EPA-AA-TEB-EF-84-2
Fleet Characterization Data Used for MOBILES
                 August  1984
         Test and Evaluation Branch
    Emission Control Technology Division
         Office of  Air and Radiation
    U. S. Evironmental Protection Agency

-------
                            Table of Contents

1.0    INTRODUCTION

2.0    REGISTRATION DISTRIBUTIONS
2.1    Light-Duty Gasoline & Diesel Vehicles
2.2    Light-Duty Gasoline & Diesel Trucks
2.3    Heavy-Duty Gasoline Vehicles
2.4    Heavy-Duty Diesel Vehicles

3.0    MILEAGE ACCUMULATION RATES
3.1    Light-Duty Gasoline Vehicles
3.2    Light-Duty Diesel Vehicles
3.3    Light-Duty Gasoline & Diesel Trucks
3.4    Heavy-Duty Gasoline Vehicles
3.5    Heavy-Duty Diesel Vehicles

4.0    TOTAL VEHICLE COUNTS

References

Appendix  1.  MOBILES July 1 Registration Distributions
Appendix  2.  MOBILE3 Gasoline/Diesel Sales Fractions for LDVs and LDTs
Appendix  3.  Annual Average Mileage Accumulation Rates for LDVs
Appendix  4.  Annual Average Mileage Accumulation Rates for LDTs
Appendix  5.  Annual Average Mileage Accumulation Rates for HDGVs
Appendix  6.  Annual Average Mileage Accumulation Rates for HDDVs
Appendix  7.  Total HDDV Registrations by Class for 1980 through 2000

-------
1.0    INTRODUCTION
The fleet  characterization  data in MOBILES  are  used primarily to  weight
emission rates  for different age  groups of  vehicles  into fleet values.
The fleet characterization data consist of:

              0    registration distributions by  age
              0    average annual mileage accumulation  rates by age
              0    gasoline/diesel sales fractions
              0    total vehicle counts by class

The first  two pieces  of  information  are  used  to  obtain  the percent  of
fleet  vehicle-miles-traveled  (VMT) accumulated  by a  given vehicle  type
for a  particular age.  These  are  referred  to  as  the travel  fractions.
The average annual mileage  accumulations are also used with  the  emission
deterioration rates and their zero mile emission levels to  estimate  the
emission rates of  a given vehicle type  for  a given calendar  year.   Both
the registration distributions and mileage accumulation rates  are part  of
the  MOBILE2/MOBILES  input  parameters.   National  average   values   are
assigned within  the  computer  program,  or  they   may  be altered  by  user
supplied rates.

Because  there  have   not  been  significant  sales  of   light-duty  diesel
vehicles and  trucks until recent  years, these vehicles are combined  with
the   gasoline   vehicles   and    trucks   in   developing    registration
distributions.   The  resulting  distributions  are  for  all  light-duty
vehicles and  light-duty trucks, irrespective  of whether they  are  gasoline
or diesel powered.  However, as gasoline powered  vehicles and trucks  have
different  emission rates than  the diesel  powered  vehicles  and trucks,
these  combined registration  distributions must  be split  into  separate
gasoline and  diesel distributions.   This  task  is  accomplished by using
the gasoline/diesel sales fractions.

The vehicle counts by class are used along  with other information,  such
as average annual mileage traveled by vehicles  in each vehicle class  and
the  percentage  of  vehicles  which  are  diesel  powered   to  yield  • the
percentage of total mobile  source  VMT accumulated by a particular vehicle
class.  These percentages are used in turn  to  determine  an  "all  mobile
sources combined" emission factor.

One basic assumption of the emission  estimation model  regarding the fleet
characterization  data  is  that  all   vehicles  are  sold,   scrapped,   and
accumulate  mileages  uniformly  across  all calendar  years.    Therefore,
it is  important  that  both  the  registration  distributions  and the  annual
    The one  exception is  the  mileage accumulation  rates for  heavy-duty
    diesel vehicles.  This is explained in section 3.5.

-------
mileage accumulation  rates reflect the  general trends  of  the  vehicles'
age distributions  and travel patterns.   EPA contracted  with Energy and
Environmental  Analysis,   Inc.   (EEA)   to   determine  how   the   fleet
characterization data may have changed since MOBILE2.  Their  final  report
(Ref.  1,  hereafter  referred  to  as   the  1983  EEA  report)   contains
estimations of registration distributions and mileage accumulation rates
based  upon  more   recently  available   data.   Most   of   the   fleet
characterization data used for MOBILES are,  at  least in part, based  upon
the updated estimates.

The  purpose  of   this  report  is  to  summarize  all  the  July   1   fleet
characterization  data used  for  MOBILES.   The  MOBILES  computer  program
uses  the   same   methodologies   as MOBILE2  to  transform   the  July   1
information to January 1' data.  This report will limit its  discussions to
the July 1  fleet  characterization data only.  Detailed discussion  on how
the  fleet  characterization  data are  used  by  the  MOBILE2  (or  MOBILES)
computer  program  are  contained  in  the  document  "Compilation  of Air
Pollutant Emission Factors:  Highway  Mobile Sources"  (Ref.  2)2.   A new
compilation document  on MOBILES  is scheduled to  be  released in October,
1984.
    Discussions on the calculation of the VMT mix are in Appendix  B.   The
    January 1  transformations  and travel weighting  fraction  calculations
    are in Appendix D.

-------
2.0
REGISTRATION DISTRIBUTIONS
In the mobile  source  emission models, there  are  eight vehicle types  for
which  the registration  fractions  by  age  are required.   These  vehicle
types are shown in Table 1.

                                 Table 1

                   Vehicle Classes  in MOBILE2/MOBILE3
Vehicle Type                           Abbr.

Light-Duty Gasoline Vehicles           LDGV
Light-Duty Diesel Vehicles             LDDV

Light-Duty Gasoline Class 1 Trucks     LDGT1
Light-Duty Gasoline Class 2A Trucks    LDGT2
Light-Duty Diesel Trucks               LDDT

Heavy-Duty Gasoline Vehicles           HDGV
Heavy-Duty Diesel Vehicles             HDDV

Motorcycles                            MC
                                                  Gross Vehicle
                                                  Weight (in Ibs)
                                                  0-6,000
                                                  6,001-8,500
                                                  0-8,500

                                                  more than 8,500
                                                  more than 8,500
Although  it  is  possible  to  develop  registration  distributions  for  the
LDDVs  and LDDTs from  historical data,  these  distributions are  probably
not representative  of  future  registration distributions since  there  were
very few  light-duty diesel sales before  the  year  of 1975.   Therefore, the
LDDV and  LDDT  distributions were  estimated  from the combined  light-duty
vehicle   and   truck    registration   distributions   and   the   expected
gasoline/diesel  sales  fractions.  For HDVs,  there were  enough sales  in
both the  gasoline and  diesel classes to  estimate separate  registration
distributions.   Therefore, in MOBILE2/MOBILES,  the  registration fractions
are given for five vehicle types:  LDVs,  LDTs,  HDGVs,  HDDVs, and MCs.

The MOBILE2  registration  distributions were  based upon vehicle counts  by
model  year  (age)  in  a given calendar  year,  that is,  the LDV  and  LDT
registrations were  from calendar year 1978  counts (Ref.  3)  and the  HDV
registrations were from calendar year 1972 counts.  This has  created  some
problems.  When  a  registration distribution is  based upon counts from a
single calendar year, it  incorporates the sales  trends  and scrappage  that
exist  at  that  time.   These  sales  and  scrappage  trends  may not  be
appropriate  for  all other calendar years.   For example,   since the sales
of  LDVs   and LDTs were  relatively high around  the year of  1978,   the
registration distributions developed from these data  weighted newer (and,
from an emissions standpoint, cleaner) vehicles more, and  older  (dirtier)
vehicles  somewhat  less.  In  fact, the  sales   of  LDGT2s   were  very  high
during the  1975-78 time  frame.   Therefore,  when  emission estimates  of

-------
LDGTls  and  LDGTZs  were  made  for  the  1985  and  later time  frames,  the
LDGT2s were projected  to have lower emissions  than the LDGTls because of
the   heavier   weighting   toward  the   newer  model   years   (and  lower
emissions).  To  eliminate or  reduce  these types  of  problems  in MOBILES,
registration distributions  were  based  upon two nonadjacent calendar years
of historical data.

Table  2  is  a  summary  of  the  data  base  references for   registration
distributions used  for MOBILE2 and MOBILES.  No  update has been done for
the motorcycle (MC) registration distributions.

                                 Table 2

                References  for Registration Distributions

Vehicle Class             MOBILES                       MOBILE2

LDVs                      1977 & 1981  R.L. Polk        1978 R.L. Polk
LDTs                      1978 & 1981  R.L. Polk        1978 R.L. Polk
HDVs                      1972 TIUS* & 1981 Polk        1972 TIUS
*   Truck  Inventory and  Use Survey, U.S.  Department  of Commerce,  Bureau
    of the Census.

The   method   used  to  develop   a  registration  distribution   from   two
representative  calendar years'  data was to derive a density  function  that
excluded  the  age 1  and  age 20 plus  fractions.   Age 1 data  were  excluded
because  the  surveys of vehicle counts  by age take place in  July, and  the
age  1 (current model  year) vehicles have  not  all been included  in these
counts.   Age   20   plus   data  were   excluded   because   registrations   for
vehicles  older than 19 years  of   age are added up  to  be  the age 20  plus
registration.   Therefore,  by  excluding  the  ages  1 and 20  plus data,  a
general  trend  of  vehicle  registrations can be better characterized.   With
the  density  function  (either  a  linear  or an  exponential)  that  fits  the
data  the best, a  set of  predicted  fractions  for ages 2  through 19  can
thus  be  derived.   It  should be  noted  that  the  registration of  vehicles
decreasing  as   the  age of  vehicles  increases.   For example,  the age  10
vehicle  fraction  is expected to  be  lower than the  fraction  of  registered
age   9  vehicles,  and  higher   than  the  fraction  of  registered  age  11
vehicles.

-------
The  age  1  fractions  were  then estimated  from the  age 2  fractions by
applying an adjustment  factor.   For LDVs and  LDTs the adjustment  factor
is 0.75,  since the sales of LDVs and LDTs are assumed to begin in October
of  the  previous  year  and  the  surveys of  vehicle counts  take  place in
July.  The adjustment factor for HDVs is 0.50, due to the fact that sales
of HDVs are assumed  to  begin in January and the  July 1 information will
account for half  year vehicle fractions.  The  age 20 plus fractions were
the  sum of  the  extrapolated values from ages 20  and up.  Therefore,  the
age 20 plus registration is expected to be a little higher than  the value
predicted for age 19.

The  set  of  smoothed  fractions was  then  renormalized.   The   resulting
MOBILES July  1  registration distributions  are tabulated in Appendix 1.
Note that the LDGV  distributions are used for  both  LDGVs  and LDDVs.  In
addition,  the  fractions  from  LDGTls  are used  for  all  LDTs,   explained
further in Section 2.2.

The  gasoline  and  diesel sales  fractions for  LDVs  and  LDTs  (given in
Appendix  2)  are  used   in  MOBILE2 /MOBILES  to  split  the   LDV  and  LDT
registration  distributions  into   gasoline   and   diesel    registration
distributions.  The  sales  fractions of diesel  and gasoline HDVs are  not
listed due to the fact  that separate registration  distributions  for HDGVs
and HDDVs are already available.

2.1    Light-Duty Gasoline and  Diesel Vehicles

The  1983  EEA final  report  tabulated  LDV  registration distributions  for
calendar years  1976 through  1982   (Table  3-1 of  the  1983  EEA report).
These fractions were generated  from R. L.  Polk data.   These  seven year
fractions  were   used  as  data  base  for  curve-fitting,  using  the  same
methodology described in  Section 2.0.   The  resulting set of  registration
distributions is  very similar  to that obtained from using  only the  1977
and  1981  data.    This  is due  to  the  fact  that  during this  seven  year
period,  1977  had  the  best  vehicle  sales (8,497,893  units),  while  1981
held the worst sales record on  light-duty vehicles  (6,253,138 units).  By
using only  the  1977 and  1981 data,  the registration distributions would
capture  the  fluctuations of  the  economic  situation.  This  is  the  data
base used for MOBILES.

A  linear density function  was  found to result in a better  fit than an
exponential curve.   The  predicted  fractions  from an  exponential  curve
exhibited unrealistically high  fractions  for ages  2  through 4  vehicles.
On the other hand, the  fitted linear function had  such  a rapid decreasing
rate that  the  predicted fractions reached  zero at the age  of  19.   This
would imply that the LDVs are no longer in use after  they are 18 years of
age — an assumption contrary to reality.  Therefore, for ages 17 and  up,
averaged values from the 1977 and  the  1981  data were used to replace  the
predicted fractions.

-------
The  renormalized  fractions  were  used  for  MOBILES  LDV   registration
distributions.   Figure  1  is  a  pictorial  view  of  the  four  sets   of
registration distributions:  one  that was used for  MOBILE2 (1978  data),
one  that  came  from the  1977 data, one  that was obtained  from the 1981
data,  and one  that  combined  the  1977  and  1981  data  sets  (used  for
MOBILE3).

2.2    Light-Duty Gasoline and  Diesel Trucks

MOBILE2 used  separate registration distributions for  light  duty class  1
and  class  2A  trucks.   However,  using separate registration  distributions
for  LDGTls and LDGT2s presented  a problem  in MOBILE2,  as discussed  in
section  2.0.   The emission  standards for LDGT2s have been  the same  as
those for LDGTls since 1979.  The sales of LDGT2s  have also  stabilized  so
that  the  1975-78  sales  of  LDGT2s  should  be  viewed  as  a  one-time
occurrence.  To  avoid the  problem of  heavier weighting on newer  model
years,  a  separate  LDGT2  registration  distribution  was   not  used   in
MOBILES.   Because of  the similarities in vehicle weight  classes  and uses
(personal  transportation  and   light  hauling),   the  LDGT1  registration
distribution was used for LDGT2s.

There have not been enough diesel truck  sales in past years  to  estimate
the  LDDT  registration distribution.   MOBILE2 used the  LDGT1  registration
distributions to  represent  both the  gasoline  class  1 and  diesel  trucks.
This was also done in MOBILES.

The  LDGT1  fractions  from the 1978 data  (used  in  MOBILE2) and  those from
the  1981  data are the data base used  for MOBILES.    The curve-smoothing
technique  discussed  in Section 2.0 was  used.   A linear  density  function
was  found to be  a  more  reasonable  fit  than  an  exponential curve.   The
exponential curve for this data set had  such a small  decreasing rate that
the  fraction  value  for  age 20  plus  was  calculated  to be higher than
the  predicted fractions  for any other age.  Therefore, a linear  function
was  used  to smooth  the LDGT1 registration distribution.   The  renormalized
registration  distribution,  together  with those   from  the 1978  (MOBILE2)
and  the 1981 data, are shown in Figure 2.
    As  described  in Section 2.0,  the  prediction on age  20 plus was  the
    sum  of  the  extrapolated  fractions  from age  20  and  up,  until  the
    fractions reached zero.

-------
2.3    Heavy-Duty Gasoline Vehicles

The most recently available  HDGV  registration data were found in  the  1981
Polk data  (Table  3-19 of the  1983  EEA report).  Fractions developed  from
this data  and the 1972  TIUS  data (used in MOBILE2)  are  used as  the  data
base for MOBILES.  An exponential function is used to smooth  the data.

The  renonnalized fractions  are  the  registration  distribution  used  for
MOBILES.   The three  sets  of fractions, one  from the  1972  data  (used  in
MOBILE2),  one from  the  1981  data,  and the one  that  is used for  MOBILES,
are shown  in  Figure 3.

2.4    Heavy-Duty Diesel Vehicles

The most recently available  HDDV  registration data were also found  in the
1981 Polk  data (Table 3-18 of  the  1983 EEA report).   Fractions developed
from this  data and  the  1972  TIUS data (used in MOBILE2)  are used as  the
data base  for  MOBILES.   An  exponential function  is  used  to  smooth  the
data.

The  renonnalized   fractions   are   the   registration   distribution   for
MOBILES.   The three-  sets  of fractions, one  from the  1972  data  (used  in
MOBILE2),  one from  the  1981  data,  and the one  that  is used for  MOBILES,
are shown  in  Figure 4.

-------
                                    9

3.0    MILEAGE ACCUMULATION RATES

The  MOBILES  mileage  accumulation  rates are  different  from the  MOBILE2
rates in three aspects:

       1.  The  MOBILES rates are based  upon  more recent data,  estimated
from  different   or  updated  surveys.    Data  source  differences   are
summarized in Table 3.

                                 Table 3

                References for Mileage Accumulation Rates

Vehicle Class                 MOBILES                    MOBILE2

LDGV                          1979 GMR-NPD*              1978 NTS**
LDDV                          same as LDGV               same as  LDGV

LDGT1                         1977 TIUS***               1972 TIUS
LDGT2                         1977 TIUS                  1972 TIUS
LDDT                          1977 TIUS                  1972 TIUS

HDGV                          1977 TIUS                  1972 TIUS
HDDV                          1977 TIUS                  1972 TIUS
*   General  Motors Research Laboratories  (GMR) derived  these data  from
    National Purchase Diary Research, Inc.  (NPD) survey.

**  National Transportation Survey (for National Science Foundation).

*** Truck  Inventory  and  Use Survey, U.S.  Department  of Commerce,  Bureau
    of the Census.

       2.  The  average annual mileage accumulations  of  a given  vehicle
type are smoothed  by fitting  the data with an  exponential  function.   The
predicted  values  of  the  fitted  data  are  used in  MOBILES.  The  MOBILE2
model used unsmoothed data.

       3.  The  HDDV  rates in MOBILE2  were the averages  over all  weight
classes  for  trucks  above 8,500  Ibs  GVW.   The MOBILES  model  used  the
mileage  distributions  weighted  by   the  projected   registrations  of  four
different  HDDV  classes.   (The  derivation  on  HDDV mileage  accumulation
rates will be discussed in section 3.5.).

-------
                                   10

The  curve-smoothing  methodology used  for mileage  accumulation rates  is
slightly different  from that  used for  registration distributions.  All
data  points  were  included in  the curve  fitting  process.   The mileage
accumulation rates  for all vehicle  types,  except  motorcycles, were the
predicted values from fitting  an exponential function.

The  mileage  accumulation  rates  used  for  MOBILES  are  tabulated   in
Appendices 3 through 6.  The MOBILE2 rates are also listed  for  comparison
purposes.   Since no  update  has  been  done  for the  motorcycle mileage
accumulation rates, they are not listed in this  report.

3.1    Light-Duty Gasoline Vehicles

In the  MOBILE2  model,  the  annual  average mileage accumulation rates for
LDGVs were based upon the  1978  National.Transporation Survey  (NTS) data
(Ref. 4).  The 1983 EEA  report  used  data from the  1981 National  Purchase
Diary Research,  Inc.  (NPD) survey.   However,   the purpose  of this NPD
survey  was  mainly  to  characterize  household  fuel purchases  and  their
travel  patterns.   The  survey  results  were criticized  for  not  including
singles  or  families  that  lived in   apartments,  boarding   rooms,   or
condominiums.  Consequently, estimates from this data set were  considered
to be biased.  For this reason,  results  from  the 1983 EEA report  were not
used.

The 1977 Nationwide Personal Transporation Study (NPTS) data  were used  as
the  data base  for  the  draft  version4  of  MOBILES.  Only  the  first ten
year  data  were  available.   These  ten  year   data  were   used  for   an
exponential  fit.  The  mileage  accumulation rates for  vehicles  eleven
years of age and up were extrapolated from the first ten year data.

In the  final version of MOBILES,  however,  the LDGV mileage  accumulation
rates  were  based  upon   a  special   1979   NPD   survey.    Motor  Vehicle
Manufacturer's  Association (MVMA),  in   their  comments  on  the MOBILES
Draft, suggested that the  1977 NPTS data might also be  biased.   Since the
NPTS  data were  based  upon subjective car owner  responses about vehicle's
mileage  during  a given  year,  rather than  the  actual odometer  readings,
the results from the NPTS  data might over-estimate  the  annual mileage (as
the respondents tended to  round  the annual mileage  to  10,000  miles).  The
1979  NPD survey data, on  the  other hand, were based upon actual  odometer
readings.    Furthermore,   GM   contracted  with   NPD   to   include  460
single-person households   in  this  data  base.   The mileage  accumulation
rates estimated  from  this data  base may be  less biased and be  a  better
reflection of actual travel and driving patterns.
    The draft version of MOBILES  was released on February 14, 1984.   The
    purpose  of  this draft   version was  to  seek  public  opinions  and
    suggestions on the  various  changes  and updates the MOBILES model  had
    made.  Consequently, some revisions were made on  the draft  version.
    The final version of MOBILES was released in June,  1984.

-------
                                   11

The 1979 NPD  survey  data were fitted with  an exponential function.  The
first year mileage accumulation value of the  1979 NPD  survey corresponded
to the average mileage accumulation  rate of vehicles between the ages of
one and  two.   However, MOBILES  needed  the  average mileage accumulation
rate  of  vehicles  between the  ages  of  zero  and  one.   This  value was
derived from the 1979 NPD survey data by redefining the data as ages two
through 21 and estimating year one from  the  fitted exponential function.

The  LDGV annual  average  mileage  accumulation  rates  used  in MOBILE2,
MOBILES Draft and MOBILES final version  are  shown in Figure  5.

3.2    Light-Duty Diesel Vehicles

MOBILE2 used the LDGV  rates for the  LDDV  mileage accumulation rates due
to  insufficient  LDDV  data  available   at   the   time   when MOBILE2 was
finalized.

The 1983  EEA  report  used the  1981 NPD  survey data to estimate the LDDV
average annual mileage accumulations  (Table 3-5  of the 1983 EEA report).
The results  suggested that  the  average annual  mileage accumulations of
LDDVs (for vehicles less  than  six  years  of  age)  would be slightly  higher
than  the  LDGV  rates  during  the  same time  frames.  This  might  be  due to
the fuel price effect on  the driving  patterns — as diesel fuel cost was
less,  diesel  vehicle  owners  tended  to drive  more  than  the gasoline
vehicle owners would.

The MOBILE3 Draft  used  the  results from the  1983 EEA  report  as the data
base.  The first four year rates were from  the LDDV rates.  The rates for
all other years were the  LDGV rates.  The  combined data set was smoothed
by  an exponential  function.   The  predicted values  were  used  for the
MOBILE3 Draft.

In the final version of MOBILE3, however, the LDGV  annual average mileage
accumulations were also  used for LDDVs.  This was  because  the number of
diesels  in  the  1979  GMR-NPD  survey  data  were  not enough  to  develop  a
realistic mileage accumulation distribution  for diesel  vehicles.

The LDDV  mileage accumulation rates used for MOBILE2, MOBILES Draft and
the final version of  MOBILES are plotted in  Figure 6.

3.3    Light-Duty Gasoline and Diesel Trucks

The annual  average mileage accumulation rates for  LDTs from the MOBILE2
model were  obtained  from the  1980 EEA  report (Ref.  5).   Data from the
1983 EEA report were  used as the data base for MOBILES.  In this 1983 EEA
report, the annual mileage  accumulations of LDTs were estimated from the
1977 TIUS (Truck Inventory and Use Survey) data.  To adjust the estimated
1977  rates to  represent  the  1981 level,  it  was assumed that the shape of

-------
                                   12

the  1977  travel distributions remain  constant,  but  the absolute  levels
were adjusted to reflect the 1977-1981 change in average annual miles  per
single  unit  truck, as  reported by  the  Federal  Highway  Administration.
This resulted in  an ajustment factor  of 1.134 for  the light-duty  truck
fleet.  The adjusted rates  were  tabulated in Tables  3-12, 3-13 and 3-15
of the 1983 EEA report.

These  adjusted  mileage  accumulation rates,  however,  were  not  used   in
MOBILES.  The adjustment factor  of 1.134 was thought to be  inappropriate
on two aspects: 1)  The  LDTs  are expected to have somewhat  higher  annual
mileage accumulation  rates  than the  LDVs because many of  the  LDTs  are
used  in commercial  applications  (such  as  in  construction) or  used   as
campers.   However,  in  comparison  with  the MOBILES   LDGV rates,   the
adjusted LDT rates seemed too  high.   For example,  the  first year mileage
of LDGTls was adjusted  to  be 19,724 miles,  while  the first year mileage
from LDGVs was 12,818 miles.   2) It  may not be appropriate  to adjust  the
mileage accumulation rates  "uniformly"  for  all twenty  years.   Thus,  the
unadjusted rates were used as the data base for MOBILES.

The  LDT  mileage   accumulation  rates  used  for  MOBILES  are   listed   in
Appendix 4.   The  MOBILE2 rates  are also  listed  for reference.  Graphic
presentation  of  MOBILES versus  MOBILE2  mileage  accumulation  rates  for
LDTs are shown in Figures 7  through 9.

3.4    Heavy-Duty Gasoline Trucks

The annual average mileage  accumulation rates for HDGVs from the MOBILE2
model  were  obtained  from the  1980 EEA  report.   Data  from  the  1983  EEA
report were used as the  data  base  for  MOBILES.   In this 1983 EEA report,
the  annual  mileage accumulations  of  HDGVs were also estimated from  the
1977 TIUS  data.   Adjustment factors were  derived  in the  same  manner  as
the  light-duty  trucks  to  adjust  the  1977  rates  to represent  the 1981
levels.   The  adjustment  factors were  1.132 for  classes 2B  through  5,
1.119  for class 6, and  0.957  for classes 7 and 8 heavy-duty trucks.   The
adjusted rates can be found in Table 3-19 of the 1983 EEA report.

The  unadjusted  rates from  the  1983  EEA  report  were  used  for MOBILES.
(The reasons  for   using  the unadjusted rates are  the  same  as  the  LDTs,
discussed in  Section  3.3.)   A separate  EEA study (Ref. 6)  on emissions
conversion factors and  fuel economy for heavy-duty trucks indicated that
the  historical  and  projected  sales  (Tables 3-7  and  3-8 of  the  EEA
conversion  factors  study)   and  use  patterns  (Table   2-2  of   the   EEA
conversion  factors study)  of  the HDGVs  over the  years  remain  fairly
stable among the various weight  classes.  Therefore,  it  is appropriate to
use  an overall  average  mileage  distribution  to  represent the   entire
heavy-duty gasoline fleet.

The  HDGV   mileage  accumulation  rates  used  for  MOBILES  are   listed   in
Appendix  5,  with  MOBILE2   rates   listed  also  as   reference.   Graphic
presentation  of  MOBILES versus  MOBILE2  rates  are  shown in  Figure  10.

-------
                                   13

3.5    Heavy-Duty Diesel Vehicles

The HDDV annual average mileage  accumulation rates from  the MOBILE2  model
were obtained  from the 1980 EEA report.   Data from  the 1983 EEA  report
were used  as the  data  base for  MOBILE3.   In  this  1983 EEA report,  the
annual mileage  accumulations  of HDDVs were estimated from the 1977  TIUS
data,  and adjusted to  1981  levels.   The  various classes  of  HDVs  with
respect to their weights are shown in Table 4.

                                 Table 4

                         Classification of HDVs

                   Weight                Gross  Vehicle
                   Class                 Weight (in Ibs)

                   2B                     8,501 - 10,000
                   3-5                   10,000 - 19,500
                   6                     19,501 - 26,000
                   7                     26,001 - 33,000
                   8 A&B                 more  than 33,000

In  the  MOBILE2 model,  the  HDDV  mileage accumulation rates  were  the
averages  over  all  weight  classes.   The  1983  EEA report suggested  that
three weight classes of HDDVs  be considered  separately, since the  sales
trends and usage  patterns  vary  among  the light  (classes 2B-5),  medium
(class 6)  and heavy  (classes  7-8) diesel  trucks.  The  adjusted  mileage
accumulation rates of  these three  classes  were  tabulated in Tables 3-20
through 3-22 of the 1983 EEA report.

In  the  draft  version  of  MOBILE3,  the  unadjusted  mileage  accumulation
rates of  the light (classes  2B-5),  medium,  and heavy HDDVs  from  the 1983
EEA  report were used.   The annual mileage accumulations of these  three
classes were then weighted by  the  estimated  registrations in each  class
by   calendar  year  to  derive  a  unique  class-combined  HDDV   mileage
accumulation distribution for each calendar year.

For  the  final  version of MOBILES,  however,  an  additional  class  of
HDDVs—class 2B was created,  or separated from the  classes  2B-5  figures.
The sales of class  2B diesel trucks are projected  to  be  increasing  in the
future years,  as  evidenced  by Table 3-8  of  the  EEA study on  emissions
conversion  factors  and fuel  economy  for heavy-duty   trucks  (Ref.  6).
Also, class  2B trucks  probably have mileage accumulation rates more like
class 2A  trucks  than classes  3-5 trucks.   Therefore, it  was  necessary to
include  a separate category  for these  class  2B  diesel trucks  in  the
mileage calculations.   Since the average annual mileage  accumulations for
class 2B  HDDVs were  not readily available in the  1983 EEA report,  the
LDGT2 mileage accumulation rates were used.

-------
                                   14

The  smoothed  mileage accumulation rates  of the  four  HDDV classes  —2B,
light  (classes  3-5),  medium  (class  6),  and  heavy  (classes  7-8), are
listed  in  Appendix 6.  The  estimated total HDDV registrations by  class
for years  1980 through  2000, obtained from another  EEA report5 (Ref.  7)
are  listed  in  Appendix  7.   Graphic  presentations  of  MOBILES   versus
MOBILE2 rates are shown in Figure 11.

An example of  how  the overall HDDV mileage  accumulations are  derived from
all weight  classes in a  given  calendar year  is  briefly  presented  here.
The  estimated total 1982 HDDV  registrations in  millions  were 0.049 for
class 2B, 0.009 for classes 3-5, 0.135 for  class  6,  and  1.599 for  classes
7-8 heavy-duty trucks  (obtained from Appendix  7).   Therefore, the  total
1982  HDDV  registrations  for all classes  are  1.792 million  (sum of all
classes, last  column in Appendix 7).   This yields registration fractions
of 2.7 percent for class 2B, 0.5 percent for classes 3-5,  7.5 percent for
class  6,  and 89.3  percent for  classes  7-8  HDDVs.   The  overall  HDDV
mileage accumulations by age for the  calendar year of  1982 were obtained,
as shown in Table  5, by weighting the annual average mileage  accumulation
rates of the  four  weight  classes (from Appendix  6) with their respective
registration fractions.
    Appendix B  of the  EEA report  entitled  "October 1983  Output of  the
    Highway Fuel Consumption Model."

-------
Age

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

                                 Table 5

                1982 Calendar Year HDDV Registration Mix*
2B
(2.
18,
16,
15,
1*,
13,

,7%)
,352
,946
,648
,449
,342
3-5
(0.
45,
39,
34,
30,
26,
(Light)
5%)
544
671
558
092
213
6
(7
53
46
41
36
31
(Medium)
.5%)
,370
,901
,190
,206
,812
7-8
(89
82,
78,
68,
62,
56,
(Heavy)
.3%)
288
984
328
263
737
Weighted
Average
78,209
71,134
64,701
58,857
53,543
12,320
11,376
10,504
9,700
8,956
22,834
19,898
17,332
15,098
13,152
27,948
24,556
21,575
18,956
16,655
51,700
47,111
42,930
39,119
35,647
48,711
44,318
40,325
36,692
33,389
8,270
7,637
7,052
6,511
6,012
11,456
9,979
8,693
7,572
6,596
14,632
12,856
11,296
9,925
8,719
32,483
29,599
26,972
24,578
22,396
30,385
27,652
25,167
22,906
20,849
5,552
5,126
4,734
4,371
4,036
5,746
5,005
4,360
3,798
3,308
7,661
6,728
5,913
5,196
4,565
20,408
18,597
16,946
15,442
14,071
18,978
17,275
15,726
14,316
13,033
    The registration  fractions for the  four  HDDV classes  are  calculated
    from  the  total  registrations  by  class  (Appendix  7)  divided  by  the
    overall HDDV  registrations  (last  column  of Appendix 7).   Mileage
    accumulation rates by age  for  the  four HDDV classes are from Appendix
    6.

-------
                                   16

4.0    TOTAL VEHICLE COUNTS

The total numbers  of  vehicles of each vehicle  type are used  to  estimate
the  portion  of  the  total  mobile  source  vehicle-miles-traveled  (VMT)
driven  by   each  vehicle  type.   In  MOBILE2/MOBILE3,  the  total  vehicle
counts are given for six vehicle  types: LDVs, LDTs,  LDGT2s,  HDGVs, HDDVs,
and MCs.  Dieselization rate adjustment  factors are used in  the  computer
program  to  split  the  LDV count  into total  LDGV  and LDDV  registration
counts,  and  also  to  split   the  LOT  count  into  total  LDGT1  and  LDDT
registration counts.

Table  6 is  a summary  of the MOBILE2/MOBILE3  total vehicle counts  by
vehicle  type.  The MOBILE2 counts were obtained from 1978  registrations
(Ref.  3),  and  the  MOBILES   counts  were  derived  from   estimates  in
Department of Energy's Highway Fuel Consumption Model (Ref.  7).

                                 Table 6

                    Total  Vehicle Counts in Millions

Vehicle Type                MOBILE3               MOBILE2

LDGV/LDDV                   105,839               102,900
LDGT1/LDDT                   18,072                12,900
LDGT2                        11,506                 5,700
HDGV                          4,650                 4,700
HDDV                          1,640                 1,000
MC                            5,600                 7,300

-------
                                   17

References

1.     Energy and  Environmental Analysis,  Inc.,  "Fleet  Characterization
       Data Final Report," prepared for the U.S. Environmental  Protection
       Agency, December 20,  1983.

2.     "Compilation of  Air  Pollutant  Emission Factors:   Highway  Mobile
       Sources," EPA 460/3-81-005,  March 1981.

3.     Energy   and   Environmental   Analysis,   Inc.,   "National   MOBILE2
       Variable   Estimates,"    prepared   for   the   U.S.   Environmental
       Protection Agency, July 1980.

4.     Correspondence from  Mr. Barry  Felrice, National  Highway  Traffic
       Safety  Administration,   to  Mr.  Michael  P.  Walsh,   Environmental
       Protection Agency, March 31, 1980.

5.     Energy   and  Environmental  Analysis,   Inc.,   "Techniques   for
       Estimating MOBILE2 Variables," prepared for the U.S.  Environmental
       Protection Agency, July 1980.

6.     Energy and Environmental Analysis, Inc., "Historical  and Projected
       Emissions  Conversion  Factors  and  Fuel  Economy   for  Heavy-Duty
       Trucks   1962-2002,"  Prepared   for  Motor  Vehicle   Manufacturers
       Association, December,  1983.

7.     Energy   and  Environmental  Analysis,   Inc.,   "The   Highway  Fuel
       Consumption Model:  Tenth Quarterly Report," prepared  for  the  U.S.
       Department of Energy, October,  1983.

-------
                                   18




                               Appendix 1




                MOBILES July 1 Registration Distributions
Vehicle
Age
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20f

LDV*
0.080
0.101
0.095
0.089
0.083
0.077
0.071
0.065
0.059
0.053
0.047
0.041
0.035
0.029
0.023
0.017
0.011
0.008
0.006
0.008
                             LOT**
HDGV
HDDV
*   The same distribution is used for LDGVs and LDDVs.




**  The same distribution is used for LDGTls,  LDGT2s  and  LDDTs.




*** Same as MOBILE2.
MC***
0.080
0.101
0.095
0.089
0.083
0.067
0.085
0.081
0.077
0.073
0.079
0.136
0.116
0.099
0.085
0.090
0.151
0.126
0.105
0.088
0.133
0.145
0.138
0.116
0.123
0.077
0.071
0.065
0.059
0.053
0.069
0.065
0.061
0.057
0.053
0.072
0.062
0.053
0.045
0.038
0.073
0.061
0.051
0.043
0.036
0.114
0.069
0.044
0.024
0.009
0.047
0.041
0.035
0.029
0.023
0.048
0.044
0.040
0.036
0.032
0.033
0.028
0.024
0.020
0.018
0.030
0.025
0.021
0.017
0.014
0.085
0.0
0.0
0.0
0.0
0.017
0.011
0.008
0.006
0.008
0.028
0.024
0.020
0.016
0.024
0.015
0.013
0.011
0.009
0.045
0.012
0.010
0.008
0.007
0.031
0.0
0.0
0.0
0.0
0.0

-------
                               19
                           Appendix 2

    MOBILES Gasoline/Diesel Sales Fractions of LDVs and LDTs
MODEL
YEAR
Pre-1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995+

LDGV
1.000
0.997
0.997
0.996
0.991
0.972
0.966
0.939
0.954
0.947
0.940
0.934
0.927
0.920
0.910
0.900
0.887
0.887
0.886
0.886
0.885
0.885
LDGV
1.000
0.997
0.997
0.996
0.991
0.972
0.966
0.939
0.954
0.947
0.940
0.934
0.927
0.920
0.910
0.900
0.887
0.887
0.886
0.886
0.885
LDDV
0.000
0.003
0.003
0.004
0.009
0.028
0.034
0.061
0.046
0.053
0.060
0.066
0.073
0.080
0.090
0.100
0.113
0.113
0.114
0.114
0.115
LDGT (1&2)
1.000
0.998
0.997
0.995
0.991
0.972
0.966
0.940
0.920
0.900
0.870
0.840
0.820
0.790
0.760
0.730
0.706
0.697
0.688
0.679
0.670
LDDT
0.000
0.002
0.003
0.005
0.009
0.028
0.034
0.060
0.080
0.100
0.130
0.160
0.180
0.210
0.240
0.270
0.294
0.303
0.312
0.321
0.330
                         0.115
0.661
0.339
Source:   Draft version  of  "Diesel  Particulate  Study,"  Table  2-6,
Standards Development and Support Branch, Emission  Control  Technology
Division, Office of  Air  and Radiation, U.S. Environmental  Protection
Agency, 1984.

-------
                                   20

                               Appendix 3

           Annual Average Mileage Accumulation Rates for LDVs


                                  MOBILES
Vehicle
Age
1 .
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20+
LDGV
Final
12,818
12,102
11,427
10,789
10,187
9,619
9,082
8,575
8,096
7,645
7,218
6,815
6,435
6,076
5,737
5,416
5,114
4,829
4,559
4,305

Draft
13,925
13,125
12,375
11,661
10,992
10,361
9,766
9,205
8,677
8,179
7,716
7,274
6,857
6,464
6,094
5,745
5,415
5,105
4,813
4,537
LDDV
Final
12,818
12,102
11,427
10,789
10,187
9,619
9,082
8,575
8,096
7,645
7,218
6,815
6,435
6,076
5,737
5,416
5,114
4,829
4,559
4,305

Draft
15,332
14,349
13,428
12,567
11,761
11,007
10,301
9,640
9,022
8,443
7,902
7,395
6,921
6,477
6,062
5,673
5,309
4,969
4,650
4,352
                                                                MOBILE2*
                                                                14,400
                                                                13,900
                                                                13,400
                                                                12,800
                                                                12,300
                                                                11,800
                                                                11,200
                                                                10,700
                                                                10,200
                                                                 9,600
                                                                 9,100
                                                                 8,600
                                                                 8,000
                                                                 7,500
                                                                 7,000
                                                                 6,400
                                                                 5,900
                                                                 5,400
                                                                 4,900
                                                                 4,300
*  In MOBILE2, the same LDGV rates are  also  used  for LDDVs.

-------
LDGT1
17,394
16,132
14,961
13,876
12,869
LDGT2
18,352
16,946
15,648
14,449
13,342
LDDT
17,552
16,262
15,068
13,961
12,936
LDGT1
15,600
14,000
12,500
11,200
10,100
LDGT2
15,800
14,200
12,600
11,400
10,300
LDDT
15,700
15,100
14,600
14,000
13,000
                                    21

                                Appendix 4

           Annual Average  Mileage Accumulation Rates for LOTS

Vehicle                MOBILES                          MOBILE2
Age

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
204-
11,935
11,069
10,266
9,521
8,830
12,320
11,376
10,504
9,700
8,956
11,986
11,105
10,290
9,534
8,833
9,400
8,600
8,000
7,500
7,700
9,500
8,700
8,100
7,600
7,100
12,100
11,000
9,700
9,300
7,400
8,189
7,595
7,044
6,533
6,059
8,270
7,637
7,052
6,511
6,012
8,185
7,583
7,026
6,510
6,032
6,600
6,200
5,900
5,500
5,200
6,600
6,300
6,000
5,500
5,200
6,800
6,100
5,500
5,000
4,600
5,619
5,211
4,833
4,483
4,157
5,552
5,126
4,734
4,371
4,036
5,589
5,179
4,798
, 4,446
4,119
5,000
4,700
4,400
4,400
4,400
5,100
4,700
4,400
4,400
4,400
4,200
3,800
3,100
3,100
3,100

-------
Vehicle
Age

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20f
22
Appendix 5
Annual Average Mileage Accumulation Rates
MOBILES
19,967
18,077
16,365
14,815
13,413
12,143
10,993
9,952
9,010
8,156
7,384
6,685
6,052
5,479
4,960
4,490
4,065
3,680
3,332
3,016


for HDGVS
MOBILE2
21,000
18,900
17,200
15,500
14,100
12,600
11,300
10,300
9,400
8,600
8,000
7,300
6,700
6,300
5,900
5,500
5,200
5,000
5,000
5,000

-------
                                   23

                               Appendix 6

           Annual Average Mileage Accumulation Rates for HDDVS

                                                             MOBILE2
Vehicle   :	MOBILES    	   (All Weight
Age       2B_         3-5 (Light)   6 (Medium)   7-8 (Heavy)    Classes)


 1        18,352     45,544       53,370      82,288         70,400
 2        16,946     39,671       46,901      74,984         66,900
 3        15,648     34,558       41,190      68,328         60,600
 4        14,449     30,092       36,206      62,263         54,200
 5        13,342     26,213       31,812      56,737         47,900
 6        12,320     22,834       27,948      51,700         43,700
 7        11,376     19,898       24,556      47,111         39,400
 8        10,504     17,332       21,575      42,930         36,600
 9         9,700     15,098       18,956      39,119         34,500
10         8,956     13,152       16,655      35,647         33,100
2B
18,352
16,946
15,648
14,449
13,342
12,320
11,376
10,504
9,700
8,956
8,270
7,637
7,052
6,511
6,012
5,552
5,126
4,734
4,371
4,036
3-5 (Light)
45,544
39,671
34,558
30,092
26,213
22,834
19,898
17,332
15,098
13,152
11,456
9,979
8,693
7,572
6,596
5,746
5,005
4,360
3,798
3,308
6 (Medium)
53,370
46,901
41,190
36,206
31,812
27,948
24,556
21,575
18,956
16,655
14,632
12,856
11,296
9,925
8,719
7,661
6,728
5,913
5,196
4,565
7-8 (Heavy)
82,288
74,984
68,328
62,263
56,737
51,700
47,111
42,930
39,119
35,647
32,483
29,599
26,972
24,578
22,396
20,408
18,597
16,946
15,442
14,071
11         8,270     11,456       14,632      32,483         32,400
12         7,637      9,979       12,856      29,599         31,700
13         7,052      8,693       11,296      26,972         31,000
14         6,511      7,572        9,925      24,578         29,600
15         6,012      6,596        8,719      22,396         27,500
16         5,552      5,746        7,661      20,408         24,600
17         5,126      5,005        6,728      18,597         20,400
18         4,734      4,360        5,913      16,946         17,600
19         4,371      3,798        5,196      15,442         17,600
204-        4,036      3,308        4,565      14,071         17,600

-------
                                   24

                                 Appendix 7

                     Total HDDV Registrations  by   Class
                              for 1980  through 2000
Calendar
Year
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
Total Registrations (in Millions)
Class 2B
0.000
0.000
0.049
0.104
0.185
0.274
0.370
0.475
0.588
0.707
0.831
0.960
1.092
1.225
1.354
1.480
1.600
1.712
1.816
1.912
1.999
3-5 (Light)
0.006
0.006
0.009
0.013
0.022
0.037
0.053
0.071
0.089
0.106
0.122
0.137
0.151
0.165
0.178
0.190
0.202
0.212
0.222
0.230
0.238
6 (Medium)
0.112
0.124
0.135
0.141
0.153
0.166
0.177
0.185
0.193
0.201
0.208
0.215
0.222
0.229
0.237
0.245
0.253
0.261
0.269
0.276
0.283
7-8 (Heavy)
1.521
1.581
1.599
1.592
1.641
1.719
1.816
1.927
2.041
2.151
2.258
2.362
2.471
2.581
2.693
2.807
2.914
3.015
3.108
3.194
3.273
Overall
1.639
1.711
1.792
1.850
2.001
2.196
2.416
2.658
2.911
3.165
3.419
3.674
3.936
4.200
4.462
4.722
4.969
5.200
5.415
5.612
5.793
Source:  EEA, prepared for  the  U.S.  Department of Energy, "The Highway  Fuel
Consumption Model, 10th Quarterly Report," October,  1983.

-------
                          Figure 1
           LDV Registration Distributions
    0.12-1

-------
 0)
 CD
cd
   0.10-
   0.08-
 ^^w
 O

'S
 3  0.06 H
CD
0)
FT*  0.04 H
O  0.02-


-------
                        Figure 3
         HDGV Registration Distributions
   0.20-,
0)



0)

O
• •—I

Cl
h—I

•s

0)
CD
0)
O
(H
0)
PL,
   0.15-
   0.10-
   0.05-
   0.00
                       1	1	1	1	1	1	1	1	1	1	1	\	T~
                       8  9 10 11 12 13 14 15 16 17 18 19 20
           234567
                     Vehicle Age (Years)
                                                         Legend

                                                        A MOBILE2

                                                        X 1981 FOLK

                                                        D MOBILE3

-------
   0.20-.
                       Figure 4
         HDDV Registration Distributions
0)
$  0.15-^
JU
(D
O
(H
0)
a*
   o.ioH
0.05 H
0.00 -—i—i—i—i—i—r

      123456
                     i—i—i—i—i—i—i—i  i   i  i  i
                     7 8  9 10 11 12 13 14 15 16 17 18 19 20
                     Vehicle Age (Years)
                                                     Legend

                                                    A MOBILE2

                                                    X 1981 POLK

                                                    0 MOBILE3

-------
                     Figure 5 : LDGV
         Vehicle Miles Traveled Distributions
   16000-.
   14000-


 0)
i—H
 8J 12000
 (3
ft
 w
JS 10000
• I—I
s
o
    8000-
    6000-
    4000
I  T I  I   I  I  I  I  I  I  I  I  I   I  I  I  I  I  I  I
1  2 3 4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20
                                               Legend
                                              A MOBILE2

                                              X M3 DRAFT

                                              D MOBILE3
                      Vehicle Age (Years)

-------
 •~^»
£
 W
   16000-1
   14000-

JO)
£  12000-
   10000-
(D
r—i
•    8000H
    6000-
    4000
                     Figure 6 : LDDV
         Vehicle Miles Traveled Distributions
                  I  I
I  I  I  I  I   I  I  I  I  I  I  I  I  T
                                   Legend
                                  A MOBILE2
                                  X M3 DRAFT
                                  D MOBILE3
          1  23 4 5 6 7 8  9 10 11 12 13 14 15 16 17 18 19 20
                      Vehicle Age (Years)

-------
                    Figure 7 : LDGT1
         Vehicle Miles Traveled Distributions
   80000-1
 CD
•—I

 I
CO

-------
                    Figure 8 : LDGT2
         Vehicle Miles Traveled Distributions
   20000 -.
 
-------
                     Figure 9 : LDDT
         Vehicle Miles Traveled Distributions
   20000-1
 CD
i—i
 0)

 Ed
w
CD
•—i
•rH

a

CD
i—i
O
   15000-1
10000-1
    5000 H
                                                         Legend

                                                        A MOBILE2

                                                        X MOBILE3
          12345
                 6789 10 11 12 13 14 15 16 17 18 19 20

                   Vehicle Age (Years)

-------
                    Figure 10 :  HDGV

         Vehicle Miles Traveled Distributions
   25000-1
   20000-
I
CO

CD
   15000-
0)  10000
I—I
o
    5000-
                                                       Legend


                                                      A MOBILE2


                                                      X MOBILE3
                    i  i  i  i  r r
I  T  I  I  I  I
         1 2 3 4  5  6  7  8  9  10 11 12 13 14 15 16 17 18 19 20


                     Vehicle Age (Years)

-------
-o

-------