SUPPORTING
STATEMENT
FOR
THE
MOBILE
SOURCE
EMISSION
FACTOR
RECRUITMENT
PART
A
1.
Identification
of
the
Information
Collection
1(
a)
Title
and
Number
of
the
Information
Collection
ICR:
Mobile
Source
Emission
Factor
On­
Highway
Recruitment
OMB
Control
Number
2060­
0078
1(
b)
Short
Characterization
The
EPA,
Office
of
Transportation
and
Air
Quality
(
OTAQ),
Assessment
and
Standards
Division
(
ASD)
administers
a
nationwide
effort
to
collect
data
on
emission
levels
from
in­
use
vehicles
recruited
from
the
general
public
voluntarily.
ASD
carries
out
this
effort
through
the
Emission
Factor
Program
(
EFP).
EFP
gathers
activity
and
emissions
measurements
either
by
sampling
a
vehicle's
emissions
and
activity
during
normal
operation
or
alternatively
during
a
driving
simulation
developed
for
laboratory
testing
using
a
chassis
dynamometer.

The
purpose
of
the
EFP
is
to
collect
a
representative
sample
of
in­
use
emissions
and
activity
data
to
assure
that
EPA's
current
vehicle
emission
model,
MOBILE6,
can
accurately
predict
the
impact
of
past,
current,
and
future
vehicle
emission
control
policies
and
standards
on
air
quality.
MOBILE6
currently
performs
the
task
of
three
historical
EPA
models,
its
predecessor,
Mobile
5,
for
gaseous
emissions,
PART5
for
particulate
emissions,
and
MOBTOX
for
air
toxic
compounds.
MOBILE6
is
a
required
product
provided
by
EPA
to
support
federal,
state,
and
local
air
pollution
agency
decisions
that
mitigate
mobile
source
emissions
and
make
transportation
policy.
Furthermore,
the
data
in
this
information
collection
shall
be
used
to
develop
MOBILE6'
s
replacement,
the
Multi­
scale
Motor
Vehicle
&
Equipment
Emission
System
(
MOVES).
MOVES
will
be
a
more
efficient,
data
driven
model
using
second­
by­
second
emission
and
activity
data
collected
from
in­
situ
sampling
of
vehicles
in
the
real
world
and
secondby
second
emissions
test
data
obtained
in
laboratory
testing.

All
suitably
collected,
documented,
and
quality­
assured
data
is
stored
in
the
EFP
data
base,
the
Mobile
Source
Observation
Data
Base
(
MSOD)
and
is
available
on
request
to
interested
parties
and
individuals
on
Compact
Disc.
The
data
in
MSOD
is
then
used
by
ASD
modeling
staff
to
develop,
create,
edit,
and
test
the
algorithms
that
are
used
in
both
MOBILE
6
and
MOVES.

MOBILE6
and
MOVES
require
EFP
to
obtain
by
either
in­
situ
sampling
or
testing
vehicle
exhaust
emission
data
for
volatile
organic
compounds
(
VOCs),
carbon
monoxide
(
CO),
oxides
of
2
nitrogen
(
NOx),
and
particulate
matter
(
PM).
Other
pollutants
that
are
required
by
EFP
to
assess
and
model
are
vehicle
VOC
emissions
from
evaporative
losses,
species
of
VOCs,
exhaust
and
evaporative,
and
PM
that
are
considered
air
toxics.
These
latter
pollutants
are
only
directly
assessed
using
laboratory
procedures.

The
aggregation
of
emissions
across
a
population
of
vehicles
to
produce
their
emission
inventory
requires
that
either
the
model
or
model
users
have
knowledge
of
vehicle
use
(
activity).
In
the
past
this
was
entirely
based
on
data
derived
from
vehicles
that
were
not
part
of
the
emission
measurement
process.
The
activity
data
was
based
on
either
chase
car
data
for
speed
in
traffic
vehicle
behavior
and
contractor
analysis
of
Department
of
Transportation
data
to
obtain
vehicle
mileage
accumulation
and
vehicle
trips.
For
this
information
collection,
activity
data
shall
be
sampled
with
and
without
emissions
data
in
the
field.

The
vehicle
types
needed
to
meet
the
goals
of
this
information
collection
are
the
three
major
emission
classes:
light
duty
vehicles
(
cars),
light
duty
trucks
(
pickups,
vans,
and
sport
utility
vehicles)
and
heavy
duty
vehicles
(
vehicles
with
gross
vehicle
weight
rating
greater
than
8500
pounds)
that
operate
on
either
gasoline
or
diesel
fuel.
These
types
shall
be
stratified
in
model
year
groups
that
correspond
to
major
changes
in
vehicle
class
emission
standards.

The
sample
size
for
these
strata
shall
be
based
on
their
contribution
to
the
emission
inventory,
the
uncertainty
in
their
average
emission
level,
and
the
precision
requirement
of
20%
of
the
average
emission
level
or
the
Tier
2
BIN
light
duty
emission
standard
whichever
is
greatest.
The
emission
inventory
shall
be
that
projected
in
2007
from
Mobile
6
and
the
uncertainty
in
each
stratum
shall
be
calculated
using
the
emissions
data
stored
in
MSOD.
This
method
will
result
in
a
stratified
sample
that
will
be
efficient,
meet
the
program
requirements,
and
be
within
the
means
of
the
resources
available
to
EFP.
The
actual
and
random
selection
of
participants
shall
be
either
from
state
vehicle
registration
lists
or
from
a
state
inspection
and
maintenance
test
lane.

Uncertainty
within
strata
is
believed
driven
by
vehicles
that
have
emissions
that
exceed
the
mean
emission
factor
for
the
strata
by
a
large
margin,
high­
emitting
vehicles.
EFP
shall
target
these
vehicles
for
recruitment
to
determine
their
effect
on
the
uncertainty
of
their
sample
stratum,
improve
the
sampling
plan's
sample
efficiency,
and
adjust
the
strata's
sample
size
accordingly
to
improve
the
precision
of
the
modeled
emission
rate
estimates.
The
targeting
shall
be
done
using
either
remote
sensing
devices
or
tail
pipe
process
analyzers
on
recruited
vehicles.

Initially,
the
information
collection
shall
be
done
in
two
specific
localities
that
have
or
will
have
the
capability
of
collecting
emission
test
data
for
pollutants
not
yet
within
the
capabilities
of
in­
situ
sampling,
namely
PM,
VOCs
from
evaporative
emissions,
and
air
toxics.
One
of
the
areas,
Kansas
City,
was
chosen
for
the
first
major
effort
by
EFP
because
it
is
an
area
where
in­
use
vehicle
emissions
have
not
been
sampled
and
where
they
are
not
subject
to
a
state
Inspection
and
Maintenance
program.
This
is
in
response
to
Office
of
Management
and
Budget
and
the
Department
of
Energy's
comments
on
EFP
practices.
Furthermore,
emission
and
activity
3
sampling
shall
be
done
in
association
with
all
in­
use
testing
programs.
This
assures
that
emissions
and
activity
affected
by
local
factors
like
climate,
topography,
operator
behavior,
fuels,
road
conditions,
and
traffic
are
assessed
for
those
areas.
This
data
shall
prove
useful
in
the
national
emission
and
activity
sampling
program
that
shall
be
executed
in
the
last
two
years
of
the
information
collection.

The
national
sampling
program
shall
select
areas
of
the
country
using
random
proportional
sampling,
proportional
to
the
area's
contribution
to
the
national
vehicle
population.
This
proportional
random
sample
shall
result
in
a
list
of
sample
areas,
participants,
their
vehicles
and
vehicle­
specific
national
weighting
factors,
based
on
the
vehicle
type's
frequency
in
the
national
population.
The
weighting
factor
shall
be
used
to
scale
the
vehicle's
sample
emissions
and
activity
to
national
estimates
of
that
vehicle's
stratum.

In
the
national
and
area
specific
to
vehicle
testing
the
program
participants
shall
be
selected
randomly
from
area­
specific
pools
of
possible
participants.
The
methods
of
recruitment
shall
be:
the
use
of
postal
cards
or
letters
sent
to
the
random
sample
of
vehicle
owners
identified
through
State
motor
vehicle
registration
lists
or
from
the
random
selection
of
owners
solicited
from
an
I/
M
test
lane.

During
this
information
collection,
EFP
shall
obtain
by
competitive
procurement,
vehicle
recruitment
services
apart
from
the
emission
sampling
and
testing
contractor.
The
contractor
shall
be
required
to
have
expertise
in
the
sampling.
Their
charge
is
to
identify
the
pool
of
possible
participants,
determine
important
demographic
data
within
the
pool,
and
use
the
demographic
data
to
increase
the
programs'
saliency,
identify
appropriate
incentives,
and
assure
rates
of
participation
that
are
greater
than
80%.
In
both
the
testing
and
sampling
programs,
EFP
will
require
the
recruitment
contractor
to
contact
negative
responders,
identify
why
they
choose
not
to
participate,
and
make
reasonable
attempts
to
get
them
to
be
willing
volunteers.

The
owner
will
not
be
asked
any
operational
perspective
or
prospective
questions
concerning
their
vehicle
for
either
the
testing
or
sampling
programs.
Participants
in
the
sampling
program,
however,
will
be
asked
to
fill
out
a
log
to
record
certain
tasks
and
conditions
that
occur
during
the
sampling
period
in
which
they
will
be
directly
involved,
have
full
knowledge
of,
and
directly
affect
vehicle
emissions
The
information
collection
will
involve
420
number
of
respondents
at
total
cost
of
1.2
million
dollars
per
year
for
three
years.

2(
a)
Need/
Authority
for
the
Collection
The
Agency
requires
States
to
submit
certain
air
quality
reports
to
EPA.
The
information
gathered
by
the
EFP
is
required
for
the
development
of
State
Implementation
Plans
(
SIPs),
Reasonable
Further
Progress
(
RFP)
reports,
attainment
status
assessments
for
the
National
Ambient
Air
Quality
Standards
(
NAAQS),
etc.
4
The
legislative
basis
for
the
Emission
Factor
Program
is
Section
103(
a)(
1)(
2)(
3)
of
the
Clean
Air
Act:

Sec.
103.
(
a)
The
Administrator
shall
establish
a
national
re­
search
and
development
program
for
the
prevention
and
control
of
air
pollution
and
as
part
of
such
program
shall
­
(
1)
conduct,
and
promote
the
coordination
and
acceleration
of,
research,
investigations,
experiments,
demonstrations,
surveys,
and
studies
relating
to
the
causes,
effects
(
including
health
and
welfare
effects),
extent,
prevention,
and
control
of
air
pollution;
(
2)
encourage,
cooperate
with,
and
render
technical
ser
vices
and
provide
financial
assistance
to
air
pollution
control
agencies
and
other
appropriate
public
or
private
agencies,
institutions,
and
organizations,
and
individuals
in
the
conduct
of
such
activities;
(
3)
conduct
investigations
and
research
and
make
surveys
concerning
any
specific
problem
of
air
pollution
in
coopera­
tion
with
any
air
pollution
control
agency
with
a
view
to
recommending
a
solution
of
such
problem,
if
he
is
requested
to
do
so
by
such
agency
or
if,
in
his
judgment,
such
problem
may
affect
any
community
or
communities
in
a
State
other
than
that
in
which
the
source
of
the
matter
causing
or
contributing
to
the
pollution
is
located;
Section
103(
b)(
1)
of
the
Clean
Air
Act
authorizes
the
Administrator
to
"
collect
and
make
available,
through
publications
and
other
appropriate
means,
the
results
of
and
other
information,
including
appropriate
recommendations
by
him
in
connection
therewith,
pertaining
to
such
research
and
other
activities."

2(
b)
Practical
Utility/
Users
of
the
Data
The
emission
data
collected
through
the
mobile
source
Emission
Factor
Program
provides
the
basis
and
updates
for
the
mathematical
on­
highway
vehicle
emission
model
MOBILE6.
Additionally,
the
data
collected
under
this
information
collection
shall
be
used
in
the
new
model,
MOVES.
MOVES,
unlike
MOBILE6,
shall
be
based
more
on
in­
situ
sampled
emission
and
activity
data
than
on
emission
data
collected
by
vehicles
tested
using
one­
to
many­
drive
cycles
simulating
vehicle
activity.
The
data
is
also
published
in
a
series
of
volumes
known
as
AP­
42,
the
compilation
of
both
mobile
source
and
stationary
source
emission
factors.

MOBILE6
is
and
MOVES
will
be
a
tool
of
Federal,
State,
and
local
government
agencies,
academic
institutions,
commercial
entities,
and
private
citizens
who
are
responsible
for
or
interested
in
identifying
highway
vehicle
contributions
to
current
and
future
inventories
of
air
pollutants.
A
list
of
current
users
is
found
in
Appendix
A.
1
MOBILE6
and
its
replacement
MOVES
are
essential
tools
for
a
number
of
Federal
air
pollution
monitoring
and
control
activities:

°
National
air
quality
policymakers
need
to
assess
current
and
future
trends
in
the
attainment
of
National
Ambient
Air
Quality
Standards.
These
assessments
cannot
1Mobile
Source
Observation
Data
(
MSOD)
Database
Update,
October
31,
2002,
Eastern
Research
Group,
Inc
5
be
made
without
use
of
MOBILE6.
For
this
reason,
MOBILE6
plays
and
MOVES
will
play
a
key
role
in
discussions
of
air
quality
strategy
within
both
the
executive
and
legislative
branches
of
the
Federal
government,
the
states,
regional
air
quality
authorities,
local
governments,
and
other
organizations
that
have
an
interest
in
rulemaking
required
by
the
Amendments
to
the
Clean
Air
Act.
This
is
a
critical
use
of
MOBILE6
and
MOVES.

°
EPA's
national
motor
vehicle
pollution
control
program
effectiveness
relies
on
analyses
provided
by
MOBILE6.
Past
and
present
emission
standards
must
be
evaluated
with
in­
use
data
to
determine
how
well
those
programs
meet
their
air
pollution
mitigation
goals.
These
assessments
determine
whether
the
agency
should
revise
present
and/
or
adopt
new
policies
and
rules
concerning
vehicle
emissions.

°
State
and
local
government
bodies
are
responsible
for
the
development
of
State
Implementation
Plans
aimed
at
assuring
that
national
ambient
air
quality
goals
are
met.
These
plans
must
incorporate
a
balance
of
mobile
source
and
stationary
source
compliance
measures
adequate
to
assure
that
air
quality
standards
are
attained
and
maintained
into
the
future.
For
areas
requiring
more
stringent
measures
to
control
motor
vehicle­
related
pollutants,
use
of
MOBILE6
is
essential
for
determining
what
vehicles
and
regions
are
best
targeted
to
specific
policies
and
programs
to
mitigate
air
pollution.
Examples
of
local
policies
and
programs
are
public
transportation
subsidies,
special
fuel
programs,
traffic
mitigation,
and
motor
vehicle
inspection
and
maintenance
(
I/
M)
programs.

3.
Nonduplication,
Consultations,
and
Other
Collection
Criteria
3(
a)
Nonduplication
In
2002
EPA
had
Eastern
Research
Group
identify
all
suitable
data
sources
available
to
populate
the
data
set
to
create
the
emission
model
MOVES1.
The
data
identified
as
meeting
the
specifications
of
the
statement
of
work
is
to
be
incorporated
in
the
prototype
version
of
MOVES.
The
data,
all
of
it
collected
on
a
second­
by­
second
basis,
was
judged
for
quality
and
appropriateness
for
MOVES.
In
general
the
data
is
useful.
There
is
a
rich
source
of
on­
highway,
heavy
duty
diesel
vehicle
data
that
addresses
past
deficiencies
in
the
EFP
program.
However,
there
remain
sampling
problems
with
the
delivered
data
sets.
The
chief
deficiencies
are:

°
On
a
quantity
basis
most
of
the
data
is
from
several
I/
M
areas
and
from
local
I/
M
2P
153­
155,
Modeling
Mobile­
Source
Emissions,
National
Research
Council,
2000,
National
Academy
Press
3A
List
of
Compounds
Emitted
from
On­
Road
and
Non­
Road
Mobile
Sources,
Report
No.
SR01­
02­
01,
Sierra
Research,
Inc,
February
21,
2001.

6
tests.
These
tests
are
of
limited
use
and
2quality.

°
The
programs
using
multi­
drive
cycles
in
their
testing
were
small
and
were
performed
in
California
on
California
vehicles.
This
has
limitations
concerning
its
application
to
the
national
fleet.
This
same
limitation
also
applies
to
the
air
toxics
data
collected
°
Many
of
the
vehicles
tested
were
lease,
state
fleet,
or
research
staff
vehicles.
It
is
not
to
say
that
these
vehicles
do
not
represent
in­
use
vehicles,
but
they
may
not
contain
a
significant
number
of
high­
emitting
vehicles.

In
2001
EPA
had
Sierra
Research
Institute
review
the
literature
to
identify
and
determine
the
quality
of
test
data
measuring
air
toxic
compounds
and
PM
from
all
mobile
sources
including
highway
vehicles3.
The
discovered
data
covering
highway
vehicles
have
been
incorporated
into
the
air
toxics
module
of
MOBILE6.
None
of
that
data
is
collected
on
a
second­
by­
second
basis,
chiefly
because
it
is
not
technically
feasible.
The
data
in
the
report
that
pertains
to
highway
sources
is
over
"
x"
years
old
and
the
volume
is
small.
Along
with
high­
emitting
vehicles,
and
spark
ignition
vehicle
particulate
matter,
emission
data
for
air
toxics
is
a
major
emission
data
gap
that
this
information
collection
is
to
address
One
major
source
for
vehicle
in­
use
emission
and
activity
data
is
the
Coordinating
Research
Council
(
CRC)
Real
World
Highway
program.
This
organization
does
test
inuse
vehicles
with
the
view
to
using
a
suitable
sample.
Much
of
CRC's
data
has
been
incorporated
in
Mobile
6
and
has
directed
recent
EFP
research.
EFP
is,
in
fact,
partnering
with
CRC
to
develop
and
manage
test
programs
as
they
apply
to
current
and
future
EPA
emission
data
needs.
This
is
being
done
to
pool
and
leverage
the
resources
needed
to
carry
out
large
national
emission
programs.
Furthermore,
CRC
and
EPA
have
had
their
differences
on
how
best
to
select
vehicles
for
testing.
EPA
has
favored
sequential
procurement
of
a
stratified
random
sample
where
CRC
tended
to
use
targeted
selection
of
certain
vehicle
types,
especially
high­
emitting
vehicles.
The
proposed
vehicle
recruitment
methods
for
this
information
collection
are
the
product
of
the
EFP
and
CRC
collaboration.
The
result
shall
be
a
more
efficient
and
representative
sample
of
real
world
vehicle
emissions.

The
first
and
major
test
program
covered
by
this
information
collection
shall
be
done
in
concert
with
CRC.
That
program
will
provide
emission
data
from
a
randomly
4The
subjects
and
agendas
of
the
various
workshops
can
be
found
at
http://
www.
epa.
gov/
otaq/
model/
mobile6/
m6wkshop.
htm
7
selected
representative
sample
of
gasoline
powered
vehicles
for
all
the
gaseous
emissions
covered
in
this
ICR,
particulate
matter,
and
the
major
air
toxics.

Similar
information
is
collected
in
the
Recall
program
also
conducted
by
OTAQ.
However,
in
that
program,
data
are
collected
only
from
properly
maintained
recent
model
year
vehicles
that
are
likely
to
be
still
under
warranty.
The
vehicle
sample
is
not
meant
to
be
representative
of
the
vehicle
fleet.
The
program's
intent
is
to
enforce
the
warranty
that
is
part
of
a
Federal
emission
standard.
The
testing
is
confined
to
the
Federal
Test
Procedure
used
for
the
certification
of
new
vehicles
and
engines.
No
other
testing
or
sampling
were
done
and
no
second­
by­
second
or
air
toxics
emissions
data
were
collected.
The
vehicle
sample
is
confined
to
recent
model
years
that
are
likely
to
be
still
under
the
emission
equipment
warranty
and
is
tightly
targeted
at
vehicles
suspected
of
not
meeting
the
in­
use
standard
.
EFP
requires
a
sample
of
vehicles
that
represent
all
maintenance
conditions
and
model
years.
EFP
requires
in
the
least
a
test
cycle
that
is
more
severe
in
its
operation
than
the
FTP
to
capture
the
larger
scope
of
emission
rates
vehicles
are
capable
of
generating
than
that
found
in
that
test
procedure
and
drive
cycle.

The
California
Air
Resources
Board
(
ARB)
conducts
an
emission
test
program
similar
to
the
Emission
Factor
Program.
EPA
accepts
the
ARB
data
as
representative
of
California
vehicles.
However,
California
vehicles
are
subject
to
their
own
unique
emission
standards.
Therefore,
data
collected
from
California
vehicles
are
not
representative
of
vehicles
in
the
other
49
states.

3(
b)
Public
Notice
Required
Prior
to
ICR
Submission
to
OMB
The
notice
that
this
ICR
is
up
for
renewal
and
soliciting
comment
was
published
in
the
Federal
Register
on
January
24,
2003
(
FR
Vol.
68,
No.
16
pp.
3524­
3526).
A
copy
of
the
notice
is
attached
here.
There
was
a
single
request
for
information
concerning
the
ICR.
All
available
planning
and
draft
documentation
was
forwarded
to
the
interested
party.

3(
c)
Consultations
Opportunities
for
public
comment
on
the
EFP
and
MOBILE6
models
are
provided
during
public
workshops.
The
workshops
are
held
periodically
to
present
results
of
EPA
analysis,
status
of
current
test
programs,
and
public
discussion
of
MOBILE6
model
issues.
A
list
of
former
workshops
is4;

Mobile6
Workshop
#
4
November
5­
7,
2002
8
Mobile6
Workshop
#
3
June
29­
30,
1999
Mobile6
Workshop
#
2
October
1­
2,
1997
Mobile6
Workshop
#
1
March
19­
20,
1997
Furthermore,
EPA
data
gathering
and
modeling
efforts
are
presented
at
the
annual
CRC
On­
Road
Vehicle
Emission
Workshop,
of
which
EPA
is
a
contributing
member.

Historically
there
have
been
three
significant
reviews
of
the
EFP
data
gathering
process
above
and
beyond
this
information
collection
review
process.
In
1982
the
University
of
Michigan's
Survey
Research
Center
(
SRC)
conducted
a
study,
under
EPA
contract,
to
identify
areas
where
sampling
methodology
could
be
improved.
The
work,
entitled
Evaluation
of
the
Survey
Methods
Used
in
the
Emission
Factors
Program
(
1982),
was
undertaken
under
the
recommendation
of
EPA's
Information
Management
Branch
to
improve
overall
response
rates
from
versions
of
the
EFP
program.
EFP
followed
guidance
and
recommendations
made
by
SRC
and
positive
response
rates
and
record
keeping
improved.
The
increased
response
is
attributed
to
the
increased
frequency
and
degree
of
resources
applied
to
successive
efforts
to
contact
sequentially
selected,
recruited,
or
eliminated
potential
participants.

Viking
Energy
Corporation
(
McLean,
Virginia)
conducted
a
review
of
the
EFP
under
contract
to
EPA's
Office
of
Policy,
Planning
and
Evaluation
(
OPPE).
The
report
entitled,
Review
of
the
Mobile
Sources
Emission
Factors
Program
was
completed
in
February
1986.
The
study
reviewed
the
purpose
of
the
MOBILE
models,
the
usefulness
of
that
purpose,
how
well
the
models
accomplish
their
goal,
and
whether
the
model
results
can
be
improved.
The
findings
of
the
study
were
listed
in
1989'
s
ICR.
As
a
result
of
the
Viking
Study
recommendations,
EPA
reviewed
and
changed
the
linearity
assumption
used
in
the
MOBILE
model
and
have
more
complete
documentation
of
MOBILE
than
was
true
for
past
models.

Following
1993'
s
ICR
submission,
EPA
was
requested
by
the
Office
of
Management
and
Budget
to:

"
conduct
additional
research
to
measure
and
correct
any
biases
arising
from
selection
artifacts.
In
addition,
EPA
should
examine
the
magnitude
of
the
effect
these
potential
biases
may
have
on
EPA
modeling
results.
The
results
of
this
research
and
any
corrective
action
taken
or
proposed
should
be
described
fully
in
the
next
request
for
approval
of
this
collection."

EFP's
response
was
to
increase
the
rate
of
positive
responses
to
its
recruiting
efforts.
EFP
began
recruitment
of
in­
use
vehicles
at
I/
M
facilities
in
Hammond,
IN
(
pre­
1996),
Dayton,
Ohio
(
1997­
98)
and
Mesa,
Arizona
(
1997­
2000).
The
effect
of
that
5Carlson,
T.
R.,
et
al.,
"
Development
of
Speed
Correction
Cycles,"
MOBILE6
Stakeholder
Review
Document
(
M6.
SPD.
001).
Prepared
for
EPA
by
Sierra
Research,
Inc.,
1997.

6P
7­
12,
Modeling
Mobile­
Source
Emissions,
National
Research
Council,
2000,
National
Academy
Press
9
activity
was
a
much
larger
percentage
of
positive
responders.
An
added
benefit
to
that
process
was
that
all
participants
had
a
laboratory
test,
minimally
a
Federal
Test
Procedure
(
FTP),
and
a
state
inspection
test,
typically
an
I/
M240
a
truncated
and
less
rigorous
emission
test
based
upon
a
hot
running
version
of
the
third
phase
of
the
Federal
FTP.
These
paired
tests
were
then
used
to
establish
correlations
with
the
emission
factors
developed
with
the
FTP
with
the
much
larger
data
set
represented
by
the
I/
M240
performed
at
the
state
I/
M
lane.
This
leveraging
of
the
laboratory
data
with
the
population
subject
to
state
inspections
resulted
in
higher
predicted
emission
factors
with
MOBILE6
than
with
MOBILE.
This
is
believed
due
to
correcting
laboratory
tests'
underpredictions
caused
by
a
`
good
volunteer
effect'.

EFP's
second
response
to
the
Viking
Energy
report
was
to
address
a
selection
bias
due
to
the
use
of
only
the
FTP
for
vehicle
testing.
The
FTP
is
a
procedure
developed
for
new
vehicle
certification
and
does
not
represent
the
universe
of
loads,
speeds,
and
ambient
conditions
that
in­
use
vehicles
are
exercised
in.
EFP,
therefore,
developed
from
chase
car
data
collected
under
a
large
variety
of
conditions,
other
driving
schedules
that
were
used
to
test
vehicles
in
different
and
often
more
severe
driving
conditions
than
the
FTP
5.
This
resulted
in
more
accurate
traffic
and
speed­
based
corrections
for
MOBILE
and
better
precision
in
its
predictions.

In
1998
Congress
asked
the
National
Research
Council
to
review
and
evaluate
MOBILE,
its
inputs,
assumptions,
structure
and
predictive
accuracy.
The
review
resulted
in
the
report
"
Modeling
Mobile
Source
Emissions"
published
in
2000.
Members
of
the
committee,
their
affiliations
and
their
recommendations
are
found
in
the
publication.
The
recommendations
that
directly
affect
this
information
collection
are6;

1.
"
EPA
should
develop
a
program
to
enable
more
accurate
determination
of
in­
use
emissions"

2.
"
EPA
should
design
and
undertake
a
large­
scale
testing
program
that
will
better
assess
real­
world
emissions
from
heavy­
duty
vehicles."

3.
"
EPA
should
promptly
update
PART5
with
the
best
available
data
on
PM
emissions
and
incorporate
it
into
a
subsequent
revision
of
MOBILE6."

4.
"
EPA
should
begin
a
substantial
research
effort
to
characterize
high
exhaust
and
7On­
Road
Emissions
Testing
of
18
Tier
1
Passenger
Cars
and
17
Diesel
Powered
Public
Transport
Busses,
Final
Report,
V
1.4,
Carl
Ensfield,
Sensors,
Inc.
October
22,
2002
8"
Characterizing
Exhaust
Emissions
form
Light­
Duty
Gasoline
Vehicles
in
the
Kansas
City
Metropolitan
Area",
Statement
of
Work,
March,
2003
10
evaporative
emitting
vehicles."

5.
"
The
best
available
data
should
be
used
to
update
MOBTOX,
which
should
be
merged
into
MOBILE6."

EFP's
response
to
the
first
recommendation
is
to
use
better
sampling
methodology
for
vehicles
and
testing.
Programs
under
this
information
collection
shall
use
a
stratified
random
sample
instead
of
the
more
traditional
random
sample
used
under
the
last
information
collection.
EFP
also
will
use
in­
situ
sampling
methods
for
all
vehicle
types
covered
by
this
information
collection.
This
will
provide
real
world
emission
data
free
from
the
artifacts
of
laboratory
testing.
The
on­
onboard
instrumentation
that
shall
perform
this
work
is
currently
available
to
EFP.
It
correlates
well
with
current
laboratory
methods
7
and
is
sufficiently
rugged
to
be
used
in
the
field
for
on­
highway
vehicles.

EFP
shall
also
collect
emission
data
using
remote
sensing
devices
(
RSD).
Though
the
precision
and
specificity
of
RSD
data
is
controversial
it
does
have
the
advantage
of
identifying
gross
emitters,
acquire
very
large
samples,
and
is
free
from
recruitment
issues.
The
use
of
RSD
shall
be
part
of
at
least
one
joint
EFP
and
CRC
test
program
in
2003.

The
second
recommendation
is
currently
being
addressed
by
a
large
test
program
started
in
late
2002
and
to
be
finished
in
2003
,
CRC
test
program
E­
55
Heavy
Duty
Vehicle
Testing
For
Emission
Inventory
which
produced
emission
data
for
75
in­
use
heavy
duty
diesel
vehicles.
Furthermore,
EPA
has
obtained
from
West
Virginia
University
emission
data
on
`
x'
number
of
in­
use
heavy
duty
vehicles
collected
in
several
regions
across
the
country
using
a
portable
testing
facilities.
Emission
and
activity
sampling
of
heavy
duty
vehicles
shall
be
done
both
EPA's
Ann
Arbor
test
facility.

EFP's
response
to
the
NRC's
recommendation
of
improving
EPA's
estimate
of
particulate
emissions,
as
a
member
of
CRC,
is
to
develop
a
test
program
to
obtain
new
particulate
emission
estimates
from
a
representative
sample
of
480
in­
use
gasoline
fueled
vehicles
from
the
Kansas
City
Metropolitan
Area
(
KCMA)
8.
This
program
is
designed
to
address
those
issues
that
past
EPA
and
CRC
test
programs
have
been
criticized
for,
recruitment
bias
due
to
recruitment
from
an
I/
M
area,
poor
rates
of
participation
in
non­
I/
M
areas,
the
lack
of
cold
start
emission
data,
and
the
selection
of
the
most
appropriate
11
drive
cycle
for
the
testing.

The
recruitment
and
testing
of
high­
emitting
vehicles,
both
exhaust
and
evaporative,
is
a
primary
goal
for
all
EFP
test
programs
and
is
in
direct
response
to
the
NRC
report.
While
sample
size
is
estimated
by
current
measurement
uncertainties
within
each
stratum
and
EFP
precision
requirements,
participants
are
selected
at
random
from
state
registration
lists.
Participants
shall
each
initially
have
their
vehicle
screened
by
either
a
remote
sensing
device,
tail
pipe
shop
analyzer,
or
onboard
instrumentations
to
determine
if
its
exhaust
emissions
are
likely
to
be
considerably
greater
than
is
considered
normal
for
its
stratum.
The
vehicles
shall
also
be
inspected
for
liquid
leaking
fuel
systems
to
determine
the
vehicle's
likelihood
of
being
an
evaporative
emission
high­
emitter.
Vehicles
that
are
identified
has
being
high­
emitters
shall
be
selectively
recruited
into
either
an
appropriate
sampling
or
a
testing
program
to
have
their
emission
levels
determined.
The
belief
is
that
these
vehicles
contribute
greatly
to
a
stratum's
uncertainty.
If
that
is
the
case
EFP
shall
increase
a
stratum's
sample
size
to
increase
the
certainty
in
the
stratum's
modeled
emission
rates.
Furthermore,
as
these
high­
emitting
vehicles
are
identified,
sufficient
demographic
and
vehicle
data
may
be
available
to
produce
new
strata
for
these
vehicles
apart
from
the
emission
standard
driven
strata
used
for
this
information
collection.

Air
toxics
measurements
are
have
become
part
of
all
present
and
future
EFP
emission
testing
programs.
When
the
capability
becomes
available
air
toxics
will
also
become
part
of
the
in­
use
sampling
programs
that
are
part
of
this
information
collection.

3(
d)
Effects
of
Less
Frequent
Collection
This
information
collection
does
not
require
periodic
reporting
or
record
keeping.
Information
is
gathered
on
occasion.
Each
individual
is
contacted
and
asked
to
lend
his/
her
vehicle
if
the
program
involves
vehicle
testing.
If
the
participants
are
being
asked
to
participate
in
a
sampling
plan
they
will
asked
to
fill
out
a
vehicle
operation
log
which
they
will
provide
to
EFP
at
the
end
of
the
sampling
period.

3(
e)
General
Guidelines
Participation
in
either
a
test
of
sample
program
by
each
owner
is
on
a
voluntary
basis.
Vehicle
owners
do
not
have
to
plan
or
retain
any
records
or
information
if
they
are
involved
in
a
testing
program.
If
the
owner/
operator
is
to
participate
in
a
sampling
program,
however,
he/
she
will
be
asked
to
keep
a
log
of
certain
events
and
actions
that
affect
the
vehicle's
emissions
during
the
sampling
period.
Those
events
and
actions
shall
be
immediately
apparent
to
the
participant
and
are
not
subject
to
prospective
or
perspective
judgment.
Once
the
sampling
period
is
over
and
the
onboard
board
equipment
is
removed
from
the
vehicle,
the
log
is
collected
from
the
owner.
The
participant
has
ended
the
need
to
record
any
information.
12
3(
f)
Confidentiality
The
only
personal
information
collected
in
this
information
collection
is
the
name,
address
and
phone
number
of
the
owner
of
individual
vehicles.
This
information
is
not
used
for
any
purpose
other
than
contacting
the
owner
schedule
the
vehicle
for
testing
or
sampling.
This
information
is
not
stored
in
the
database,
but
on
CD
and
in
data
packets
which
are
kept
in
secured
files.
The
operation
log
used
for
vehicle
sample
will
not
record
locations
only
activities
and
events
that
affect
vehicle
emissions,
vehicle
fueling,
number
of
passengers,
vehicle
payload,
and
vehicle
maintenance.

Vehicle
registration
information
is
received
from
either
the
State
or
from
a
contractor
who
has
agreements
with
States
on
maintaining
the
confidentiality
of
vehicle
owners.
This
information
is
received
on
a
CD
or
computer
tape
and
used
by
EPA
personnel
and/
or
contractor
to
randomly
screen
vehicles
to
develop
mailings.
The
CD
is
locked
in
a
secured
filing
cabinet
with
other
paper
files.
The
computer
is
password
secured
and
known
only
by
the
Project
Officer
and
contractor
personnel
doing
the
screening.
Both
the
computer
files
and
paper
packets
which
includes
the
personal
information
are
destroyed
in
3
years.

Contractors
(
as
listed
in
part
B
1(
c))
are
involved
in
contacting
the
owner
for
participation
in
EFP.
They
will
instruct
the
owners
in
how
to
fill
out
the
vehicle
logs
for
those
who
participate
in
the
sampling
program.
These
contractors
are
monitored
by
two
project
officers.
All
the
contracts
that
deal
with
vehicle
recruitment
have
confidentiality
clauses
that
the
contractors
must
adhere
to.

3(
g)
Sensitive
Questions
The
only
personal
information
involved
in
this
information
collection
is
the
name,
address
and
phone
number
of
the
owner
of
individual
vehicles.
All
that
information
shall
be
kept
in
secured
paper
files.
The
vehicle
information
number
of
the
vehicle
(
VIN)
shall
be
kept
in
MSOD,
but
in
a
non­
public,
password
protected
table
that
shall
not
be
published
to
general
and
non­
EFP
EPA
users
of
the
information.
The
purpose
of
this
information
is
to
request
answers
to
the
participated
in
either
an
emission
and
activity
sampling
or
emission
testing
program
schedule
vehicle
testing.
None
of
the
vehicle
log
entries
are
of
a
sensitive
nature
and
the
log
does
collect
any
data
that
can
be
traced
directly
to
the
owner
or
the
vehicle
operator.

4.
The
Respondents
and
the
Information
Requested
4(
a)
Respondents/
SIC
Codes
Respondents
to
the
EFP
are
members
of
the
general
public
that
own
motor
vehicles.
While
some
businesses
may
be
asked
to
participate
because
they
have
a
certain
13
type
of
vehicle,
only
the
registered
owner
of
the
vehicle
or
their
representative
will
be
asked
to
respond
to
a
solicitation.
There
are
no
SIC
codes
that
correspond
to
an
individual
vehicle
owner.

4(
b)
Information
Requested
4(
b)(
i)
Data
Items,
Including
Record
keeping
Requirements
Any
vehicle
owner's
response
to
a
solicitation
to
participate
in
the
program
is
on
a
voluntary
basis.
Vehicle
owners
do
not
have
to
plan,
generate,
or
retain
any
records
or
information
if
they
are
participants
in
a
testing
program.
If
they
are
participants
in
a
sampling
program
they
will
be
asked
to
keep
a
vehicle
operating
log
for
as
long
as
the
vehicle
is
being
samples.
Exhibit
A
is
a
copy
of
that
log.

4(
b)(
ii)
Respondent
Activities
EPA
shall
solicit
the
general
public
for
participation
in
Emission
Factor
Program
(
EFP)
through
the
random
selection
of
vehicle
owners
using
State
motor
vehicle
owner
registration
lists
(
may
be
derived
from
several
sources).
Potential
participants
have
no
obligation
or
burden
other
than
responding
than
the
keeping
of
the
vehicle
log
if
they
are
participants
in
a
sampling
program.
Participation
in
the
program
by
each
owner
is
on
a
voluntary
basis.
When
EPA
or
its
contractor
returns
a
loaned
vehicle
or
the
sampling
of
the
vehicle
is
completed,
the
owner's
role
in
the
program
ends.
Positive
respondent
activities
in
the
program
are
to:
read
materials
or
discuss
the
program
with
an
EPA
representative,
keep
a
vehicle
log
if
they
are
participants
in
a
sampling
program
questionnaire,
make
the
vehicle
available
for
testing,
and
return
any
loaner
vehicle
to
EPA.

Initial
contact
with
individuals
identified
on
a
State
vehicle
owner
registration
list
is
by
means
of
introductory
letters
(
Exhibit
B.)
Respondents
reply
by
returning
a
preprinted
card
which
indicates
their
willingness
to
participate
in
the
program
(
Exhibit
C).
If
there
is
no
response
to
the
initial
mailing,
a
follow­
up
letter
(
Exhibit
D)
is
sent,
and
attempts
are
made
to
contact
the
potential
participants
by
telephone.
Failing
this,
any
further
attempts
to
solicit
the
individual
are
halted.
In
any
case,
solicitation
of
an
individual
stops
immediately
if
there
is
unwillingness
to
participate
in
the
program.

A
positive
response
may
result
in
an
additional
contact.
At
this
time,
vehicle
descriptive
questions
(
e.
g.,
length
of
vehicle,
air
conditioning,
etc.)
are
asked
either
over
the
phone
or
in
person,
in
accordance
with
the
participant's
wishes.

Willing
participants
deliver
their
vehicles
to
the
testing
facility
or
where
the
sampling
equipment
is
to
be
installed,
or,
at
their
preference
arrange
for
program
representatives
to
pick
up.
While
the
vehicle
undergoes
emissions
testing
by
EPA
(
a
period
of
less
than
one
week),
the
owner
has
the
option
of
using
a
loaner
vehicle
plus
14
receiving
a
small
compensation
of
$
200
or,
if
they
choose
not
to
use
a
loaner
vehicle,
they
will
receive
a
higher
compensation,
$
275
day
for
the
use
of
their
vehicle.
If
the
participant
is
in
a
sampling
program
they
shall
receive
an
additional
$
50.
Their
vehicle
shall
also
be
washed
and
the
tank
filled
with
fuel.

The
solicitation
process
for
vehicles
that
arrive
at
state
inspection
lanes
is
a
twostep
process.
First,
qualifying
vehicles
entering
the
lane
are
identified
and
approached
by
EPA
representatives.
Vehicle
owners
are
then
asked
if
they
would
like
to
participate
in
the
EFP
and
a
cash
incentive
(
up
to
$
250/
week)
and
a
loaner
vehicle
are
offered
to
the
vehicle
owner.
If
the
owner
agrees
to
participate
in
the
program,
he/
she
is
asked
vehicle
descriptive
questions
(
e.
g.,
length
of
vehicle,
air
conditioning,
etc.).
Generally
the
participant
exchanges
their
vehicle
for
a
loaner
vehicle
while
they
are
at
the
inspection
lane.
The
vehicle
is
then
taken
to
the
test
lab
to
undergo
testing
by
the
EPA
contractor
over
a
one
week
period.
A
count
of
all
vehicle
owners
responses
is
kept,
as
well
as,
the
number
of
vehicles
using
the
emission
lanes.
(
part
B,
section
3
provides
some
more
detail
on
the
purpose
and
analysis
methodology
for
the
inspection
lane
type
of
program.)

5.
The
Information
Collected:
Agency
Activities,
Collection
Methodology,
and
Information
Management
5(
a)
Agency
Activities
The
EFP
is
required
to
identify
and
quantify
the
effects
of
changing
trends
in
vehicle
populations
and
use
that
affect
emission
inventories.
The
methods
of
determining
the
effects
on
inventories
are
developed
following,
in
order,
the
first
suitable
method.

°
EFP
first
determines
if
the
data
can
be
developed
by
using
theory
and
engineering
principles.
This
applies
to
phenomena
based
on
physical
laws,
such
as
the
effect
of
vehicle
mass
on
fuel
economy.
This
method
does
not
require
this
information
collection
request.

°
The
second
tier
of
data
collection
is
that
obtained
from
past
empirical
studies.
This
is
data
previously
collected
by
EPA,
vehicle
and
engine
manufacturers,
academia,
and
other
investigating
organizations.
For
some
segments
of
the
vehicle
fleet
there
is
a
great
deal
of
historical
data
that
can
be
and
is
exploited
by
EFP
to
provide
data
to
update
and
develop
EPA's
emission
models.
This
does
not
require
this
information
collection
request.

°
The
third
possible
source
of
information
is
to
develop
emission
data
based
on
the
laboratory
testing
of
leased
vehicles
or
engines
at
EPA
of
EPA
contractor
test
facilities.
This
type
of
work
is
applicable
when
one
is
determining
the
effects
of
specific
factors
where
the
vehicle's
membership
in
the
in­
use
fleet
is
not
desirable
or
necessary.
This
research,
generally
called
A
to
B
comparisons,
are
done
to
15
isolate
the
effect
of
one
or
two
variables
under
carefully
controlled
conditions.
This
type
of
information
collection
is
not
subject
to
information
collection
request.

°
The
fourth
method,
requiring
this
ICR,
is
to
develop
new
data
from
in
situ
sampling
of
in­
use
vehicle
emission
and
activity.
This
is
relatively
new
to
EFP,
but
has
the
advantage
of
collecting
real
world
data
as
opposed
to
laboratory
emission
testing
which
is
the
measurement
of
vehicle
emissions
based
on
a
simulation
of
prescribed
operating
conditions.
This
method
is
ideal
for
vehicle
activity
data
collection.
However,
there
are
limitations
to
sampling;
not
all
pollutants
can
be
measured
in
the
field.

°
The
last
method
of
data
collection
and
also
subject
to
this
ICR,
is
to
measure
vehicle
emissions
by
the
testing
of
in­
use
vehicles
in
a
laboratory
setting
using
facilities
and
protocols
to
imitate
a
variety
of
traffic,
operating,
and
road
conditions.
This
has
been
the
traditional
track
for
the
data
collected
under
this
ICR.

The
methods
selected
are
built
around
issues
that
data
is
to
address.
In
the
case
that
in­
use
emission
data
is
needed
EFP
shall
identify
a
population
of
vehicles
to
be
sampled,
a
sample
size,
an
area
or
region
in
which
the
work
is
to
be
done,
the
emissions
to
be
measured,
a
method
in
which
the
data
is
to
be
collected,
and
resources
needed
to
complete
the
work.
In
general
EFP
programs
subject
to
this
ICR
will
be
both
sampling
and
testing
programs
performed
concurrently.

5(
b)
Collection
Methodology
and
Management
All
phases
of
the
EFP,
including
vehicle
procurement,
vehicle
emission
and
activity
sampling,
emission
testing,
and
data
analysis,
are
guided
by
quality
assurance
plans.

The
development
of
a
statement
work
for
all
aspects
of
the
work
is
the
first
step.
Current
EFP
programs
involve
consultation
with
stakeholders
such
as
CRC
whose
members
are
from
the
vehicle
manufacturers,
academic
organizations,
the
US
Department
of
Energy,
US
Department
of
Transportation,
and
other
offices
within
EPA.
This
level
of
consultation
is
required
because
CRC
members
are
providing
considerable
funding
to
the
project,
have
fundamental
interest
that
the
data
generated
by
EFP
programs
is
rigorous
and
useful
to
all
members,
and
have
the
necessary
expertise
to
design
and
manage
these
type
of
programs.

The
statement
of
work
is
required
to
meet
basic
quality
assurance
requirements
from
the
agency's
contracts
rules.
EPA
and
EPA
contractors
mus
document
all
aspects
of
any
testing
program
such
that
the
data
meets
minimal
QA
requirements
for
accuracy
and
precision.
The
SOW
therefore
relies
on
prescribed
measurement
procedures
that
are
specified
in
the
Code
of
Federal
Register,
the
use
of
consensus
test
methods
set
by
the
16
Association
Standards
for
Testing
of
Materials
Participants
and
their
vehicles
shall
be
recruited
randomly
from
the
a
stratified
sample
from
either
an
I/
M
lane
or
state
vehicle
owner's
list.
Specific
procurement
criteria
and
quality
assurance
procedures
for
their
implementation
are
the
subject
of
each
work
assignment.
However,
in
all
cases
sample
size
for
each
stratum
shall
be
based
on
the
precision
requirement
of
this
information
collection
and
uncertainty
in
the
emission
measurement
for
each
stratum.
This
is
to
assure
that
an
adequate
and
efficient
sample
is
used
for
all
EFP
test
programs.

Vehicle
sampling
and
testing
shall
use
quality
assurance
procedures
described
in
detail
in
CFR
86
subpart
B,
C,
D,
M,
N,
and
O.
In
the
case
of
vehicle
sampling
where
new
methods
are
to
be
used
or
developed,
extensive
lab
testing
to
sampling
instrumentation
correlation
shall
be
done
as
part
of
the
sampling
program.
This
shall
be
done
to
establish
a
precision
and
accuracy
statements
associated
with
the
data
the
program
produces.

Specific
test
requirements
and
quality
assurance
steps
are
outlined
in
the
individual
work
assignments.
Furthermore,
all
significant
changes
in
the
work
are
documented
and
reviewed
by
stake
holders
before
their
execution.

Data
to
be
produced
by
all
EFP
programs
are
those
specified
in
its
statement
of
work.
Examples
of
and
the
specified
format
for
the
information
collected
for
sampling
and
testing
programs
is
found
in
Exhibit
E.
The
data,
once
placed
into
the
specified
formats,
is
reviewed
by
the
testing
or
sampling
contractor,
delivered
to
EFP
staff,
where
the
data
is
reviewed
by
the
principal
investigator,
then
processed
through
program­
specific
preprocessing
and
quality
assurance
programs
before
it
is
loaded
into
the
database
MSOD.
The
data
checking
programs,
loading
programs,
and
the
design
of
MSOD
are
being
reviewed
and
subject
to
continuous
improvement
to
increase
data
quality
and
allow
for
new
data
types
to
be
stored.

The
following
type
and
format
for
the
information
collected
for
sampling
programs
is
found
in
Exhibit
E
and
that
collected
for
emission
testing
programs
in
Exhibit
F.
The
data,
once
placed
into
the
specified
formats,
is
reviewed
by
the
testing
or
sampling
contractor,
delivered
to
EFP
staff
where
the
data
is
reviewed
by
the
principal
investigator,
then
processed
through
program­
specific
preprocessing
and
quality
assurance
programs
before
it
is
loaded
into
the
database
MSOD.
Once
in
the
format
is
subject
to
review
by
database
users
and
ASD
modeling
staff
and
contractors.

The
vehicle
owner
may
receive
his
vehicle's
sampled
or
tested
emission
results
upon
request.
EFP
staff
will
help
the
owner
interpret
the
results
and
answer
any
questions
concerning
them
or
the
program
in
which
they
participated.
Access
to
the
database,
MSOD
is
available
to
the
public
upon
request.
The
only
section
of
the
database
that
is
17
confidential
is
a
table
of
Vehicle
Identification
Numbers
(
VINS).
There
is
no
names
or
addresses
of
participants
in
the
MSOD.
Auto
manufacturers,
the
oil
industry,
and
other
pollution
control
agencies
including
state
and
local
governments
are
the
main
users
of
the
database.

The
results
of
most
of
EPA's
analyses
of
the
data
are
published
in
the
form
of
updates
to
the
current
emission
model
MOBILE6
and
as
input
to
the
MOVES.
The
data
is
also
used
in
the
document
AP­
42,
the
compilation
of
mobile
source
emission
estimates,
or
various
technical
reports
on
specific
analyses.
Further,
EPA
holds
public
workshops
periodically
to
present
the
results
of
analyses
and
their
potential
impact
on
the
EPA
emission
models.

5(
c)
Small
Entity
Flexibility
The
information
collection
does
not
involve
small
businesses
or
small
entities.
The
information
collection
solicits
individual
vehicle
owners.
When
selecting
vehicles
from
vehicle
registration
lists,
EFP
shall
limit
mailings
of
materials
to
owners
of
vehicles
targeted
for
testing.
In
addition,
the
vehicle
log
is
only
requested
from
active
participants
in
EFP
emission
and
activity
sampling
programs.

5(
d)
Collection
Schedule
The
EFP
is
an
ongoing
program.
The
current
schedule
is
being
set
to
address
issues
brought
to
EFP
from
the
report
"
Modeling
Mobile
Source
Emissions"
by
the
NRC
and
the
development
the
model
MOVES.
The
immediate
EFP
projects
are
to
address
particulate
emissions,
high­
emitting
vehicles,
heavy
duty
vehicles,
and
air
toxins
with
the
most
current
and
accurate
emission
and
activity
data
possible.
The
EPA
emission
models
must
be
revised
periodically
to
reflect
the
changing
current
mix
of
vehicle
types,
technologies,
and
operating
conditions.

6.
Estimating
the
Burden
and
Cost
of
the
Collection
6(
a)
Estimating
Vehicle
Owner
(
Respondent)
Burden
and
Inquiry
Burden
The
attached
materials
represent
those
items
required
to
be
filled
in
by
the
vehicle
owners.
First,
for
the
traditional
EFP,
a
random
selection
of
owners
are
mailed
an
introductory
letter
(
Exhibit
B)
which
includes
a
question
and
answer
sheet
that
is
mailed
back
to
us
in
a
postage
paid
envelop.
This
"
package"
is
mailed
only
to
those
owners,
selected
from
State
motor
vehicle
registration
lists,
who
have
registered
vehicles
of
interest
to
the
program.
If
there
is
no
response
to
the
first
package
a
follow­
up
letter
(
Exhibit
D)
is
sent.

For
individuals
solicited
from
an
inspection
lane,
an
EPA
representative
explains
18
the
program
and
asks
the
vehicle
owner
to
participate.
This
process
replaces
the
steps
outlined
for
contacting
vehicle
owners
by
mail.

In
either
case,
when
the
participant
is
part
of
a
sampling
program
they
shall
be
asked
to
enter
data
into
the
vehicle
operation
log,
an
example
of
which
follows
below.
program
and
asks
the
vehicle
owner
to
participate.
This
process
replaces
the
steps
outlined
for
contacting
vehicle
owners
by
mail.

In
either
case,
when
the
participant
is
part
of
a
sampling
program
they
shall
be
asked
to
enter
data
into
vehicle
operation
log,
an
example
of
which
(
Exhibit
A
­
Vehicle
Log)
is
shown
below.
19
Item
Key
On
Date
Key
On
Time
Key
Off
Date
Key
Off
Time
Check
Box
for
Fuel
Level
Number
of
Passengers
>
16
years
Number
of
Passengers
11
­
16
years
old
Number
of
Passengers
1
­
10
years
old
Trip
Type
Payload
Units
Date
Time
Date
Time
1/
4
1/
2
3/
4
Full
Number
Number
Number
Commute
Shopping
Errands
Items
in
Vehicle
Example
2­
Apr
11:
50
AM
2­
Apr
12:
20
P
M
1/
4
1/
2
3/
4
x
Full
1
2
Commute
Shopping
x
Errands
G
ro
ce
rie
s
Example
2­
Apr
3:
10
P
M
2­
Apr
3:
20
P
M
1/
4
1/
2
x
3/
4
Full
1
1
Commute
Shopping
Errands
x
Dry
C
le
aning
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
1/
4
1/
2
3/
4
Full
Commute
Shopping
Errands
Tables
1­
2
list
the
inquiry
burden
hours
for
both
types
of
vehicle
solicitation
utilized
in
the
EFP.
The
burden
hour
estimates
were
determined
based
upon
in­
field
experience
of
both
solicitation
types
and
an
estimate
on
the
time
necessary
make
entrees
into
the
vehicle
log.
The
mileage
figures
of
15
miles
for
the
trip
to
the
test
site
or
inspection
lane
was
calculated
from
past
programs
and
grown
to
account
for
a
measured
trends
of
increasing
commuting
distance.
It
was
assumed
that
25
mph
represents
an
average
speed
in
an
urban
area.
The
trip
mileage
and
speed
were
combined
to
determine
the
length
of
time
the
vehicle
owner
spends
returning
the
loaner
vehicle.
The
estimated
annual
burden
in
hours
and
costs
for
the
two
recruitment
schemes
are
summarized
in
Tables
1
and
2.
The
total
program
annual
burden
in
hours
and
costs
is
summarized
in
Table
3.
20
Table
1
­
Estimating
Hours
Burden
and
Cost
to
Vehicle
Owner
at
I/
M
Lane
6(
a)
and
(
b):
Estimating
Burden
and
Cost
to
Vehicle
Owner
at
I/
M
Lanes
Annual
Labor
Hours
and
Cost
to
Vehicle
Owner
Vehicle
Number
Owner
Owner
of
Contacts/
Total
Costs
Hours
Respondents
Hours
$
31.88
per
hour
Program
Discussion/
Invitation
(
1)
0.25
120
30
$
956
Exchanging
Vehicles
and
Paper
Work
with
Contractor
(
2)
0.5
100
50
$
1,594
Return
Owners
Car
from
Test
Site
(
3)
0.6
100
60
$
1,913
Sampling
Program
Participant
Vehicle
Log
Training
(
4)
0.17
25
6
$
179
Fill
Out
Vehicle
Log
(
4)
0.82
33
27
$
863
Total
(
Hours)
172.7
Total
(
Cost)
$
5,505
(
1)
Total
Approaches
by
Vehicle
Recruiters
(
2)
Owners
are
provided
remuneration
and
free
loaner
vehicle
(
3)
Assumes
an
average
one
way
trip
length
of
15
miles
at
an
average
speed
of
25
mph
(
4)
Assumes
100
positive
responses
from
owners
and
33
owners
will
participate
in
a
sampling
program.
Vehicle
log
is
provided
by
EPA
21
Table
2
­
Estimating
Hours
Burden
and
Cost
to
Vehicle
Owners
from
Registration
List
6(
a)
and
(
b):
Estimating
Burden
and
Cost
to
Vehicle
Owner
from
Registration
Lists
Annual
Labor
and
Cost
to
Vehicle
Owner
Vehicle
Number
Total
Vehicle
Owner
Owner
of
Mailings
Responden
t
Costs
(
4)

Hours
or
Responses
Hours
$
31.88
per
hour
Read
Information
in
Mailing
(
1)
0.03
300
9
$
287
Fill
Out
and
Return
Post
Card
0.083
50
4
$
132
Follow
Up
Phone
Calls
and
Visits
to
Respondents
1
240
240
$
7,651
Exchanging
Vehicles
and
Paper
Work
with
Contractor
0.5
240
120
$
3,826
Drive
To
and
From
Test
Site
(
2)
1.2
240
288
$
9,181
Sampling
Program
Participant
Vehicle
Log
Training
(
3)
0.17
50
9
$
271
Fill
Out
Vehicle
Log
(
3)
0.82
50
41
$
1,307
Total
(
Hours)
710.7
Total
(
Cost)
$
22,656
(
1)
Assumes
1200
mailings
(
2)
Assumes
an
average
one
way
trip
length
of
15
miles
at
an
average
speed
of
25
mph
(
3)
Assumes
240
positive
responses
from
owners
and
50
owners
will
participate
in
a
sampling
program.
(
4)
No
Capital
costs,
EFP
provides
vehicle
log
The
total
annual
respondent
burden
is
reported
below
in
Table
3.

Table
3
­
Total
Respondent
Burden
for
EFP
(
Hours
and
Costs)

Contacted
Participants
Hours
Costs
I/
M
Lane
Recruitment
Totals
120
100
172.7
$
5,505
Registration
List
Recruitment
Totals
300
240
710.7
$
22,656
Totals
420
340
883
$
28,161
Respondents
are
provided
remuneration
for
their
time
and
inconvenience
(
See
part
A
section
4
(
b)
)
and
that
cost
is
therefore
an
Agency
cost.

6(
b)
Estimating
Inquiry
Costs
The
inquiry
collection
costs
are
incorporated
in
Tables
1­
2
listed
above.
The
respondents
are
provided
remuneration
for
their
time
and
inconvenience
(
See
part
A,
section
3
(
b)(
iii))
and
that
cost
is
therefore
an
Agency
cost.
22
6(
c)
Estimating
Agency
Burden
and
Cost
The
Agency
cost
burden
is
determined
by
the
contract
dollars
spent
and
the
salary
paid
to
EPA
personnel
and
contractors.
The
contract
costs
cover
inspection,
repair
and
testing
of
vehicles,
as
well
as,
data
entry,
quality
control,
and
storage.
On
an
annual
average,
EPA
devotes
4
person­
years
to
planning
and
monitoring
the
program,
processing
the
data,
and
design/
maintaining
the
database.
The
salaries
used
for
the
three
classification
was
GS
13
for
project
manager
and
Contracts
Specialist,
GS
9
for
technicians,
and
GS
6
for
clerical
support.
All
pay
categories
were
are
at
maximum
step
for
each
grade
for
FY03
Detroit
area
General
Schedule.
Contractor
costs
are
based
on
current
costs
incurred
in
typical
contractor
supplied
services
to
EPA
in
the
same
geographic
area.
The
estimate
of
agency
costs
are
estimated
is
in
Tables
4
for
work
provided
by
EPA
staffed
,
Table
5
for
contractors,
and
totals
for
both
in
Table
6.
23
Table
4
­
Agency's
Estimated
Hours
and
Costs
Burden
­
Agency
Activities
6(
c)
Estimating
Annual
Agency
Hours
and
Costs
Burden
­
EPA
Staff­
Supplied
Work
Manager
Contracts
Specialist
Technician
Clerical
Capital/
Startup
O
&
M
Labor
Hours
Costs
Hourly
Labor
Costs
per
Task
$
70
$
70
$
41
$
30
Write
Statement
of
Work
(
SOW)
for
Test
Programs
100
100
10
20
$
0
$
0
180
$
15,060
Vehicle
Selection/
Recruitment
(
1)
200
0
0
5
$
0
$
0
205
$
3,663
Expand/
Maintain
Database
20
0
4500
10
$
10,000
$
0
4530
$
207,731
Vehicle
Repair
and
Maintenance
5
0
5
20
$
0
$
0
30
$
2,208
Testing
Vehicles
100
0
200
0
$
0
$
0
300
$
8,501
Quality
Control/
Quality
Assure
Data
100
40
200
10
$
0
$
0
350
$
18,287
Monitoring
Test
Programs
100
0
10
0
$
0
$
0
110
$
7,434
Review
Programs
and
Write
Reports
100
5
10
5
$
0
$
0
30
$
7,935
Instrumenting
Vehicles
for
Sampling
10
0
100
0
$
0
$
0
400
$
4,778
Analyze
Test
Data
300
0
1500
5
$
0
$
0
1505
$
82,355
Hours
Costs
Totals:
7460
$
342,892
(
1)
Assumes
vehicle
selection
by
registration
lists
and
I/
M
lanes
as
specified
in
6(
a)
&
6(
b).
24
Table
5
­
Agency's
Estimated
Hours
and
Costs
Burden
­
Contractor
Activities
6(
c)
Estimating
Annual
Contractors
Hours
and
Costs
Burden
­
Contractor­
Supplied
Work
Manager
Technician
Clerical
Capital/
Startup
O/
M
Labor
Hours
Costs
Hourly
Labor
Costs
$
69
$
42
$
27
Write
Statement
of
Work
(
SOW)
for
Test
Programs
10
10
0
$
0
$
0
20
$
1,108
Vehicle
Selection/
Recruitment
(
1)
300
1000
2500
$
0
$
500
3800
$
130,103
Expand/
Maintain
Database
5
40
5
$
0
$
0
50
$
2,166
Vehicle
Repair
and
Maintenance
100
3000
20
$
0
$
0
3120
$
134,112
Testing
Vehicles
200
9000
0
$
0
$
0
9200
$
393,876
Quality
Control/
Quality
Assure
Data
50
500
300
$
0
$
0
850
$
32,564
Monitoring
Test
Programs
200
10
40
$
0
$
0
250
$
15,207
Review
Programs
and
Write
Reports
50
10
10
$
0
$
0
70
$
4,119
Instrumenting
Vehicles
for
Sampling
100
2000
0
$
0
$
0
2100
$
91,338
Analyze
Test
Data
200
200
0
$
0
$
0
400
$
22,164
Hours
Costs
Totals:
19840
$
825,648
(
1)
Assumes
vehicle
selection
by
registration
lists
and
I/
M
lanes
as
specified
in
6(
a)
&
6(
b).

Table
6
­
Agency's
Estimated
Burden
for
the
Future
­
Totals
6(
c)
Estimating
Annual
Agency
Burden
­
Totals
Hours
Dollars
EPA
Staff
7460
$
342,892
Contractors
19840
$
825,648
Total
Agency
Hours
27300
Total
Agency
Costs
$
1,168,540
6(
d)
Estimating
Vehicle
Owner
and
Inquiry
Universe
and
Total
Burden
Costs
There
is
no
universe
labor
or
burden
costs
associated
with
the
vehicle
owner
or
inquiry
other
than
what
has
been
listed
in
6(
a)
and
6(
b).
The
Agency
cost
for
testing
owner's
vehicles
has
been
listed
in
6(
c).
25
6(
e)
Bottom
Line
Burden
Hours
and
Costs
The
total
burden
hours
and
costs
are
provided
in
Table
7.

Table
7
­
Estimated
Average
Annual
Bottom
Line
Burden
and
Costs
Item
Respondents
Agency
Totals
Estimated
Annual
Burden
in
Hours
883
27,300
28,183
Estimated
Annual
Burden
in
Dollars
$
28,160
$
1,168,540
$
1,196,701
6(
f)
Reasons
for
Change
in
Burden
The
calculated
respondent
burden
per
respondent
has
increased
due
to
the
introduction
of
the
vehicle
log
for
emission
and
activity
sampling
programs,
longer
commutes
to
and
from
test
sites,
and
inflation.
The
overall
burden
however
has
decreased
for
testing
programs
due
to
the
elimination
of
the
questionnaire
since
the
last
submission
of
the
ICR
and
to
budget
reductions
with
a
resulting
decrease
in
the
number
of
vehicles
to
be
tested.
Some
reductions
in
burden
to
the
Agency
have
occurred
because
of
more
automated
data
collection
methods.
All
test
and
field
sampling
data,
save
the
vehicle
log,
are
collected
electronically
from
the
testing
contractors.

6(
g)
Burden
Statement
Burden
Statement:
The
annual
public
reporting
and
recordkeeping
burden
for
this
collection
of
information
is
estimated
to
average
2.6
hours
per
response.
Burden
means
the
total
time,
effort,
or
financial
resources
expended
by
persons
to
generate,
maintain,
retain,
or
disclose
or
provide
information
to
or
for
a
Federal
agency.
This
includes
the
time
needed
to
review
instructions;
develop,
acquire,
install,
and
utilize
technology
and
systems
for
the
purposes
of
collecting,
validating,
and
verifying
information,
processing
and
maintaining
information,
and
disclosing
and
providing
information;
adjust
the
existing
ways
to
comply
with
any
previously
applicable
instructions
and
requirements;
train
personnel
to
be
able
to
respond
to
a
collection
of
information;
search
data
sources;
complete
and
review
the
collection
of
information;
and
transmit
or
otherwise
disclose
the
information.
An
agency
may
not
conduct
or
sponsor,
and
a
person
is
not
required
to
respond
to,
a
collection
of
information
unless
it
displays
a
currently
valid
OMB
control
number.
The
OMB
control
numbers
for
EPA's
regulations
are
listed
in
40
CFR
part
9
and
48
CFR
chapter
15.

To
comment
on
the
Agency's
need
for
this
information,
the
accuracy
of
the
provided
burden
estimates,
and
any
suggested
methods
for
minimizing
respondent
burden,
including
the
use
of
automated
collection
techniques,
EPA
has
established
a
public
docket
for
this
ICR
under
Docket
ID
No.
OAR­
2003­
0006,
which
is
available
for
public
viewing
at
the
Office
of
Air
and
Radiation
Docket
in
the
EPA
Docket
Center
(
EPA/
DC),
EPA
West,
Room
B102,
1301
Constitution
Ave.,
NW,
Washington,
DC.
The
EPA
Docket
Center
Public
26
Reading
Room
is
open
from
8:
30
a.
m.
to
4:
30
p.
m.,
Monday
through
Friday,
excluding
legal
holidays.
The
telephone
number
for
the
Reading
Room
is
(
202)
566­
1744,
and
the
telephone
number
for
the
Office
of
Air
and
Radiation
Docket
is
(
202)
566­
1742).
An
electronic
version
of
the
public
docket
is
available
through
EPA
Dockets
(
EDOCKET)
at
http://
www.
epa.
gov/
edocket.
Use
EDOCKET
to
submit
or
view
public
comments,
access
the
index
listing
of
the
contents
of
the
public
docket,
and
to
access
those
documents
in
the
public
docket
that
are
available
electronically.
Once
in
the
system,
select
"
search,"
then
key
in
the
docket
ID
number
identified
above.
Also,
you
can
send
comments
to
the
Office
of
Information
and
Regulatory
Affairs,
Office
of
Management
and
Budget,
725
17th
Street,
NW,
Washington,
DC
20503,
Attention:
Desk
Office
for
EPA.
Please
include
the
EPA
Docket
ID
No.
OAR­
2003­
0006
and
OMB
control
number
2060­
0078
in
any
correspondence.
27
PART
B
1.
Survey
Objectives,
Key
Variables,
and
Other
Preliminaries
1(
a)
Survey
Objectives
The
primary
objective
of
the
EFP
is
to
obtain
the
national
average
and
variance
of
the
key
variables
of
the
study
for
each
stratum
in
the
sample
frame.
Other
EFP
objectives
are
listed
below:

°
Determine
the
effect
of
vehicle­
specific
attributes
like
technology
and
age
on
emission
and
activity
on
a
second­
by­
second
basis.
These
attributes
make
up
the
source
bins
that
will
be
used
in
the
model
MOVES.

°
Determine
the
effect
of
environmental
conditions
like
temperature,
road
grade,
and
road
conditions
on
emission
and
activity
on
a
second
by
second
basis.

°
Determine
the
effect
of
fuels,
inspection
and
maintenance
programs
on
emission
and
activity
on
a
secondby
second
basis.

1(
b)
Key
Variables
The
key
variables
of
the
information
collection
are
vehicle
emissions
VOC,
PM,
N)
x,
CO
2,
CO,

the
group
of
compounds
known
as
the
air
toxics
compounds,
and
vehicle
activity.
The
emissions
are
those
typically
emitted
from
all
motor
vehicles.
The
variable
activity
are
the
metrics
that
describe
the
frequency,

duration,
temporal,
and
spacial
use
of
a
vehicle.
Activity
measurement
also
includes
the
ambient
conditions
and
vehicle
load
under
which
the
vehicle
is
being
used.
The
emissions
are
expressed
in
emission
factors
on
either
a
gram
per
mile
or
gram
per
second
basis.
Activity
is
expressed
in
miles
per
hour,
time
of
day,
road
grade,

temperature,
barometer,
and
location
on
a
second
be
second
basis.

1(
c)
Statistical
Approach
28
A
statistical
approach
is
required
for
the
Emission
Factor
Program.
Similar
vehicles
do
not
emit
pollution
at
the
same
rate
in
the
same
way
under
all
conditions.
It
is
necessary
to
identify
an
appropriate
sample
of
vehicles,
sample
and
test
them
for
emissions
using
appropriate
sampling
and
testing
protocols,
and
then
distribute
the
emission
results
to
those
vehicles
in
the
fleet
that
the
sample,
the
sampling,
and
testing
protocols
are
meant
to
represent,
since
it
is
necessary
to
test
vehicles
to
measure
the
emission
levels.
Other
sampling
types
do
not
provide
an
accurate
measure
of
emissions
for
an
in­
use
vehicle
fleet.
A
census
would
be
impractical.

Anecdotal
collection
methods
would
not
produce
the
data
needed,
and
vehicle
emission
measurement
is
not
a
common
practice
with
the
general
public.

At
present
there
is
a
single
contractor
doing
this
work
for
EFP:

EG&
G
Automotive
Research
Solicit
vehicle
owners
and
2565
Plymouth
Road
administer
questionnaire
Ann
Arbor,
MI
48105
in
Ann
Arbor,
MI
Headquarters:

5404
Bandera
Road
San
Antonio,
TX
78238
1(
d)
Feasibility
Respondent
obstacles
to
a
successful
EFP
is
sufficient
salience
and
incentives
to
potential
participants
to
assure
participation
rates
of
80%.
A
second
obstacle
is
that
participants
who
are
members
of
an
emission
and
activity
sampling
program
will
be
requested
to
keep
a
log
of
events,
activities,
and
facts
that
directly
affect
vehicle
emissions.

Funding
for
EFP's
Kansas
City
testing
program
in
FY
2003
and
FY2004
is
$
600
k
from
EPA's
Office
of
9State
Motor­
Vehicle
Registration
2000,
R.
L.
Polk
&
Co.,
26955
Northwestern
Highway,
Southfield,
Michigan
48034
29
Transportation
and
Air
Quality
(
EFP's
parent
office),
$
500
K
from
the
EPA's
Office
of
Air
Quality
Planning
and
Standards,
$
400
K
from
the
Department
of
Energy,
$
200
K
from
the
Department
of
Transportation,
and
$
405
K
from
the
membership
of
the
Coordinating
Research
Council.
EFP
has
committed
five
engineers,
two
scientist,

two
contractor
employees,
and
one
man
year
of
contract
management
staff
to
this
program.
The
parent
division
has
delegated
one
center
director
to
supervise
and
direct
the
engineering
and
scientists.
New
sampling
equipment
has
been
purchased
for
the
field
work
at
a
cost
of
over
$
100
K
dollars.
Sufficient
travel
funds
for
2003
have
been
allocated
to
this
program
for
site
visits
to
Kansas
City
and
one
to
the
potential
contractors
for
the
work.

Timing
of
results
for
decision
making
­­
The
results
of
the
EFP
are
used
primarily
in
the
computer
model,
MOBILE6.
The
MOBILE
models
are
usually
updated
every
three
to
five
years
with
the
EFP
data
that
has
been
accumulated.
The
data
from
the
Kansas
City
program
is
projected
to
be
used
minimally
as
a
new
input
to
Mobile
6
and
more
likely
as
a
new
input
into
the
model
MOVES
in
late
2004.

2.
Survey
Design
2(
a)
Target
Population
and
Coverage
The
target
population
for
the
EFP
consists
of
on­
highway
vehicles
registered
for
operation
on
local,
state,

and
federal
road.
The
registration
of
those
vehicles
is
subject
to
the
rules
established
in
state
within
the
United
States.
Motorcycles
shall
be
excluded
as
a
vehicle
emission
class
for
this
information
collection
because
of
their
low
impact
on
the
total
emission
burden.
They
account
for
0.1939
of
the
on­
road
vehicles
and
less
than
1.0
%
of
the
mileage
traveled
and
no
more
than
1.0
percent
of
the
pollutants
from
all
on­
road
vehicles.
Obtaining
in­
use
emission
tests
for
them
is
not
cost
effective.
EFP
has
and
shall
obtain
alternative
data
sources
for
EFP
30
modeling.
These
alternative
sources
shall
be
data
from
research
testing
done
for
regulatory
work
at
EPA
and
from
certification
testing.

The
target
vehicle
emission
classes
that
use
fuels
other
than
gasoline
and
diesel
fuel
shall
be
excluded
from
the
information
collection.
Though
these
vehicles
make
up
an
increasing
number
of
the
fleet
they
still
represent
a
small
fraction
of
the
total
vehicle
fleet
and
contribution
to
the
emission
inventory.
Emission
data
for
these
vehicles
have
been
the
subject
of
research
and
that
data
is
available
to
EFP
to
provide
sufficient
information
upon
which
to
model
their
emissions
in
an
adequate
manner.
Resources
are
best
spent
to
sample
those
vehicles
using
the
most
typical
fuels,
gasoline
and
diesel
fuel.

The
coverage
shall
be
selected
counties
within
the
United
States.
The
number
of
counties
shall
be
greater
than
50,
but
no
more
than
100.
The
counties
shall
be
selected
based
on
their
representativeness
of
the
national
vehicle
registration
for
the
key
variables
and
the
availability
of
testing
facilities.

2(
b)
Sample
Design
2(
b)(
i)
Sampling
frame
In
all
instances
the
sampling
frame
for
EFP
testing
and
sampling
is
a
state
or
some
subset
of
a
state
vehicle
registration
list,
usually
specific
to
the
area
where
the
emission
testing
or
sampling
is
to
take
place.

2(
b)(
ii)
Sample
size
Sample
size
is
determined
by
the
variability
of
the
emission
factor
in
the
EFP
database
MSOD
and
required
precision
of
the
value
desired
for
modeling
a
particulate
emission
for
each
sample
stratum.
The
desired
precision
is
20
%
of
the
average
emission
factor
for
each
stratum,
over
the
course
of
the
data
collection
or
the
Tier
2
BIN
standard,
whichever
is
larger.
This
conditional
precision
is
due
to
very
large
variances
on
a
percentage
basis
for
very
clean
vehicles
which
would
result
in
very
large
sample
sizes
that
would
result
in
very
little
increase
in
the
precision
of
the
aggregate
inventory
for
all
strata
in
the
sample.
Table
A.
1
shows
the
relative
31
contribution
of
the
proposed
sample
strata
to
the
overall
on­
highway
generated
emission
inventory.
32
Vehicle
Strata
Fuel
Strata
Model
Year
Strata
2007
Mobile
VOC
Tons
2007
Mobile
CO
Tons
2007
Mobile
NOx
Tons
2007
Mobile
PM
Tons
HDV
Diesel
1980­
1990
17643
130012
245594
18314
HDV
Diesel
1991­
1995
14878
74315
425493
10309
HDV
Diesel
1996
and
Newer
28957
130327
750623
10922
LDT
Diesel
1980­
1990
490
856
517
67
LDT
Diesel
1991­
1995
0
0
0
0
LDT
Diesel
1996
and
Newer
0
0
0
0
LDV
Diesel
1980­
1990
208
455
415
49
LDV
Diesel
1991­
1995
5
12
11
1
LDV
Diesel
1996
and
Newer
23
79
78
6
HDV
Gasoline
1980­
1990
15972
59040
26432
271
HDV
Gasoline
1991­
1995
27848
120155
56785
529
HDV
Gasoline
1996
and
Newer
18287
159962
76725
903
LDT
Gasoline
1980­
1990
596512
4595802
172364
986
LDT
Gasoline
1991­
1995
411751
3311941
267618
799
LDT
Gasoline
1996
and
Newer
426324
5834171
622965
1692
LDV
Gasoline
1980­
1990
248671
1386020
109702
236
LDV
Gasoline
1991­
1995
394043
2660195
237277
674
LDV
Gasoline
1996
and
Newer
282904
3879941
298509
1214
Table
A.
1
­
Emission
Inventory
Contribution
of
EFP
Sample
Strata
Appropriate
sample
size
for
each
stratum
is
found
in
Table
A.
2.

Sampling
size
for
activity
data
is,
at
this
point,
unknown.
It
is
believed
a
sufficient
number
of
vehicles
will
be
sampled
in
the
KC
work
to
develop
appropriate
statistics
to
identify
an
efficient
sample
and
sample
size.
A
target
of
100
vehicles
across
all
the
strata
per
year
is
the
sampling
goal
for
this
work.

2(
b)(
iii)
Stratification
variables
The
stratification
of
the
sample
is
based
on
the
model
year
group
(
grouped
by
similar
emission
standards),

vehicle
type,
and
fuel
type).
For
this
information
collection
the
Light
Duty
Trucks
shall
be
defined
for
this
information
collection
as
all
trucks
defined
in
40
CFR
part
86
subpart
A
as
light­
duty
trucks
1,
light­
duty
trucks
33
2,
light­
duty
trucks
3,
and
light­
duty
trucks
4.
Heavy
Duty
Vehicles
shall
be
for
this
information
collection
all
those
vehicles
that
meet
the
definitions
for
the
engines
that
power
them
in
40
CFR
part
86
subpart
B
(
Otto­
Cycle
Complete
Heavy­
Duty
Vehicles)
and
40
CFR
part
86
subpart
D.
Motor
Cycles
are
those
vehicles
that
conform
to
that
definition
found
in
40
CFR
part
86
subpart
E.

The
fuel
types
are
those
mentioned
on
the
title
40
CFR
part
86
subpart
A.
However,
the
vast
majority
of
on­
highway
vehicles
operate
on
gasoline
and
diesel
fuel.
Vehicles
using
these
fuels
will
be
the
target
population
of
this
information
collection.
Furthermore,
fuels
used
in
this
information
collection
shall
be
either
in­
use
gasolines
and
diesel
fuels
or
one
of
several
standard
test
gasoline
or
diesel
fuels
used
in
past
emission
factor
testing.

Vehicle
emissions
deteriorate
as
the
vehicle
ages
or
accumulates
mileage.
The
vehicle
age
also
encapsulates
a
vehicle's
emission
technology
and
therefore
its
emission
standard.
Therefore,
the
most
practical
manner
to
stratify
the
key
variables,
emission
control
technology
and
age)
is
to
identify
model
years
that
capture
when
emission
standards
were
constant.
It
also
simplifies
the
sampling
frame.
The
strata
for
this
information
collection
are
summarized
in
the
following
Table
A.
2:
34
Vehicle
Strata
Fuel
Strata
Model
Year
Strata
n
used
for
calulating
sample
size
for
VOC,
CO,
CO2,
NOx
ICR
Sample
Size
for
VOC
ICR
Sample
Size
for
CO
ICR
Sample
Size
for
CO2
ICR
Sample
Size
for
NOx
n
used
for
calulating
sample
size
for
PM
ICR
Sample
Size
for
PM
n
used
for
calulating
sample
size
for
VOC
Evap
ICR
Sample
Size
for
VOC
Evap
HDV
Diesel
Pre
1980
10
66
19
8
32
10
26
HDV
Diesel
1980­
1990
53
472
28
4
11
53
48
HDV
Diesel
1991­
1995
36
518
11
38
23
32
194
HDV
Diesel
1996
and
Newer
27
5
0
1
7
27
52
LDT
Diesel
1980­
1990
9
270
1
9
25
9
89
LDT
Diesel
1991­
1995
36
35
2
1
18
36
50
LDV
Diesel
Pre
1980
6
42
0
5
12
6
12
LDV
Diesel
1980­
1990
29
32
1
7
20
29
47
LDV
Diesel
1991­
1995
4
0
0
0
0
4
0
LDV
Diesel
1996
and
Newer
1
0
0
0
0
1
0
HDV
Gasoline
1980­
1990
64
21
22
1
31
0
HDV
Gasoline
1991­
1995
292
443
347
2
31
4
0
70
434
HDV
Gasoline
1996
and
Newer
47
131
14
1
36
1
0
17
392
LDT
Gasoline
Pre
1980
54
150
86
5
28
23
131
14
172
LDT
Gasoline
1980­
1990
2000
136
112
4
67
184
522
2790
101
LDT
Gasoline
1991­
1995
4488
95
8
3
37
132
5
82
6
LDT
Gasoline
1996
and
Newer
162
35
1
4
130
9
1
0
LDV
Gasoline
Pre
1980
507
154
116
6
40
42
215
60
96
LDV
Gasoline
1980­
1990
13692
402
109
4
50
296
376
20334
79
LDV
Gasoline
1991­
1995
9920
57
14
3
39
220
1
934
207
LDV
Gasoline
1996
and
Newer
132
21
16
1
31
13
0
0
Table
A.
2
­
Sample
Strata
and
Size
What
the
table
does
not
show
are
vehicles
to
be
tested
and
sampled
that
are
selected
to
correct
non­
sampling
errors
(
see
2(
c)(
ii)).
These
vehicles
are
chosen
because
they
are
high­
emitting
vehicles,
new
technology
vehicles,

and
strata
in
the
above
table
that
have
been
characterized
with
sample
sizes
less
than
20
observations.
The
new
35
technology
vehicles
are
associated
with
new
emission
standards.
For
this
information
collection
new
emission
standards
occurred
for
light
vehicles
(
cars
and
trucks)
in
model
year
2000
(
addition
of
Supplemental
Federal
Test
Procedure)
and
heavy
duty
vehicles
in
model
year
2004
(
primarily
for
NOx
control).

2(
b)(
iv)
Sampling
method
Vehicles
shall
be
selected
randomly
using
two
stages
and
probability
proportional
sampling
at
both
stages
from
a
particular
EPA
region.
This
technique
shall
be
used
to
produce
weighted
sample
emissions
for
each
stratum
that
shall
be
applied
to
the
national
vehicle
fleet.
The
first
sampling
stage,
the
primary
sampling
unit
(
PSU),
shall
be
counties
or
areas
within
a
particular
EPA
region.
The
region
selected
initially
for
this
ICR
is
the
Kansas
City
Metropolitan
Area
(
KCMA)
in
EPA
Region
7.
It
was
selected
because
of
its
involvement
in
an
ongoing
Federally
funded
transportation
congestion
mitigation
program
with
Department
of
Transportation
(
DOT).
Much
of
the
ground
work
identifying
vehicles
has
already
been
done
and
related
to
possibly
pertinent
demographic
data
such
as
family
income,
zip
codes,
and
vehicle
miles
traveled.
Another
PSU
shall
be
an
area
50
miles
around
the
city
of
Ann
Arbor,
Michigan.
This
area
was
selected
because
of
EPA's
test
facilities
in
the
city
of
Ann
Arbor
and
all
the
special
resources
that
facility
has.
There
is
an
anticipated
national
vehicle
emission
and
activity
program
to
begin
in
the
later
years
of
the
ICR
where
counties
(
PSUs)
shall
be
selected
using
proportional
random
samples
to
deploy
onboard
emission
and
activity
measurement
instrumentation.
This
approach
will
free
us
from
the
need
for
having
laboratory
facilities
near
by
and
will
allow
EFP
to
develop
a
national
sample
of
vehicles
to
monitor
and
use
for
modeling
purposes.

The
value
of
proportional
random
sampling
is
that
each
PSU
shall
have
sample
weights
applied
to
it,
and
its
contribution
to
the
sample
will
carry
those
weights
along
with
the
emission
and
activity
data.
This
allows
EFP
to
collect
a
more
efficient
sample
and
scale
the
data
in
an
appropriate
manner.
36
The
actual
selection
of
participants
within
the
PSU
will
be
based
on
the
desired
sample
size
per
stratum.

However,
a
participating
vehicle
shall
be
assigned
a
weighing
factor
much
like
the
PSU
to
reflect
the
stratum's
relative
contribution
to
the
national
fleet
as
reflected
by
the
national
distribution
of
strata
at
the
time
of
emission
test
or
sampling
program.

2(
b)(
v)
Multi­
stage
Sampling
Multi­
stage
sampling
shall
be
used
in
any
national
vehicle
emission
and
activity
sampling
program.
The
first
stage
shall
be
the
selection
of
the
PSU
using
proportional
random
sampling
with
replacement.
The
second
stage
of
sampling
shall
be
a
stratified
random
sample
of
vehicles
from
vehicle
registration.
The
stratum
is
based
on
the
three
vehicle
classes,
two
fuel
types,
and
model
year
groups
that
reflect
past,
current,
and
future
emission
standards.

2(
c)
Precision
Requirements
2(
c)(
i)
Precision
targets
The
precision
target
for
this
information
collection
is
a
sample
of
vehicles
that
will
produce
a
95
%

confidence
interval
about
the
average
emission
factor
of
20
%
or
the
emission
rate
or
the
5th
Bin
of
the
Tier
2
standard
for
Light
Duty
Vehicles
whichever
is
greater.
That
target,
along
with
a
stratum's
emission
factor
uncertainty,
shall
be
used
to
determine
sample
sizes
for
each
stratum.

2(
c)(
ii)
Non­
sampling
Error
Unusual
problems
requiring
special
sampling
­
MOBILE
is
most
often
used
to
predict
future
emission
estimates
for
the
mobile
source
fleet
and
obtain
better
emission
estimates
on
high­
emitting
vehicles.
These
vehicles
may
not
be
found
in
large
numbers
in
the
general
fleet,
but
their
impact
on
present
may
either
be
disproportionate
because
of
their
high
emission
rates
(
the
high­
emitters)
or
they
may
be
very
important
in
the
future,
the
new
technology
vehicles.
Therefore,
special
efforts
beyond
random
selection
are
required
to
obtain
37
these
vehicles
.

2(
d)
Vehicle
Log
Design
Vehicle
log
information
is
used
to
interpret
vehicle
load
during
emission
and
activity
sampling
with
onboard
instrumentation.
Currently,
onboard
instrumentation
can
not
tell
directly
how
many
passengers
are
onboard,
if
the
vehicle
is
carrying
cargo,
the
type
trip
taken,
and
if
any
vehicle
maintenance
preceded
a
trip.
The
goal
is
to
use
the
data
to
help
to
characterize
load's
effect
on
emission
levels.
Vehicle
load
Questionnaire
responses
help
to
confirm
the
emission
measurements
made
on
the
vehicle.

The
log
requires
three
types
of
information
to
be
collected;

°
Key
on
and
key
off
events
are
needed
to
coordinate
the
log
entries
with
the
onboard
instrumentation.

°
Vehicle
load
information
is
collected
by
identifying
how
much
fuel
is
onboard,
the
number
of
passengers,

the
type
trip
taken,
and
the
type
cargo
that
is
being
carried.

°
Short
description
of
any
maintenance
that
preceded
a
key
on
event.

The
key
on
and
key
off
events
require
the
recording
the
time
of
day
and
date.
The
vehicle
load
information
is
handled
by
a
check
box
for
fuel
level,
a
number
for
passengers
between
the
age
of
0
and
11,
12
and
17,
and
adults.
Any
cargo
is
characterized
by
the
type
trip
the
vehicle
is
doing,
commuting,
shopping,
or
errands.

3.
Pretests
and
Pilot
Tests
A
pretest
of
the
vehicle
log
was
performed
in
spring
of
2003
to
evaluate
how
well
the
form
is
designed,

the
accuracy
,
and
usefulness
of
the
data,
and
saliency
of
the
form
to
the
participant.
This
will
determine
if
the
form
needs
improvement,
is
adequate
to
the
task,
and
whether
some
incentive
may
be
required
to
assure
that
it
is
filled
out
in
a
timely
and
accurate
manner.

4.
Collection
Methods
and
Follow­
up
38
4(
a)
Collection
Methods
There
are
three
types
of
collection
in
the
EFP.
The
first
is
the
emission
measurements
conducted
in
the
laboratory.
They
are
needed
to
characterize
a
vehicle's
emissions,
a
set
of
the
information
collection's
key
variables.

The
second
type
collection
is
emission
and
activity
measurements
sampled
in
the
field
using
onboard
instrumentation.
This
is
one
of
two
methods
of
EFP
to
obtain
emission
data
and
the
only
method
of
obtaining
activity
data,
both
of
which
are
key
variables
of
the
information
collection.

The
third
collection
method
is
the
vehicle
log.
It
is
required
to
obtain
vehicle
load
information
during
emission
and
activity
sampling.
Currently
there
is
not
appropriate
instrumentation
to
measure
a
vehicle's
payload
during
the
trip
other
than
direct
weighing
of
the
vehicle
before
and
after
loading
or
by
querying
the
vehicle
operator.
The
former
method
is
not
operationally
practical
so
a
vehicle
log
must
be
kept
to
obtain
that
information.
The
participant
shall
be
instructed
and
given
an
appropriate
incentive
to
fill
out
the
log
as
completely
as
possible.
The
log
is
designed
to
filled
out
as
the
vehicle
is
operated
and
will
not
require
that
the
participant
to
recall
historical
information.

4(
b)
Survey
Response
and
Follow­
Up
The
response
rates
for
the
EFP
historically
is
less
than
25
%
using
vehicle
registration
lists
and
90
%

when
participants
were
recruited
from
an
inspection
and
maintenance
test
lane.

A
summary
of
registration
list
response
rates
since
1995
is:

°
Ultimate
tested
sample
size
­
520
°
Asked
to
participate
­
2361
°
Response
rate
22
%
39
In
the
1980s,
a
concern
had
been
raised
about
the
Emission
Factor
Program
regarding
the
historically
low
response
and
participation
rate
by
vehicle
owners.
This
hesitancy
to
participate
by
vehicle
owners
is
largely
understandable,
since
the
request
to
borrow
a
vehicle
for
several
weeks
is
far
from
trivial.
There
has
been
intensive
analysis
of
this
issue
by
the
University
of
Michigan's
Survey
Research
Center
in
the
early
1980s
and
by
a
focus
interview/
survey
conducted
early
this
year.
To
maximize
response
rates,
a
follow­
up
mailing
and
phone
call
to
the
owner
is
made,
the
owner
is
provided
with
a
new
model
loaner
vehicle
and
a
small
monetary
incentive,
and
the
owner's
vehicle
can
be
picked
up
and/
or
delivered
by
the
EPA
contractor.
Further
activities
will
be
tried
to
improve
the
solicitation
materials
and
promote
the
program
in
the
Ann
Arbor
area.

The
participation
rate
in
the
EFP
for
vehicle
owners
who
are
solicited
from
inspection
lane
is
much
greater
than
the
rate
in
the
traditional
EFP.
In
the
last
EFP
test
program
where
lane
recruitment
was
used
the
rates
were
since
1995:

°
Ultimate
tested
sample
size
­
195
°
Asked
to
participate
­
214
°
Response
rate
91
%

The
participation
rates
in
Hammond
is
discussed
in
section
5(
b)
of
part
A.

There
are
probably
many
reasons
that
the
participation
rate
is
higher
in
this
type
of
EFP.
Some
of
the
reasons
may
be:
personal
contact
between
the
EPA
representative
and
the
vehicle
owner;
the
owner
can
observe
an
emissions
test
that
is
similar
to
those
his/
her
vehicle
would
be
subject
to;
the
specific
loaner
vehicle
that
the
owner
will
receive
is
available
for
immediate
examination;
if
the
vehicle
fails
the
State
inspection,
EPA
agrees
to
fix
the
vehicle.

During
this
information
collection,
vehicle
recruitment
using
registration
lists
will
be
handled
by
a
marketing/
research
organization,
not
a
vehicle
testing
organization.
The
goal
is
better
participation
rates
and
10"
Characterizing
Exhaust
Emissions
from
Light­
Duty
Gasoline
Vehicles
in
the
Kansas
City
Metropolitan
Area",
Statement
of
Work,
March,
2003
40
owner/
participant
demographics.
This
process
shall
be
used
in
the
Kansas
City
10study
in
FY
2003
and
FY
2004.

The
process
shall
entail
a
recruitment
pilot
that
will
be
used
to
identify
appropriate
sample
frame
demographics.

The
demographics
shall
then
be
used
to
determine
what
incentives
and
program
descriptions
would
make
the
program
more
appealing
and
salient
to
possible
participants.
This
would
form
the
basis
of
a
suitable
demographically
specific
recruitment
strategy
that
would
be
tested
and,
once
proven,
applied
to
the
total
program.

The
contractor
would
also
determine
the
source
of
non
or
negative
respondents
and
formulate
methods
to
recruit
them
in
later
stages
of
the
testing
and
sampling
program.

5.
Analyzing
and
Reporting
Survey
Results
5(
a)
Data
Preparation
The
accuracy
of
the
emission
rates
are
ensured
by
adherence
to
standard
test
procedures
and
the
agency
requirement
that
all
emission
measurements
are
managed
under
a
quality
assurance
action
plan
that
is
specific
for
all
individual
testing
and
sampling
programs
performed
by
EPA.
The
procedures
for
emission
testing
(
e.
g.,

Federal
Test
Procedure
and
Highway
Fuel
Economy
Test)
were
developed
as
part
of
various
regulations
and
were
therefore
subject
to
intensive
review
by
both
EPA
and
the
regulated
industry.
Emission
sampling
programs,

though
relatively
new
to
mobile
source
programs,
have
gone
through
two
prototype
programs
and
are
now
under
extensive
review
by
CRC.
All
the
test
procedures
utilized
in
the
EFP
are
described
in
the
test
procedure
manuals
maintained
by
the
Testing
Services
Division.
The
specific
procedures
applicable
to
the
EFP
are
referenced
in
the
April
3,
1987
memo,
"
Revised
Memo
of
Understanding
­
EFP
Test
Procedures."

The
data
for
the
EFP
are
entered
into
the
computer
data
base
in
several
ways.
First,
the
basic
vehicle
41
identification
information
is
coded
on
data
forms.
This
information
is
keyed
into
a
computer
file
through
data
entry
checking
program
specific
for
vehicle
information.
The
emissions
data
are
entered
in
a
variety
of
ways,

depending
upon
the
capabilities
of
the
test
facility
supplying
the
data.
In
general,
the
majority
of
the
emission
and
activity
data
is
collected
and
transferred
to
the
EFP
data
base
electronically.
All
of
the
emissions
data
collected
at
the
Ann
Arbor
test
facility
is
transmitted
directly
from
the
measurement
equipment
to
the
computer
files.
The
emissions
data
provided
by
contractors
is
done
in
the
same
manner.
All
data
is
eventually
placed
in
files
compatible
with
MSOD
input
files.

All
input
files
are
checked
by
the
testing
contractor
using
a
sequence
of
EFP
quality
control
computer
programs.
These
programs
are
systematically
tailored
for
and
applied
to
all
EFP
emission
and
activity
data.
The
programs
are
applied
to
the
test
data
by
the
testing
contractor
and
EFP
staff
before
the
data
is
loaded
into
the
database
MSOD.
Furthermore,
MSOD
is
a
fully
functional
relational
database
with
business
rules
enforced
by
relational
constraints
implied
by
the
database's
design.

All
aspects
of
quality
control
other
than
data
entry
and
final
review
are
governed
by
the
quality
assurance
plans
referred
to
in
part
A,
section
4(
b).

5(
b)
Data
Analysis
Generally
the
basic
emission
rates
are
estimated
by
linear
or
piecewise
linear
regression
of
emissions
(
dependent
variable)
versus
vehicle
mileage
(
independent
variable)
using
all
the
data
collected
in
past
EFPs
pertaining
to
a
given
vehicle
type.

The
basic
emission
rates
provide
estimates
for
average
urban
summer
conditions
for
ambient
temperature
and
vehicle
speed.
The
MOBILE
user
needs
to
predict
emissions
at
other
than
these
average
test
conditions.

EFP
data
are
also
used
to
estimate
correction
factors
for
nonstandard
test
conditions.
Again
in
this
situation,
least
42
squares
regression
procedures
are
used
to
determine
the
prediction
equation
for
emissions
(
dependent
variable)
as
a
function
of
ambient
temperature
or
average
speed
(
independent
variables).
The
regression
equation
is
then
normalized
to
the
average
urban
summer
conditions
to
apply
as
a
correction
factor
to
the
basic
emission
rate
equation.

5(
c)
Data
Reporting
The
EFP
emission
results
on
an
individual
vehicle
are
made
available
to
the
vehicle
owner
upon
request.

All
the
data
recorded
on
the
EFP
data
base
is
available
upon
request
on
CD­
ROM.
Eventually
it
will
be
available
from
the
EPA
Internet
website.

The
EFP
data
are
analyzed
and
used
in
a
series
of
computer
models,
the
MOBILE
models.
The
documentation
of
MOBILE
is
a
supplement
to
a
series
of
reports
known
as
AP42.
The
last
formal
update
to
AP42
(
Supplement
A
to
the
Compilation
of
Air
Pollutant
Emission
Factors
­­
Volume
II:
Mobile
Sources)
was
published
January
1991.
Portions
of
this
supplement
(
Appendices
G,
H,
I,
J
and
K)
have
been
updated
since
that
time
are
available
on
the
EPA
Web
Site
(
www.
epa.
gov).
EPA
is
in
the
process
of
updating
the
MOBILE
model
itself
and
technical
reports
relevant
to
the
analysis
are
posted
on
the
EPA
Website
for
stakeholder
review.
Public
MOBILE
workshops
were
held
in
March
and
October
1997.
Completion
of
the
updated
MOBILE
model
is
expected
in
fiscal
year
2000.
