MC 76-02
Technical Support Report for Regulatory Action
Motorcycle Usage
January 1976
Notice
Technical support reports for regulatory action do not necessarily
represent the final EPA decision on regulatory issues. They are in-
tended to present-a technical analysis of an issue and recommendations
resulting from the assumptions and constraints of that analysis. Agency
policy considerations or data received subsequent to the date of release
of this report may alter the recommendations reached. Readers are
cautioned to seek the latest analysis from EPA before using the in-
formation contained herein.
Standards Development and Support Branch
Emission Control Technology Division
Office of Mobile Source Air Pollution Control
Office of Air and Waste Management
U.S. Environmental Protection Agency
-------
Abstract
The 1974 Gallup Motorcycle Survey is analyzed to obtain average
urban usage data. Data obtained include average trip length, frequency,
and speed plus areas commuted through. A qualitative comparison to the
NPRM Motorcycle Driving Cycles is made. Areas where additional data
would be useful are identified.
Prepared by
Project Manager
Motorcycles
Approved - Branch Chie'f/ SDSB
/ J
Distribution:
ipproved - Division Di/ector
D. Alexander
,E. Brune
> J. DeKany
C. Gray
D. (iardin
K. Hellinan
W. Houtman
T. Huls
R. Jenkins
G. Kit trudge
ฃ. Rosenberg
R. S t ahtuan
E. Stork
G. Thompson
E. Vogt
M. Williams
-------
Introduction
This report presents an analysis of motorcycle usage data available
from the 1974 Gallup Motorcycle Survey. In addition to simply present-
ing these data, an attempt is made to compare motorcycle usage V7ith the
Federal Test Procedure (FTP) driving, cycle being proposed in the NPRM
for testing motorcycles. It is recognized that insufficient survey data
exist to fully quantify how motorcycles are used; the attempt in this
report is to study the available data, qualitatively compare the data to
the driving cycles, and .recommend areas in which additional data would
be useful.
o The average urban trip length for street legal motorcycles is
5.2 miles.
o The average trip frequency is 3.2 trips per day. (Motorcycle use
is seasonal; these data are representative of use in the month of
May.)
o The calculated average commuting trip speed is 29 miles per hour.
o Small displacement motorcycles are used less frequently and over
shorter distances at lower speeds.
o Over one half the large displacement motorcycles but slightly less
than 20 percent of the small displacement motorcycles use the
freeways for commuting.
o It is recommended that the light duty vehicle weighting factors
be used for calculating emissions from uncontrolled and 1978 level
controlled motorcycles. The effect of weighting factors. on cal-
culated emissions will become more important as emissions .from
motorcycles are controlled, and the determination of more accurate
weighting factors specific to motorcycles would be recommended.
o The usage data obtained frsffl fifee? Gallup Survey is insufficient
to determine if a diffefettfi driving cycls for motorc-vcles L.-\
needed. Because the usage data indicate higher average speeds and
shorter trip lengths for motorcycles, obtaining m.ore detailed usage
data to further quantify driving characteristics is recommended.
A subset of the Gallup National Motorcycle Survey data was used for
this analysis. The subset contained cintn which were the intersection of
the nulli-iv.'ip.g categories of responses and classifications:
-------
Street and Dual Purpose Motorcycles
Engine Displacement >_ 50 cc
. Registered
Working Odometer
Reside in a city with population of ^L 50,000
This subset is representative of street legal motorcycles whose owners
reside within cities with populations greater than or equal to 50,000.
This data set was chosen so that the data would be as representative of
urban motorcycle usage as possible.
The specific Gallup survey questions upon which the following data
are based are presented in the Appendix.
Average Trip Length
Tables I and II present a summary of trip characteristics for
commuting and non-commuting trips. The overall average trip length is
calculated by weighting the trip lengths by both the frequency of trips
and by the percent of the population which is making the specific type
of trip. The latter weighting is required because motorcycles are not
used for all types of trips (only 59 percent of owners use motorcycles
to commute, for example).
The average trip length, based on the data in Table I and II is
calculated to be:
Average urban trip length
ป - '
CC 21 50 5.2 miles
CC >^ 170 5.4 miles
50 ฃ CC < 170 4.0 miles
The trip, length shown above is representative of.urban use. For
the average trip length for "non-commuting trips greater than 10 miles
one way", (a survey question, see the Appendix), it was assumed the
first 10 miles of each trip was urban and therefore 10 miles was used
for that trip length. No actual average length for trips greater than
10 miles if known since, the survey question was "open ended"; only trip
frequency was asked.
Frequency of Trips
Motorcycles differ from light duty vehicles in chat their ur^a is
frequently seasonal or may be of a recreational nature only- Thus the
average number of trips per day, as determined from the survey data.
will vary depending on whether those data indicating ?:ero trips of a
specific type are included in the average. For example, ^9 percent of
the owners surveyed did not take any 2 to 10 mile non-commuting trips
in the last week. Because respondents were asked questions about usage
in the month just prior to the survey, these data are representative of
use in the month of May. The results may be different for other seasonal
periods.
-------
-3-
The average frequency of all types of trips combined, which in-
cludes in the calculation those owners not using their motorcycles for
a specific type of trip in the last week, (i.e., zero trip response),
was calculated by summing the number of trips for each category for each
survey response. The average number of trips per day (zero trip motor-
cycles included) is:
Average Trip Frequency in May
CC >^ 50 3.1 trips/day
CC _> 170 3.6 trips/day
50 ฃ CC < 170 1.7 trips/day
Weekly Mileage
As a check on the validity of the average trip frequency and length,
the calculated weekly mileage is compared to the estimated weekly mileage
(a survey question, see the Appendix). The estimated average weekly
mileage includes inactive motorcycles.
Comparison of Weekly Mileage, Miles
Calculated^ Estimate^ Estimate(2) (3)
Total Weekly Mileage On Road Weekly Mileage
CC ^ 50 -113 109 93
CC _> 170 136 128 115
50 ฃ CC < 170 48 45 33
(1) Miles per trip times trips per day times 7 days per week.
(2) A Survey Question
(3) Off road riding not included
The weekly mileage calculated from the average trip lengths and
frequencies compare closely with the survey's estimated total weekly
mileage. The calculated mileage is higher than the estimate of weekly
en rone! mileage, hovrevcr. - Consider inf; that th-? results -;": =.!. -vl^ted
rroiu estimates, i:he agreement is gouci and support.;', t. he ca leu.I CM, _;.!
values OL: trip arid length frequency.
Av_er;.i ^ Soced
The survey askea r.ov lon^ it. took to travel <:.: work or ^ci:oo'L.
Using these data plus the average commuting trip .length, an average
speed was calculated. The average speed for coinnuif i.ng cr-ips is:
Calculated Average. CoiiiUiut ing T'-'ip f
CC _> 50 29.2 mph
CC >_ 170 30. 2 mph
50 _< CC < 170 22.7 mph
-------
-4-
The large displacement motorcycles show a higher average speed,
probably due to traveling part of the trip at highway speeds (see next
section). Breaking the commuting trips into school and work trip
speeds, the large displacement work trip is the only type of commuting
trip with a significantly higher average speed.
Calculated Average Commuting Trip Speed
To Work To School
CC >_ 170 32.8 mph 23.1 mph
50 <_ CC < 170 22.8 mph 22.4 mph
No data on non-commuting trip speeds are available from the survey.
V
Areas Commuted Through
Commuters were asked what areas they commute through. The results
are presented in Table III. The data show about one half commute through
downtown and industrial areas, and about 80 percent commute through
suburban and residential areas. There is little difference among the
two engine displacement classes. Approximately 20 percent commute
through rural areas. Only three percent commute solely in rural areas,
which is a result of analyzing only the responses from city dwellers.
Fifty-five percent of the large displacement owners commute on the
freeways, while only 19 percent of the small motorcycles are used on the
freeways for commuting. The latter figure reflects the fact that in
many states the smaller displacement motorcycles are legally prohibited
from using the freeways.
Hot/Cold Weighting Factors
The data needed to determine the hot/cold start weighting factors,
which are used in the total emissions calculations, are not available
from the Gallup Survey. In the NPRM, the light duty vehicle value of
0.43 cold starts per start is used.
The importance of accurately knowing the weighting factors depends
on the ratio of cold transient to hot transient emissions. For simplicity,
consider the equation for -.-.'?iehter! mnss omissions jn vhich t'm- c 1. s': p. r> r.?.
Y + n, Y, + Y
ct h ht s
V _ - ^
I
v.n
XvThere
Y T_I = Total weighted mass emissions of n ?;ivfn pollutant.
Y = Mass emissions from the cold transient portion of
the driving cycle.
-------
-5-
Y, = Mass emissions from the hot transient portion of
the driving cycle.
Y = Mass emissions from the stabilized portion of the
s
driving cycle.
n = Total number of starts per day.
n = Number of cold starts per day.
c
n, = Number of hot starts per day.
h
Substituting 1 - n /n for ix/n and taking the derivative of Y with
. c t h t wm
respect to n /n gives;
C- L,
d Ywm Y - Y, ,.
. - 7 v = ct ht
The change in total weighted emissions due to a change in the weighting
factor n /n is equal to the difference between cold and hot transient
emissions. If the hot and cold transient emissions are equal, then the
weighting factors have no effect on the calculated weighted emissions.
A review of emission data collected at EPA using the NPRM test
procedure shows that the ratio of cold to hot start emissions is approxi-
mately one for most motorcycles tested. Quantitatively, 85 percent of
the ratios are in the range of 0..8 to 1.2, Within this range, a change
in cold weighting -factor of .3 (from .43 to .73, for example) results in
less than a five percemt change in calculated weighted emissions.
The level of transient emissions for uncontrolled motorcycles tends
to mask any differences in cold and hot transient emissions. Since the
proposed 1978 level of control results in only moderate decreases in .
emissions for most motorcycles, the ratio of cold to hot transient
emissions will most likely remain at apprbximately one. Thus it is
recommended that the light duty vehicle weighting factors be used for
calculating emissions from uncontrolled and 1978 level controlled motorcycles
As motorcycles are controlled more stringently, the differences in cold
and hot transient: emissions may become -sign if ion; t . and the de ter;-.i;..a!: ion
of weighting 'factors specific to motorcycles may be needed.-
Comparison with the FTP Driving Cycle
The result? ore? .'':"! fed in Che ^vevlous sect ii.:!1^ are b-?..C:?c.! or. surv^v
clata, and their accuracy depends or, the respondents ability to estimate
frequencies, distances, and times- In addition, the data are not complete;
for example the average trip velocity can be indirectly calculated only
for commuting trips. Wi.th these quali f icationp :in mind, a comparison of
the Gallup survey results to the FTP for each engine displacement category
is presented below. The motorcycle FTP driving cycle speed versus time
-------
-6-
for the large displacement motorcycles is identical to the light duty
vehicle driving cycle. The high speed portion of the transient test
(bags 1 and 3) has been scaled doxm by the factor 0.64 for the small
displacement motorcycles.
50 < CC. < 170 CC > 170
Parameter Gallup FTP Gallup FTP
Urban trip length,
miles 4.0 6.8 5.4 7.5
Trip average speed, m , ,
mph 22. 7k ' 17.8 30.2VJJ 19.7
(1) Commuting trips only.
The trip lengths obtained from the survey are shorter than represented
by the test procedure, and the average commuting speeds are higher. The
average speed would probably be reduced, especially for the large motor-
cycles, if data on non-commuting trips were available and were included.
The argument that motorcycles are constrained by traffic to the same
average speeds as light duty vehicles is still strong. The reduction of
the maximum test speed for small displacement motorcycles appears justified
based on their lower trip speed and the small percentage of reported
freeway usage (Table III).
The Gallup data presented here is not sufficient to decide if a
different driving cycle for motorcycles is needed. It does suggest,
however-V-- that trip-length for motorcycles may be less than LDVs. The
average^speed, being based only on commuting trips and'an indirect com-
putation is insufficient to claim motorcycles in general travel at
higher average speeds than LDVs. The importance of this parameter on
emissions suggests -that obtaining further data on average speed should
be pursued.
-------
Table I
Gallup Usage Data - Commuters
(Street Legal Motorcycles in Cities)
n i
'arameter
Jommute to work only
Percent that commute to work
Miles to work, 1 way
# Trips in last 20 work
days (2)
Jommute to school only
Percent that commute to
school
Miles to school, 1 way
// Trips in last 20 school
days (2)
Commute to school and work
Percent that commute to both
Miles to work, 1 way
# Trips in last 20 work
days (2) .,
Miles to school , 1 way "
# Trips in last 20 school
days (2)
Calculated data
Average speed to work, mph
Average speed to 'school, mph
All Commuters
Percent that commute.-
Co-.nrnur.inp trip miles. 1 way
if Trips in last 20 v-vrk /
school days (2)
Question
Number
16
17
19b
16
20
22b
16
17
19b
20
22b
-
16
--
~~
50 < CC <170
26
5.6
16.7
4
5.8
21.2
9
5.9
17.4
2.6
10.4
23
22
39
5. 3
16
CC _> 170
51
10.2
ป
13.8
3
6.4
19.5
12
8.7
13.7
7.0
12.9
33
23
66
9 . /'
14
CC >_ 50
45
9.6
14.0
3
6.2
20.0
11
8.2
14.4
6.3
12.5
32
23
59
8.7
1-4
(2) Average of those who took trips; zero trips not included
-------
8
Table II
Gallup Usage Data - Non-Conimuting Trips
(Street Legal Motorcycles in Cities)
Question
Parameter Number 50 < CC <170 CC > 170 CC > 50
n the last 7 clays
// of trips < 2 miles 1 way 26 6.8 8.2 7.9
// of trips 2-10 miles 1 way 27 5.9 6.5 6.4
// of trips > 10 miles 1 way 28 7.2 4.1 3.9
(1) Average of those making non-commuting trips; zero trips not included.
.'crcent making at least 1 trip in last 7 days
//of trips < 2 miles 1 way 26 42 56 53
// of trips. 2-10 miles 1 way 27 30 57 51
// of trips > 10 miles 1 way 28 20 47 41
-------
Table III
Gallup Usage Data - Areas Commuted Through
(Street Legal Motorcycles in Cities)
Parameter
Percent of sample that commute
Percent of Commuters that commute
through:
Downtown, industrial areas
Suburban, residential areas
Rural areas
Rural areas only
Percent that commute, on freeways
Question
Number
16
24
50 < CC<170
39
CC > 170 CC > 50
66
59
23
49
78
15
0
19
48
80
23
3
55
48
80
22
3
49
-------
Appendix
Questions from Gallup Survey
ASK ALL OWNERS:
15- Approximately what percentage of the miles on this cycle do you think vere for street
riding?
"
Do you ...
Confute to work? 1[ ]
Commute to school? 2[ ]
Ride for other
purposes? 3[ )
Kone 0[
Other (specify) x[ ]
5 ways do you use this
1 If
1 2[
ป*
1 3t
1 o[
1 x[
motorcycle
] .
]
)
]
]
on the street? (read list)
1[ )
2[ ]
' 31 1 '
0[ )
xf 1
f-ftST RECENT.
2nd
IF USLD TO "COMMUTE TO WORK" IN Q. 16, ASK QUESTIONS 17 - 19b:
17. HC.V r-any miles one way is this trip?
Miles __ Miles _ _
16. Jiov long doea it take one way?
3rd
Kiles
Minutes
Hours
Minutes
Hours
Minutes
Hours
19a. If the weather is good do you ride your motorcycle to work -on a regular basis?
Yco 1[ ] 1[ ) . '.. 1[ ]
No 2[ ) 2[ ]':..'. 2[ ]
39b. How many round trips to work did you make during the past twenty .working days?
Nuaber
Number '
Number
IF U5LO TO ."COMMUTE TO SCHOOL" IN Q. 16, ASK QUESTIONS 20 - 22b:
iO. ;iov nany Eiles one way is this trip?
Miles _ __ Miles _ !-!iles
n. Kov long does i-; tal-ie cr.e vay? .
Hours
Minutes
Hours
Minutes
Hours
Minutes
22ft. If the veather is good do you rids your nvotcrcycle tc school on a re^vilar basis?
No 2[ J .-". 2[ ] ' 2[ ]
22b. How many round trips to school did yciu make during the past twenty school days?
. : :. : - ' ^ . A "- '. '
Mucber ' ' ' ' "'; Number "*'"''' Number
-------
Appendix - Cent.
ASK ALL COMMUTERS EITHER TO WORK OR SCHOOL QUESTIONS 23 & 24:
23. I>o you commute either on freeways, expressways, or turnpikes vith this cycle7
Yes 1[ ) H ] 1M
Bo 2[ ] 2[ ) 2l ]
2*. Which of these areas do you commute through:
Downtown business or
Industrial areas? l[ ) l[ ) ( it ]
Suburban or resident-
ial areas? 2[ ] 2[ ] 2[ ]
Rurr.l areas? 3t ) 3[ ] 3[ )
$< ALL OWNERS:'
26. Other than for commuting ... in the past seven days how many trips did you make th&t
vcre within .tvo miles from your home? - . .
ff of Trips # of Trips ff of Trips ,'- '
27. Other than for commuting ... in the past seven days how many trips did you make that
vere within -tvo to teii'miles from your home?
if of Trips ' ff of Trips__ ff of Trips
23. Other than for commuting ... in the past seven days how many trips did you make that
vcre over ten miles froia your home?' :-:
ft of Trips # of Trips # of Trips
29- Coneidering now your motorcycle usage for all purposes to include commuting and other
purposes ...
Thinking now about your average weekly usage during the past four weeks,, how many miles
did you ride in an average week? .
.Miles . Miles . :
------- |