5802
Development of a Highway Driving Cyc.le
^ for Fuel Economy Measurements
by
Ronald E. Kruse
and
C. Don Paul sell
March, 1974
Environmental Protection Agency
Office of Air Programs
Office of Mobile Source Air Pollution Control
Emission Control Technology Division
Procedures Development Branch
Ann Arbor, Michigan 48105
-------
Development of Highway Driving Cycle for
Fuel Economy Measurements
Introduction;
This report describes the program that was conducted to develop
a driving cycle that represents typical vehicle operation on all types
of highways.
Purpose:
The purpose of this program was to measure road speed versus time
profiles of vehicle operation on all types of highways and non-urban
roads and to reduce these profiles to characteristic parameters which
could be used to develop a composite driving cycle.
Objective:
The objective of this program was to produce a driving cycle which
could be used to measure vehicle fuel economy under typical highway
operation as simulated on a chassis dynamometer.
Background:
The EPA has for several years recognized that the light duty vehicle
emission certification procedure provides reliable, reproducible informa-
tion which can be utilized for calculation of vehicle fuel eccnonyl. The
certification test procedure incorporates a chassis dynamometer that
exercises the test vehicle to simulate the power required of the vehicle
during an urban drive in a major metropolitan area?. The carbon mass
emissions from these tests can be used to calculate the average urban
fuel economy; this calculation equally applies to all the vehicle types
tested during the certification process and permits the effect of
vehicle design parameters on urban fuel economy to be assessed. Publi-
cation of these urban fuel economy data for all classes of vehicles
provides the consumer with one piece of information he can include as
a criterion for determining the suitability of any given vehicle for
filling his needs. The fact that more than half of the total vehicle
miles accumulated are traveled in urban areas reflects the importance
of knowing urban fuel economy.
The average vehicle owner tends to ignore urban ("around town")
fuel economy because it is usually less than highway Fuel economy and
because high-./ay fuel economy is more conveniently measured. Thus, the
typical vehi:le owner has conditioned himself to expect fuel economy
data to refer to highway type operations md the publication of urban
fuel economy data does not provide the information reletive to his per-
sonal experience. Highway travel accounts for more than 405! of the
-------
-2-
total vehicle miles traveled making highv.-ay fuel economy a useful and
valid criterion for judging vehicle performance. An appropriate dyna-
mometer vehicle exercise which simulates typical highv:ay operation
could also be employed to measure highv.-ay fuel econcny data. Publication
of both equally valid fuel economy rates v;ould be useful information for
many individuals.
Highway Driving Characterization:
The Department of Transportation segregates road systems into
either of two categories on the basis of principal area characteristics.
The two categories are urban and rural (highway), which are differenti-
ated because of functional differences in land use road networks, and
travel characteristics3. DOT experience indicates that this differen-
tiation in characteristics occurs in places of 5.COO population. Rural
(highway) road nftworks are adequate if place populations are less than
5,000 and urban traffic networks are required if the place populations
exceed 5,000. In order to characterize road types within either cateqory
the Department of Transportation has developed a "Functional Classifica-
tion Concept" which classifies each highway, road, or street accordina
to the principal service that it renders. This system of classification
develops a hierarchy of route types. Lowest in the hierarchy are the
local roads and streets, where trips begin and end. These trip ends
are characterized by low speeds, unlimited access, and penetration of
neighborhoods. At the top of the hierarchy are the arterials designed
to acccraiodate high volumes of through traffic. Intermediate facilities
or collectors accommodate the necessary transition from local roads and
streets to arterials. Outside urban areas, the main road type classifi-
cations are:
A. Principal arterial system
a. Interstate
b. Other principal arterials
B. Minor arterial system
C. Collector
a. Major collectors
b. Minor collectors
D. Local system.
The development of rural systems classification starts at the top of
the hierarchy and works down. First the principal ana minor arterial
-------
-3-
systcms are developed on a statewide basis. Then the collector and
local classifications are developed on a more localized (county)
basis.
On the basis of the above classification scheme, the percent of
total highv/ay vehicle miles traveled has been calculated for each
road type:
TABLE 1
Percent of highway vehicle
Type of Highv/ay miles traveled
A. Principal arterials 39.5
B. Minor arterials 22.4
C. Collectors 23.9
D. Locals 14.2
T00~%
Highway operation represents between 40 and 50% of total vehicle
miles traveled, a value vvhich continually decreases as urbanization
increases. These percentages are the basis for constructing a compo-
site highway driving cycle to simulate all types of highway operation.
For this study, five routes incorporating each road type to be
traveled during the characterization were selected by EPA personnel.
Figure 1 is a map of the general area. Figure 2 illustrates a sample
route which was designed to cover a variety of road types for equipment
check out tests. On the first run of this route the data recording
equipment functioned properly, but the vehicle experienced a fuel sys-
tem failure. The test equipment was transferred to the stand-by
vehicle and the replacement vehicle and equipment were checked out
on the dynamometer. Since the equipment had functioned properly on
the sample route and everything functioned well when checked on the
dynamometer, the route shown on Figure 3 was run first. This is
primarily a type B (minor arterial)route with 612 type B roads, 28%
type A (major arterial) roads and 11% type C (collector) roads. The
second route, Figure 4, Is a type A route with 100% type A roads.
Figure 5 illustrates a type C route with .% type C roads, 22% type
D (local) roads, 17% type A roads and 17% type B roads. The fourth
data collection run was a rerun of the sample route, Figure 2. This
route consists of 47% type D roads, 43% type C roads and 10% of type
A roads. T'ie fifth route was run on a freeway in Ohio subject to 55
MPH speed limits, consists of 100% type A roads.
-------
-4-
During this data collection process, 460 feet of chart were used,
which at 4 inches of chart travel per minute represents about 23 hours
of data, collected over a total distance of about 1050 miles. During
all travel, an observer accompanied the driver to make notes about
the trip and to log pertinent data.
Vehicle Instrumentation:
The vehicle used to collect data in this program v/as a 1971 Ford
Ranchvagon with a 429 CID-4V engine, 3 speed automatic transmission,
and a 2.75 ratio rear axle. This vehicle had been previously instru-
mented for a study of vehicle operation and driving profiles. The
instrumentation included a manifold vacuum transducer, digital timer
(seconds), driveshaft torquemeter, and driveshaft speed pickup. The
signals from the driveshaft were scaled and recorded on a stripchart
moving at a rate o 4 inches per minute to produce the sane tine base
as the federal urban driving cycle. All of the instrumentation vas
calibrated and checked on a chassis dynamometer to verify true speed
and torque readings. The vehicle contained a static inverter power
supply to provide 120 volt, 60 cps electricity. This supply was used
on all calibrations and testing.
The true road speed was checked against the vehicle speedometer
to permit a quick calibration of the recorder on the road. A panel
meter which indicated driveshaft speed also facilitated a third check
on true speed and calibration stability. Calibration checks indicated
good stability throughout the entire program.
The torquemeter had a shunt resistor which was used to calibrate
the gain of the torquemeter. The torque readings were scaled to
measure from -200 to +800 foot-pounds. Torque readings were used to
assess the variation in throttle position for various velocity pro-
files. No problems were incurred with this measurement.
Data Verification and Analysis:
For ease of analysis, the 460 feet of recorder chart gathered
during this experiment were displayed on the walls of the office
hallway at the EPA Ann Arbor laboratory. The charts were properly
identified according to route number and were reviewed and verified
by the route observers. There was one observer on each drive and
three observers were used in the program. These observers reviewed
their own traces and verified comments. They identified route seg-
ments according to type of road, A through D, determined which seg-
ments represented urban (population above J.OOO) drl/ino and deleted
the urban segments. Data reduction consisted of tabulating route
speeds at 15 second (1 inch) intervals to determine the rcaxiirum,
minimum and average segment speeds. Total segment time, distance,
-------
-5-
nurcber of stops, number of major speed deviations per mile for each
segment were calculated. A speed deviation was defined as an excur-
sion greater than +_ 5 I-!PH from a line connecting end-point velocities
on six inch intervals (1.5 min) of the entire segment.
These data were compiled from all of the charts for each road
type and the average characteristics were determined for each road
type. These data are presented in Table 2.
TABLE 2
Average Hiohway Characteristics
Road Type
A
B
C
D
Average
Speed
MPH
57.16
49.42
45.80
39.78
Composite3 49.43*
*Composite
Speed
Stops/mile
0.0100
0.0575
0.1260
0.2360
0.08
1
Speed
Deviations/
mile
0.070
0.439
0.484
0.598
0.327
C.395/A + .224/B + .239/C + .142/0}
After these road type characteristics and the composite highway
trip characteristics had been determined, a driving cycle selection
committee was designated. This committee v/as composed of the three
observers and three other EPA staff engineers. The committee reviewed
the data, decided that a nominal 10 mile highway route would be opti-
mum for laboratory testing and agreed on a method for obtaining the
route. The committee split into three groups of 2 persons each, one
observer and one other engineer. Each group was to select and com-
bine the appropriate lengths and types of road segments to produce a
route with characteristics equivalent to the actual composite charac-
teristics. Each group traced the selected sections of the actual speed
versus time charts to come up with the composite route. After the
three candidate routes were prepared, the committee reconvened and eval-
uated the relative merits of each route. As might be expected, the
three routes v/ere quite comparable, with each having special features
which that group felt were particularly important. After a thorough
analysis and discussion, the committee constructed a composite route
which contai.ied the best features of all ซhree routes. Table 3 pre-
sents the average characteristics of the composite route. Figure 7
is a photoreduction of the driving chart arid presents a graphical
illustration of the speed-time trace as read from right to left, be-
cause of the direction of chart paper travel.
-------
TABLE 3
Characteristics of Composite Hi g hway Dr i v i ng Cyc 1 e
Segment
Length
(IN)
9.5
11.5
l/.O
12.5
50.5
Inches
Segment
D
C
A
B
Overall
Total
Average
Speed
[MPH]
41.157
43.841
56.096
48.421
48.595
MPH
Distance
Traveled
(Miles)
1.629
2.101
3.973
2.522
10,225
Miles
Elapsed
Time
(MIM)
2.375
2.675
4.250
3.125
12.625
Minutes
% Total
Miles
15.93
20.55
38.85
24.67
100.0 %
t
cr
END
START
FIGURE 7
Composite Highv.gy Driving Trace
-------
-7-
TAELE 4
Comparative Analysis of Cycle Characteristics
Average Speed " Miles Traveled
Road Type Goal Actual Diff. Goal flctual Diff.
A 57.16 56.10 -1.06 39.5 38.8
B 49.42 48.42 -1.00 22.4 24.7
C 45.80 43.84 -1.96 23.9 20.6
D 39.78 41.16 +1.38 14.2 15.9
Composite 49.43 48.59 -0.84 10G.O 100.0 0.00
Table 4 compares the final characteristics of Table 3 with the goals
shown in Table 2. It is readily apparent that the highway driving cycle
closely approximate' the real world conditions. All average speeds are
within +_ 2.0 MPH of the real vorld average and the percentages of the dis-
tance traveled in each segment are within +_ 4% of the DOT values.
During the construction of this cycle, the committee decided to use
actual on-road traces to represent each segment. This decision placed
two restrictions on the end points of the segments; the slopes and speeds
had to be continuous at the segment junctions. Furthermore the committee
thought the most realistic sequence of road segments would be DCAB. The
cycle would start from an idle, contain four soeed deviations (one
each in B and D, two in C) and end with a deceleration to a stop and
idle. For the convenience of the driver, who also controls the CVS sampl-
ing, a 2 second idle period was included at the beginning and the end of
the cycle. The on-road data indicated the average idle time was 0.063
minutes/mile for all road types traveled.
Obviously, a change in any of these criteria for one segment impacts
on the characteristics of the adjacent segments as well as the overall
composite cycle characteristics.
One general observation about the B and C segments should be made.
It was sometimes difficult to distinguish whether a road was strictly a
type B or type C. Since their characteristics are very similar, a rigid
distinction and duplication in the cycle was not considered critical.
The driving cycle shov/n in Figure 7 was constructed from all of these
criteria and is considered to be an accurate representation of all the
types of highwpy driving normally encountered.
The characteristics of this highway driving cycle were determined by
tabulating tha velocities at each .1 inch of chart \Miich represents 1.5
seconds.
This tabulation was converted to a digital table which listed the high-
way driving cycle velocities for each of the 758 one second intervals. The
trace was then scaled to the same chart paper u_ed for the federal urban cycle.
-------
References
1. A Report on Automotive Fuel Economy, U. S. Environmental Protection
Agency, Office of Air and Water Programs, Mobile Source Air Pollution
Control, October 1973.
2. Development of the Federal Urban Driving Schedule, Society of
Automotive Engineers 730553.
3. Part II of the 1972 National Highway Needs'Report, House Document
No. 92-266, Part II.
-------
.- !-"- '--.-.-' r^ -O "- i'-'--Z'i- '"'-'^V'f^ M i cJwcunT
O>' i ::>' -ป-'''":' V_-V;. I,
" ' ' /"> . '? - . . _'.
' l^i "cMt'ioiie a ;-J-t I ^ V * " ., I ^fy-~ / ^A'^ " ^T~^~ * '' ~; ' JA-- ^- |^"l t wjiw"-^ '*, 'Jj'.'.,'" .r ' 1 \' \ ' ' T
1 c,:,_-':- :'v,'^'T v^^^1-".^. V ^ซ"^ '' t'' ฉ i AKH ARDOR^ .|^/);'-;4 '{ -'! ~i~'"tf$''?' ;-^ i'1 r,^^--! p-
'"3, [ A'b:5^ \l'v S *-ซ' j'^"'-- ;1'.J?^ซ!t"CGe.ซU|kt :^.-..^, i f'':t^ViปilVnt(3BeRC,mBF-^ri.-^ ; FIGURE 1
-iu^ poLXli-.-J iw^1 ,'@J _! r^j B,^rls^.-6M?,^/>,,. JHI^' ;>/.;:^.;-A ; '. GENERAL AR
* V: ,;"""/ r. H^-!ฐ.l "-r^-lL-L^ ' ซ- r\'(:'"U^|^'i \-3 ^^o^k-^'' : TRAVELED
i. .. x , i. .' ' ,>, So ''. r ' ' ;s\ fT'i v.'F i **"L.Rocv' > < i
.,...; . ; ' H - -i l'*MtปrS1cW,J.-:,!B"*'ป" ; ) ^ rl //>A ;OAV^_ .._,-S--ca- V ,1 I 'I
( X"""," '';i ' ' ,.>, VJ'1. r y '''A fn<] irp L *ji"L ^ ' ^ \.".. v
'>..* f.J -i l'^;>r ic-,.,!-;- ฐ'^n ; ) ^ r1 x -% Jฐivii_-'-"i -r.r, .7'. ;. / L I lL~
... j i. , tt^:^^uLฃ^i -^- -,-' " .-f-fvu.--}-(-\ ic.ซ... y .^-, -rx-
"' -V j " -I-.- .,,o.lili,r-|-.-.^ -, (;,,,.,,1^ - . . -; -,.,. /w ^ / I S\ \ ) ซ>*";- \ ^V"
rS r'vrK0T '^^"^'5?i^'l^' -HJ- r'l-ijr"^^,, \ซ ^:
ฑ:.';':5_._.Lj_ >.ปซ..ซ M ^j W^/ * on-l!-l-ฎ HM'/'i !!- p^xVS \^f/i ""n x>
^}Ti f T^^^^S^?*^^ '!
, ,!_;-'.Hi !tl!!{ .|-71.LJ.!!!l-; ' ฐ .^V ! *.,J 1 L>S:'1" 'r'/V, A
3"'" V;'c-'''^l-'%J^^i~^^^^^ ' r^f r^Q^'^l^^^'^'^^f^rs^^^s^
'
!
1
*
FIGURE 1
GENERAL AREA
TRAVELED
*
i
i
8
L/< ' .r. :
< '.'.n
NJ7.CVJ; r+
'l\ (I'O XT -- ' f ,i | : r.jmo"ii v^:^ x<- -w ^s. ,,,,
=c,7^Ti.."''r^!' l^Q ^^'^"T"?ฎ'^ ^rar^^^t5] >x-.
s
l.ii,' r..ป
10
n
-------
.v> v? I i . ....1
"Wซ2&%Sv;. , ..,., ''">
FIGURE 2
SAMPLE RUN
"D" ROUTE
-------
FIGURE 3
"B" ROUTE
-------
FIGURE 4
"A" ROUTE
-------
.,/..^^'
.M'4 :'-'
-ov:."v-
'';oฃ^+
/'k;
jrซi.
vli IK
*f
FIGURE 5
C" ROUTE
-ซ iSS
^i-^^^y i5Safis^jBa^>iซ^M^i^
-------
~)
?_ ^
CVi-i ,
... uorn
" i .
J:~' j'
-^^^^---^-"-.-"O-^ -. /. -
. "" * '.' '
H' _r' /TT\ -* -
3r4-c^- -
Ann Arjsr J^ .^ ;"i" ; Wi"r-
FIGURE 6
"A-55 MPH"
ROUTE
------- |