS GRADE+50/ 0

7-31- 67

® US 40 SL,WB. PM, AT MC KINLEY DR . PED. O'PASS GRADE=O% 8-2-67

@ US 40 S.L., EB. AM, AT MC KINLEY DR. PED. O' PASS GRADE= 0%

8-3-67

® 1-70 K.C.,EB. PM, AT LISTER AVE . GRADE+O.4%

8-16-67

® 1-70 K.C., WB. AM, AT LISTER AVE . GRADE+ 0.4%

8-23-67

(i) AVERAGE ALL LOCATIONS ---- ----

10 20 30 40 50 60 70 DENSITY-VEHICLES/ LANE/MILE

~P~~n V~ n~N~TTV QDTI:'r''rvn , r\0ATTAl\10

" 80 90 100

33 VEH. ILN. 1M!., (LEVEL C) 47 VEH. ILN. 1M!., (LEVEL D)

59 VEH./LN . /M!., (LEVEL E) 75 VEH. ILN. /M! . , (LEVEL F)

FIGURE 5. TYPICAL URBAN FREEWAY OPERATING UNDER VARIOUS TRAFFIC DENSITY LEVELS .

~ ct I w :I ~

\ \

~ z ~ Q: ct U

..... z ~~ !Go ct Q:Q: 00 g ~8 lL ffiQ:

m

C:=J

..... z ::> IO

J,~ j ::> ...J

LEGEND LEVELS OF SERVICE

LEVEL C OR BETTER SPEED RANGE (MPH)

50 OR GREATER LEVEL D 40 TO 50 LEVEL E 30 TO 40

LESS THAN 30 VEHICLES/LANE/MILE -20

LEVEL F DENSITY CONTOURS

MIL

~~ ~ ...Jz wd ~~O liw ~~ 8~ ~i=;<5 Wm 8 ~

Z~Q: ~(J') ....,

Q:

LJ

FIGURE 6·. TRAFFIC DENSITY CONTOURS AND LEVELS OF SERVICE DURING THE MORNING PEAK PERIOD ON THE EASTBOUND LANES.

W

~ W ~ ~

(J') W

0 ~ ~ I .

W ~ <!)o

> -~ W

...J m

Vim Q:

:> <!)

I

W

~ Z (J')

Q:

Z ~ z ::> b3 m

D

O· .. .' .... ..

". . , ~ .

2 ~

I ~ 2 .....

\ \

~ z ~ 0:: <l U

..... Z

~Sf !!?o 0::0::

9 lJ...

..... Z ::>

<l

IO

00

l<l

~S VlO <lO::

ffio:: u ::>

CD

....J

~~ ~ l1iz We

VlZo lJ...w

~~ ~ Q <l

8~ WCD

z!cio g

~ z~o::

0::

WVl "'")

LEGEND LEVELS OF SERVICE SPEED RANGE (MPH)

LEVEL C OR BETTER 50 OR GREATER LEVEL D 40 TO 50 LEVEL E 30 TO 40 LEVEL F LESS THAN 30 DENSITY CONTOURS VEHICLES/LANE/MILE -20-

MILES

WEST BOUND

TRAFFC

o

o --30" 'j

FIGURE 7. TRAFFIC DENSITY CONTOURS AND LEVELS OF SERVICE DURING THE AFTERNOON PEAK PERIOD ON THE WESTBOUND LANES.

W ::> Z W ~ ~

Vl W

e ~ ~ I .

W ~ (!)O

>

....J

-~ W

CD

~CD 0:: I

;>

~ (!)

Vl

W

Z 0::

Z :;;c 0 ::> z

~

on a time-distance chart. To reflect the equivalent level of

service, the density contours have been color coded. With the

use of color, the density charts are virtually self-explanatory.

Red indicates the lowest level of service (level F) , while green

indicates the best le vel that could be determined (level C or

better) •

It should be remembered that these charts represent opera-

tion on only one day, and while no unusual incidents were detected

on that day, daily variations in operating conditions do occure

The eastbound lanes (Figure 6) carry the heavy traffic and

are the lanes of interest during the morning peak period . Con~

gestion , or impending congestion (levels E and F) , is prevalent W'~~..f

throughout most of the study area -e a s t of the reversible lanes .

These conditions a~pear to prevail during the 7 to 8 A. M. peak

hour. This is the same general area where these conditions were

observed during the 1964 study, except in 1964 conditions were

much less severe.

It is difficult to pinpoint critical locations where con-

gestion first appears. The congestion appears at nearly the

same time in the entire area between Florissant Road and the

reversible lanes. The highest densities were found between

Bermuda Road and Lucas-Hunt RoadQ The congestion found west of

this section could be a result of back-up from this section.

Operating conditions in the westbound lanes on the day rep­

resented in Figure 7 were rather extreme and difficult to analyze .

The extent of congestion is readily apparent. The area was con-

gested throughout the entire study period. This makes it impossible

1 ?

to determine the location or time when congesti on first occurs .

This area extends from the western terminus of the reversible

lanes to the crest of a 5 percent grade mi dway between Jennings

Sta tion Road and Lucas-Hunt Road. At times the congestion backs

up into the reversible lanes section as far as Kingshighway

Boulevard .

The congestion extends over the same area as found in the

1964 study , although in 1964 it was much less severe .

It is not apparent whether the congestion found on the 5

percent gra de west of Jennings Station Road originates there ,

or if it begins at the curve at Jennings Station Road and the

grade merely prevents the traffic from regaining normal operat­

ing speeds .

The western terminus of the reversible lanes , where five

lanes merge into three , is the location of c onsi d e rable conges­

tion . Vehicles in the reversible lanes form a single lane and

merge with the three through lanes . This is the apparent reason

for the high densities that occur periodically in the area . For

a brief period , this congestion in the merging area tied up

traffic to the extent that it wa s "metering" traffic and down­

stream traffic began to operate under free flow conditions

(indicated by the pool of green in the middle of the red) .

Operating conditions on the day the westbound lanes were

filmed were somewhat poorer than anticipated . The early appear­

ance of such heavy congestion may be an indication that some

incident , which caused a slowdown in traffic movement , may have

occurred earlier in the day_

13

Summary

Traffic volumes are continuing to increase on the Mark Twain

Expressway . October daily traffic now exceeds 100 , 000 vehicles.

Heavy densities and low levels of service are found through­

out most of the A. M. peak hour on the eastbound lanes between

Florissant Road and the beginning of the reversible lanes .

During the P.M. peak period , the westbound lanes are heavily

congested between the reversible lanes and a point midway between

Jennings Station Road and Lucas-Hunt Road. While not covering

as long an area, the P.M. congestion is much more severe than

that found during the morning period . Any detailed study of

critical areas would include most of this area .

14

111I11111 111111~lll~il~llli~I~11I111 111111I1 RD0016302