WHICH VEHICLES HAVE A HIGHER PROBABILITY OF FAILING A CALIFORNIA
SMOG CHECK INSPECTION PRIMARILY CONSISTING OF A DIAGNOSTIC SCAN OF
THE VEHICLE’S ON-BOARD COMPUTER SYSTEM?
A Thesis
Presented to the faculty of the Department of Public Policy and Administration
California State University, Sacramento
Submitted in partial satisfaction of
the requirements for the degree of
MASTER OF PUBLIC POLICY AND ADMINISTRATION
by
William Dean Thomas
SPRING
2014
© 2014
William Dean Thomas
ALL RIGHTS RESERVED
ii
WHICH VEHICLES HAVE A HIGHER PROBABILITY OF FAILING A CALIFORNIA
SMOG CHECK INSPECTION PRIMARILY CONSISTING OF A DIAGNOSTIC SCAN OF
THE VEHICLE’S ON-BOARD COMPUTER SYSTEM?
A Thesis
by
William Dean Thomas
Approved by:
__________________________________, Committee Chair
Su Jin Jez, Ph.D.
__________________________________, Second Reader
Robert Wassmer, Ph.D.
____________________________
Date
iii
Student: William Dean Thomas
I certify that this student has met the requirements for format contained in the University format
manual, and that this thesis is suitable for shelving in the Library and credit is to be awarded for
the thesis.
__________________________, Department Chair
Robert Wassmer, Ph.D.
Department of Public Policy and Administration
iv
__________________
Date
Abstract
of
WHICH VEHICLES HAVE A HIGHER PROBABILITY OF FAILING A CALIFORNIA
SMOG CHECK INSPECTION PRIMARILY CONSISTING OF A DIAGNOSTIC SCAN OF
THE VEHICLE’S ON-BOARD COMPUTER SYSTEM?
by
William Dean Thomas
An important component of California’s smog check program is the policy of directing the
highest polluting vehicles to specialized smog check stations for testing. The State Bureau of
Automotive Repair identifies these ‘directed vehicles’ through the use of a regression model that
identifies the potentially highest polluting vehicles based upon past tailpipe emissions readings of
the same type of vehicles. However, beginning in 2014, the testing procedure for a large portion
of the vehicles in California will no longer include a tailpipe emissions measurement.
The revised testing procedure will rely upon a scan of the on-board computer diagnostic
and control system that controls and constantly evaluates the function of the engine and emissions
control systems present on most vehicles manufactured since 1996. The revised procedure will
also include a visual inspection of the emissions control devices present on the vehicle.
Consequently, there is a need for a regression model capable of identifying the vehicles with the
highest likelihood of failure based upon the results of the scan of the diagnostic system and the
visual inspection.
In this thesis, I developed a binomial logistic regression model that predicts which vehicles
are highly likely to fail the vehicle computer diagnostic scan or visual inspection procedure
comprising the new inspection procedure. The regression analyses described herein accurately
v
identified a group of approximately 40% of the vehicles subject to smog check inspections that
have a higher likelihood of failure than the remaining vehicles subject to testing. Implementation
of the regression models described in this thesis, or similar models, will enable the Bureau to
continue to identify approximately 30% of the fleet of vehicles as directed vehicles.
_______________________, Committee Chair
Su Jin Jez, Ph.D.
_______________________
Date
vi
ACKNOWLEDGEMENTS
I would like to thank Dr. Jez and Dr. Wassmer for their guidance and support in this
endeavor. Their always-accurate suggestions were immensely important in keeping this project
moving in the right direction and helping to achieve final product.
Additionally, I cannot write an acknowledgements page without recognizing the support
provided by my family. Numerous times over the last months, I have been physically present at
family functions without mental presence because of my focus on which regression approach
works best or how to best clean 8 million observations. Moving forward, they will no longer hear
“I can’t go on that camping trip because of the thesis” or other missed functions.
Finally, I must acknowledge the ever-present love and support of my wife, Suzanne. She
too has taken a back seat to this thesis for several months, an action I will happily spend the next
several months making up for.
Thank You All!
vii
TABLE OF CONTENTS
Page
Acknowledgements ................................................................................................................ vii
List of Tables ............................................................................................................................ x
Chapter
1. INTRODUCTION .............................................................................................................. 1
Origins of the Smog Check Inspection Program ......................................................... 1
The Creation of Test-Only Stations ............................................................................ 4
Directed Vehicles ......................................................................................................... 5
Smog Check Inspection Procedure .............................................................................. 7
Further Program Analysis ............................................................................................ 8
Recommendations for Program Improvement ............................................................. 9
Modern Vehicle Technology ..................................................................................... 11
Research Question ..................................................................................................... 12
Organization of Remainder of Paper ......................................................................... 12
2. LITERATURE REVIEW .................................................................................................. 14
Smog Check Overview ............................................................................................... 14
Modeling to Predict Emissions Failures ..................................................................... 16
On-Board Diagnostics, Generation Two (OBD II) Systems ....................................... 19
Empirical Studies of OBD II Systems ........................................................................ 21
Conclusion .................................................................................................................. 25
3. METHODOLOGY ............................................................................................................ 26
Smog Check Test Data ............................................................................................... 27
viii
Regression Models: OBD II Test Failure ................................................................... 28
Regression Models: Visual Inspection Failure ........................................................... 34
4. ANALYSIS ........................................................................................................................ 36
Predicting OBD II Test Failures ................................................................................. 37
Predicting Visual Inspection Failures ......................................................................... 43
Reliability and Validity ............................................................................................... 48
5. CONCLUSION .................................................................................................................. 53
Predicting Likelihood of Failure of the OBD II Functional Test ................................ 54
Predicting Likelihood of Failure of the Visual Inspection .......................................... 56
Significance of Findings ............................................................................................. 57
Recommendations for Identifying Directed Vehicles ................................................. 58
Suggestions for Improved Statistical Modeling .......................................................... 60
Conclusion .................................................................................................................. 60
Appendix A: Test Volume for Pre-Model-Year 2000 Vehicles .............................................. 62
Appendix B: Distribution of Vehicle Look-Up Table Identification Numbers. ...................... 63
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test, with Basic
Vehicle Descriptions ....................................................................................... 102
Appendix D: VLT ID Numbers Demonstrating Zero Failures of the Visual Inspection, with
Basic Vehicle Descriptions. ............................................................................ 138
Appendix E: VLT ID Numbers Most Likely to Fail the Visual Inspection, with
Basic Vehicle Descriptions ............................................................................. 144
References ............................................................................................................................. 149
ix
LIST OF TABLES
Tables
Page
1.
Attrition rate of older vehicles .........................................…………………………..…….7
2.
Distribution of Failing Vehicles in Current Inspection Procedure . ... ………………………26
3.
Distribution of Vehicle Manufacturers within the dataset. ...... ……………………………33
4.
Distribution of Vehicle Model-Years within the dataset ........... …………………………. 34
5.
List of Vehicles Demonstrating Zero OBD II Functional Test Failures…………………….38
6.
Results of Regression Model with Vehicle Make as the Explanatory Variables …………….41
7.
Odds Ratios for Vehicle Model-Years ..................................... ………………………….43
8.
Results of Regression Model with Vehicle Make as the Explanatory Variables…………….46
9.
Results of Regression Model with Model-Year as the Explanatory Variables……………....48
10.
Detail of Chi-Squared Values for Regression Model ….……… ………………………….51
x
1
Chapter 1
INTRODUCTION
Nearly thirty years ago, the State of California began a program of biennial smog check
inspections on most motor vehicles registered and operated in the state. Since that time, we have
become accustomed to having to get a smog check every other year when we renew our
registration or when we transfer ownership of our vehicle to another Californian in a private party
transaction. What many people may not be aware of are the various statutes governing the
implementation of the program, including designating the Bureau of Automotive Repair
(hereinafter referred to as the Bureau) as the oversight and regulatory enforcement entity for the
program, appropriately referred to as the “Smog Check Inspection Program”.
An on-going component of the Smog Check Inspection Program is the requirement to
identify vehicles highly likely to fail the inspection. Advances in automotive technology and
program changes often require adjustments to the model used to identify these vehicles. This
thesis proposes new methodologies for identifying these vehicles and details the analyses
conducted to validate these methodologies.
The following sections of this chapter include a description of the smog check inspection
program, the origin of the program, general and specific requirements of the program, and past
and future changes to the program. There is also a discussion of the methodologies used for
identifying vehicles highly likely to fail a smog check inspection, the policies behind those
methodologies, and the need for revisions to the methodologies. Chapter 1concludes with a
summary of the remaining chapters in this thesis.
Origins of the Smog Check Inspection Program
California led the way in the automotive pollution control realm by implementing a
program to identify high polluting vehicles in the mid 1960’s (Eisinger, 2010). This initial smog
2
check program consisted of nothing more than a basic visual inspection of the basic emission
control components common to vehicles produced at that time. Throughout the 1970’s, while
California continued to rely upon a basic visual inspection, other states began implementing what
were referred to as “inspection and maintenance programs”, commonly referred to as I/M
programs, which actually measured vehicle emissions rather than simply conducting a visual
inspection.
Eisinger (2010) describes California’s attempt at an I/M program conducted in the Los
Angeles area between 1979 and 1984 as “unpopular and inconvenient” (p. 29), due to the amount
of time required to conduct an actual emissions measurement compared to the mere minutes
required for the visual inspection, which is what California motorists were accustomed to.
Although the United States Environmental Protection Agency (EPA) pushed for expansion of the
I/M program throughout the state as part of 1977 amendments to the Clean Air Act of 1970,
California resisted. Elected officials and environmental policy makers were in a difficult
situation; maintaining leadership in environmental policy would require inconveniencing millions
of motorists. The federal government forced the issue when in December of 1980 they delayed
the transfer of federal highway funds to California as a sanction for failing to comply with EPA
mandates (Eisinger, 2010). In 1981 and in response to the withholding of highway funds,
California State Senator Robert Presley proposed a new approach to emissions inspection
programs in Senate Bill 33.
Senate Bill 33 (Presley, Chapter 892, Statutes of 1982) mandated the implementation of a
smog check inspection program in the State of California. This program of testing vehicles for
compliance with emission control regulations by privately owned smog check stations began in
March of 1984, and although changes to the program have occurred over the years, the program is
still in place today. The purpose of the program is to identify high-polluting vehicles and require
3
the repair or retirement from operation in California of those vehicles. The program became
necessary because of the failure of various regions within the state to meet United States
Environmental Protection Agency (EPA) air quality standards (Eisinger, 2010). The program
never achieved the expected results and after a 1987 study highlighted the extent of the failure of
the program, the State implemented changes aimed at improving the smog check inspection
program (Eisinger, 2010).
However, the enhancements still failed to achieve the desired results and the Federal
Clean Air Act amendments of 1990 enacted even more stringent air quality requirements,
resulting in an even greater number of California regions placed in a ‘non-attainment’ status;
indicating these regions fail to meet minimum air quality standards. California’s model of
independently owned automotive repair facilities performing all but a very small portion of the
smog check inspections in the state was inconsistent with the EPA’s desired model of
governmentally contracted centralized inspection facilities (United States Environmental
Protection Agency, 1992). USEPA (1992) policies demonstrated a preference for government
controlled inspection facilities that are completely separate of repair functions, presumably with
the intent to exert more regulatory oversight over the program and to dissuade fraud in the
program. However, based upon the number of vehicles in the state and after the failure of the
1970’s centralized inspection program in the Los Angeles Area, California officials were certain
the centralized inspection model would result in failure (Eisinger, 2010). Consequently, the
Bureau, the Air Resources Board of the California Environmental Protection Agency (ARB), and
the state legislature vigorously opposed the centralized testing model and pushed for alternative
solutions. One of those solutions, eventually drafted into legislation and becoming law, expanded
4
the existing inspection program by creating a network of privately owned ‘test-only’ stations1
(California State Assembly, 1994).
The Creation of Test Only Stations
The test-only stations met the EPA’s goal of separation of testing and repair functions, as
statute prohibits test-only stations from performing any type of repairs. As a solution to the
aforementioned disagreement between the EPA and California over the model for smog check
inspections, California proposed and the EPA approved the test-only station model (California
State Assembly, 1994). Statute prohibits test-only stations from performing any type of
automotive repairs other than testing services (Health and Safety Code, § 44014.5.(b)).
In addition to limiting the types of services provided at test-only stations, statute also
directs a certain portion of the vehicles required to receive a smog inspection to test-only stations.
These vehicles, referred to as directed vehicles, account for approximately 30% of the vehicles
tested each year and are statistically more likely to fail a smog check inspection (BAR, 2012). I
discuss the method for identifying directed vehicles later in this chapter. Finally, regulations
permit test-only stations to certify ‘gross polluting’ vehicles, which are vehicles exhibiting
exceedingly high emissions levels. Regulations prohibit regular, non-test-only stations, referred to
as “Test and Repair” stations, from certifying directed and/or gross polluting vehicles. The theory
behind test-only stations was that by separating the repair and inspection functions of the smog
check program, there would be no incentive for the test-only station to fraudulently certify a
vehicle; thereby, ensuring the integrity of the smog check inspection program. The incentive for
being a test-only station was the privilege of having a certain portion of the fleet directed to these
stations (California State Assembly, 1994).
Health and Safety Code section 44010 establishes the mechanism for privately owned ‘stations’ which shall be
referred to as smog check stations and are authorized to certify vehicles pursuant to the Motor Vehicle Inspection
Program established by the Code. Throughout this thesis, the terms station or stations are referring to smog check
stations.
1
5
These test-only stations met California’s goal of continuing to allow privately owned
facilities to perform practically all (a small portion, 0.45%, of inspections are performed at
Bureau contracted referee stations) smog check inspections while meeting the EPA’s goal of
separating inspection and repair functions (California State Assembly, 1994). The statutes
authorizing privately owned test-only stations were contained in SB 521 (Presley, Chapter 29,
Statutes of 1993) and became effective in March of 1994. Test-only stations first began operation
in 1997 and eventually grew to comprise the significant portion of smog check stations, 34%, but
performing a majority, nearly 65% of the approximate 12 million initial inspections annually
(Bureau of Automotive Repair, 2012b).
Directed Vehicles
The practice of identifying and directing vehicles for testing at specific stations was also
created by the smog check program amendments implemented on March 30, 1994, pursuant to
Senate Bill 521 (Presley, Chapter 29, Statutes of 1993). As stated above, test-only stations receive
the privilege of testing directed vehicles. Beginning in 1996 and continuing until 2012, the
Bureau utilized a model referred to as the high-emitter profile (HEP) model (Bureau of
Automotive Repair, 2012a and Choo, Shafizadeh, and Niemeier, 2007) to identify directed
vehicles. Although the exact specification for the HEP model remains protected as intellectual
property, it reportedly is a logistic regression model that predicts whether a vehicle is likely to
generate high emissions based upon certain vehicle design variables (Choo, Shafizadeh, and
Niemeier, 2007).
In 2012, the Bureau reported that beginning in 2013, the primary factor for identifying
directed vehicles would be the model-year of the vehicle (Bureau of Automotive Repair, 2012a).
Specifically, all model-year 1976 through 1999 vehicles will receive directed vehicle status.
However, these model-years are diminishing as a percentage of the total volume of tests
6
performed each year and the Bureau has indicated the volume of directed vehicles will remain
consistent at the current level of approximately 30 percent of the total volume of tests each year
(Bureau of Automotive Repair, 2012a).
A concern with identifying directed vehicles by model-year is simple attrition. As
vehicles age and owners replace them with newer vehicles, the number of older vehicles tested
each year decreases. As an example, in 2011, model-year 1976 to 1999 vehicles accounted for
just over 4.5 million, or 41.80% of the initial inspections performed; while in 2012, these same
vehicles accounted for just over 4.0 million, or 35.54% of the initial inspections (Bureau of
Automotive Repair, 2012b and 2013). Table 1 below highlights the attrition rate of older vehicles.
Appendix A, “Test Volume for Pre-Model-Year 2000 Vehicles”, details the information
presented in this graph. As the graph shows, it will only be a few years before the number of
vehicles meeting these criteria is too small to provide a sufficient set of directed vehicles.
Additionally, California Health and Safety Code section 44010.5 requires the Bureau to direct the
vehicles with the highest probability of generating the highest emission levels. Although it is
reasonable to assume that the oldest vehicles will generate the highest emissions, this is not a
statistically valid approach as required by the code. Finally, beginning in 2014 and as a result of
revisions to the smog check inspection procedure, the measurement of emission levels is longer a
component of the smog check inspection.
7
Table 1
Attrition rate of older vehicles
Percentage of Pre-2000 Vehicles
120%
100%
80%
60%
Percentage of Pre-2000
Vehicles
40%
20%
0%
2001 02
03
04
05
06
07
08
09
10
11
12
Smog Check Inspection Procedure
The current smog check inspection procedure consists of three components. First is a
visual inspection wherein the technician visually inspects all emissions related to components to
verify they are present on the vehicle, properly installed, and are free from any visual defects. The
second portion of the inspection is the functional test. Depending on the model-year of the
vehicle, one or more of the following functional tests will be conducted on the vehicle: LowPressure Fuel Evaporate Test (LPFET), which tests the fuel-evaporative control system (primarily
the fuel-tank) for leaks; ignition timing test, which measures the ignition timing of the engine;
fuel cap test, which is separate from the LPFET and tests the sealing integrity of the fuel cap; OnBoard Diagnostic System, Generation II (OBD II) test, which tests the vehicle’s engine and
emissions control computer; and a visible smoke test, which as the name implies requires the
technician to inspect for smoke emanating from the engine or exhaust of the vehicle. The final
8
portion of the inspection is the emissions measurement portion of the test. (Bureau of
Automotive Repair, 2009)
Two distinct methods are used for measuring vehicle emissions. The first is the twospeed idle (TSI) test which measures the emissions with the vehicle stopped and the engine
running at 2500 revolutions-per-minute (rpm) and at idle speed, generally 600 to 800 rpm. The
second came about as a response to the Federal Clean Air Act Amendments of 1990, and is an
acceleration simulation mode test of vehicle emissions (Singer & Wenzel, 2003). The
acceleration simulation mode (ASM) test places the vehicle on a treadmill-like device allowing
the smog check technician to operate the vehicle under a load, simulating driving conditions;
specifically vehicle acceleration at 15 miles per hour (mph) and 25 mph (Choo et al., 2007). Both
tests are still utilized today, along with the visual inspection and a functional test of certain
emissions control components (Bureau of Automotive Repair, 2009). The ASM test is utilized in
the more populous areas of the state, referred to as “enhanced” areas (Bureau of Automotive
Repair, 2009) and accounted for 76% of the tests performed in 2012 (Bureau of Automotive
Repair, 2013)
Further Program Analysis
SB 521 also mandated that BAR perform random, roadside inspections of vehicles
throughout the state to confirm compliance with vehicle emissions control laws and to obtain
empirical data for smog check program analysis (Health and Safety Code §44024.5). An analysis
of the roadside inspection data gathered between 2000 and 2002 revealed that in the category of
1974 to 1995 model year vehicles that had failed an initial smog check inspection, were repaired
and subsequently certified as passing, 40% of those failed a roadside inspection within one year
after certification (Austin, McClement, & Roeschen, 2009).
9
To confirm the accuracy of this analysis, Sierra Research conducted a study of 2003 to
2006 roadside inspection data and found that 1976 to 1995 model year vehicles were highly likely
to fail a roadside inspection within one year after having failed an official smog check inspection,
receiving repairs, and certified as passing; at a level of 49% (Austin, McClement, & Roeschen,
2009). The researchers hypothesized several possible reasons for these results and concluded that
improper and poor smog check inspection procedures on the part of technicians is the primary
cause of the high roadside inspection failure rates. Further research led to the conclusion that the
test-only stations are the predominate source of improper testing procedures. Although it is
unclear as to why this is the case, one hypothesis is that the expertise needed to only perform
smog check inspections is far less than what is required to diagnose and repair vehicles, leading
to poor performance among test-only technicians.
Recommendations for Program Improvement
The Sierra Research analysis concluded with several recommendations for improvements to
the smog check inspection program. The primary recommendations are as follows:

The establishment of performance standards designed to evaluate the inspection practices
of technicians and stations

Elevated monetary penalties, the ability to issue orders of abatement, and other penalty
enhancements for those found to be improperly or fraudulently inspecting vehicles

Modernized testing procedures for vehicles with advanced emissions control technologies

A complete revision of the model for directing specific vehicles to test-only stations for
inspection, so that only those stations demonstrating the highest performance in smog
check inspections receive the privilege of testing directed vehicles.
The recommendations and other improvements to California’s smog check inspection
program were drafted into new legislation and enacted into law as California Assembly Bill 2289
10
(Eng, Chapter 258, Statutes of 2010). AB2289 directs the Bureau to implement a program for
certifying smog check inspection stations based upon inspection performance and subsequently
directing vehicles most likely to be high emitters and gross polluting vehicles only to those
stations meeting the aforementioned performance standards. In addition, AB2289 directed the
Bureau to implement a new testing model wherein most 2000 model-year and newer vehicles
would receive an inspection that consisted only of a computer scan of the vehicles’ diagnostic
system and a visual inspection of the other applicable emission components. This eliminates the
emissions testing portion of the smog check inspection for these vehicles. This testing model is
consistent with the USEPA’s current preferred inspection model and is already utilized in 31
states and jurisdictions throughout the United States (Environmental Protection Agency, 2013).
Inspection procedures for 1999 model-year and older vehicles remain unchanged. (AB2289)
In developing the performance standards as required by AB2289, the Bureau developed a
regression model that determines the inspection effectiveness of technicians and stations based
upon certain testing behaviors and the probability of a vehicle certified by a particular technician
passing a subsequent smog check inspection. These performance measures encompass the
Bureau’s ‘STAR” program, which became effective on January 1, 2013. The STAR program
identifies smog check stations that meet the applicable performance criteria as ‘STAR Certified’,
and affords these stations the opportunity to inspect and certify directed and gross polluting
vehicles. The performance criteria encompass two sets of measures: short-term and long-term
measures. The short-term measures evaluate behaviors such as bypassing a portion of the test or a
high number of aborted tests, among other factors. The long-term measure looks at a vehicle
certified by a station in the previous test-cycle, 18 to 30 months prior to the current test, and
determines the probability of that vehicle passing. A high confidence level in the vehicle passing
as it should results in a higher score, on a scale of 0.00 to 1.00 and results in a higher performance
11
rating for the station. Consequently, test-only stations are no longer able to ‘self-certify’ as able to
inspect these vehicles and are only able to avail themselves of the increased revenue opportunity
in testing directed and gross polluting vehicles by demonstrating acceptable inspection
performance levels and becoming STAR certified. (BAR, 2012c)
Modern Vehicle Technology
The above statutorily required modernized testing procedures for the Smog Check
program include the elimination of the acceleration simulation mode test for model-year 2000 and
newer vehicles equipped with the on-board diagnostics, generation II (OBD II), system (Lyons
and McCarthy, 2009). The OBD II system is a microprocessor based electronic control system
that monitors and controls the engine operation and the emissions control systems in most lightduty vehicles manufactured since 1996 (Supnithadnaporn, Noonan, Samoylov, and Rodgers,
2011). Light –duty vehicles are passenger cars and light trucks under 14,000 pounds gross vehicle
weight rating. This testing method is scheduled for implementation in late 2013 (Bureau of
Automotive Repair, 2012d).
Using various sensors, the OBD II system monitors the emissions control and engine
management systems to determine if the potential exists for a malfunction that may result in
elevated emissions (Lyons & McCarthy, 2009). If a malfunction is identified, the system may
illuminate the vehicle’s ‘malfunction indicator lamp’ (commonly referred to as the ‘check engine
light’) on the dash and store pertinent information related to the malfunction (Lyons & McCarthy,
2009). Lyons and McCarthy (2009) conducted research and determined that this system is quite
accurate and an inspection system that scans and reports this information, and causes the vehicle
to fail an emissions inspection as a result of a detected malfunction, is equally as effective as the
acceleration simulation mode test. However, a major concern with the Lyons and McCarthy study
is that their sample set consisted of only 74 vehicles, which when compared to the greater than 20
12
million light duty vehicles on California’s roadways is not statistically valid. Conversely, the
EPA also considers the OBD II system to be quite reliable (Environmental Protection Agency,
2013).
Research Question
The program analysis recommending the use of the OBD II diagnostic system for smog
check inspections and redesigning of the model for identifying directed vehicles create a critical
need for a regression model capable of predicting which vehicles have a higher probability of
failing the revised smog check inspection based solely upon the OBD II functional test results.
This leads to the research question for this thesis: Which vehicles have a higher probability of
failing a California smog check inspection primarily consisting of a diagnostic scan of the
vehicle’s on-board computer system? To answer this question, I conducted a regression analysis
to determine which vehicles are most likely to fail the OBD II functional test of the current smog
check inspection. The OBD II functional test is equivalent to the new test procedure that begins in
late 2013.
Although the primary component of the new inspection procedure is a scan of the on-board
diagnostic system of the vehicle, a visual inspection is also part of the procedure. To insure an
accurate model for identifying all vehicles likely to fail the procedure, I developed and evaluated
a second set of regression models aimed at predicting which vehicles would fail this portion of
the new smog check inspection procedure.
Organization of Remainder of Paper
In Chapter 2, I review the available literature regarding on-board diagnostic systems and
their use in smog check inspection programs. Other states are already using this methodology,
and the results generated from their experiences provide useful background for the current
research. Chapter 3 outlines the methodology employed in the conduct of this research, describes
13
the data collection process, and defines the dataset used in the regression analyses. I also describe
the functional form of the regression model selected and discuss the statistical tests performed on
the data to confirm the accuracy and significance of the results. Chapter 4 contains an in-depth
discussion of the results of the regression analysis and proposes an improved model for
identifying directed vehicles. In Chapter 5, I draw conclusions derived from the analysis, discuss
the significance of the findings, and make recommendations for future studies and or policy
initiatives, as appropriate.
14
Chapter 2
LITERATURE REVIEW
Chapter 1 of this thesis provides an introduction to the smog check program in California,
the policies associated with the program, and future direction of the program. These details served
as the impetus for this thesis. In preparation for developing the regression models employed in the
current research, I reviewed numerous articles related to the smog check program and automotive
on-board diagnostic systems. Chapter 2 provides a summary of the articles and a progression of
facts that provide the basis for the statistical analyses described later in this thesis. The first
section of this chapter outlines a summary of the smog check program as described by the
available literature. The next section discusses the various research projects previously completed
that present approaches to statistical models aimed at predicting emission failures. Following is a
section a literature that provides an overview of automotive and light truck on-board diagnostic
systems or OBD II systems in industry vernacular. Finally, I discuss the literature reporting the
results of various empirical studies of OBD II systems, their functionality, and their effectiveness
at identifying emissions test failures.
Smog Check Overview
In response to EPA mandates for the State of California to comply with the Clean Air Act
Amendments of 1990, the State updated its Smog Check program in 1997 and implemented the
acceleration simulation mode (ASM) test described in the introduction to this paper (Eisinger,
2010). The ASM continues as the mandated smog check inspection procedure used in California.
As with the original Smog Check inspection procedure, this updated test consists of the same
three parts: an emissions measurement conducted at the tailpipe of the vehicle while under load,
a visual inspection of the emissions components, and a functional test of certain emissions related
devices or functions such as engine timing (Choo, Shafizadeh, and Niemeier, 2007).
15
The ASM test is a variation of the federal I/M 240 test, which is a 240 second driving
sequence that simulates Los Angeles area rush hour traffic, that was in use in other jurisdictions
(Eisinger, 2010). It is important to note that the development of the ASM test was a compromise
between the EPA, which was pressuring for a centralized testing structure consisting of I/M 240
testing and California officials who felt a centralized testing structure would be ineffective in the
State and I/M 240 testing leads to vehicle failures (Eisinger, 2010). This pattern of conflict and
compromise between State regulators and the EPA has exemplified the Smog Check program
since its inception.
In the mid-1980’s, California implemented the Smog Check program, which mandated a
biennial emissions test for most vehicles registered in the State (Singer & Wenzel, 2003). This
original Smog Check inspection consisted of three parts: a visual inspection of the emissions
control devices, a functional test of a small portion of the devices, and an emissions measurement
of the vehicle’s exhaust (Choo, Shafizadeh, and Niemeier, 2007). The emissions measurement
portion of the inspection measured the level of pollutants in the vehicle’s exhaust with the vehicle
stationary and the engine running at or below 1000 revolutions per minute (rpm) and at 2500 rpm
(Singer & Wenzel, 2003). This emissions measurement test is the "two-speed idle" test and has
expanded to the current ASM test to meet changes in EPA requirements.
Another of the compromises between the EPA and the State of California in creating the
current smog check program was the implementation of test only stations and the direction of a
portion of the total fleet of vehicles registered in the State to these stations. The test only stations
model served to address the EPA’s demands for centralized testing, which would be impractical
in California because of the number of vehicles tested each year (Eisinger, 2012). The prevalent
theory at the time of creation of this category of directed vehicles and the test-only stations was
that there would be no incentive for these stations to game the system by improperly failing
16
vehicles to generate repair revenue or by improperly certifying the vehicles after collecting
payment for repairing a vehicle when those repairs were ineffective (Eisinger, 2010).
Modeling to Predict Emissions Test Failures
The on-going dilemma, however, has been in determining which vehicles are appropriate
for testing at test only stations. The Bureau eventually selected the high emitter profile (HEP)
model developed in 1996 by Radian International corporation discussed earlier (Choo,
Shafizadeh, and Niemeier, 2007). However, a later study demonstrated that this model is only
slightly more accurate than random chance at identifying vehicles likely to fail a smog check
inspection in California (Choo, Shafizadeh, and Niemeier, 2007). Although various research
projects produced other models for predicting failures, the Bureau continued to use the HEP
model until mid-2012 (Bureau of Automotive Repair, 2012a and Choo, Shafizadeh, and
Niemeier, 2007).
Moghadam and Livernois (2010), Bin (2003), and Washburn et al. (2001) all conducted
studies designed to develop a regression model capable of accurately predicting which vehicles
are most likely to fail an emissions inspection with the goal of allowing the majority of vehicle
owners whose vehicles consistently pass an emissions inspection to forego the inconvenience of
obtaining a required emissions inspection. The problem with the models developed in these
studies is that all of these models focus on predicting which vehicles are more likely to fail an
emissions test based upon emissions readings. California’s smog check program is changing,
reportedly in late 2013, to a testing program consisting only of a visual inspection and a scan of
the vehicle’s on-board diagnostic system, referred to as the OBD II system. This change renders
invalid any failure prediction model based upon emissions readings and necessitates the
development of a model for predicting which vehicles may fail a smog inspection that does not
utilize emissions measurements.
17
Moghadam & Livernois (2010) determined the current emissions inspection model
utilized in Toronto, Canada achieves the greatest levels of emissions reductions; however, the
program is not efficient from a cost standpoint. The researchers conducted a multi-leveled
analysis in order to arrive at this determination. First, they conducted a probability analysis to
determine at what point in a vehicle’s lifespan it would fall out of compliance with emissions
standards. The researchers adjusted this value depending upon whether the vehicle had undergone
emissions related repairs. Next, the researchers applied cost estimation functions to the level of
emissions reduction achieved per dollar spent. Finally, utilizing the above data, the researchers
performed a calculation to determine the lifetime (vehicle life span) costs of maintaining
emissions compliance. Based upon their findings, they argue that a 2% reduction in the goal for
attainment of emissions reductions would result in a 10 – 13% reduction in costs of the program.
However, there probably exists little political support for a program that reduces the goals for
pollution reduction. This research is important to the current study because the researchers
determined that the prevailing theory of focusing inspections on the oldest vehicles in the fleet is
not always the most effective approach because these vehicles often do not experience as much
use in their remaining years of usage. In fact, the researchers found that by focusing on median
aged vehicles, society achieves far greater emissions reductions per dollar spent.
Bin (2003) also conducted a study aimed at determining which vehicles are most likely to
fail an emissions inspection. The cost savings to society achieved by only directing those vehicles
most likely to fail an inspection for testing and excluding all other vehicles from testing
requirements served as Bin’s motivating factor in his research. Bin determined there is a positive
correlation between the emissions levels of a vehicle and the vehicle’s age, number of miles on
the vehicle, smaller engine sizes, and certain vehicle manufacturers. The study found that age was
the most significant of the variables with a one-year increase in the vehicle’s age increasing the
18
likelihood of an emissions failure by 1 to 2%. This study is relevant to this research because it
confirms that vehicle age affects emissions levels, supporting the need to use year of manufacture
as an explanatory variable in the analysis.
Washburn, Seet, and Mannering (2001) also conducted a study to identify vehicles likely
to fail an emissions inspection; however, their study found, at a statistically significant level, that
increased age and mileage will cause a decrease in a vehicle’s emissions levels. This is
inconsistent with most other studies that generally conclude increased vehicle age and odometer
reading cause an increase in emissions levels (Bin, 2003; Wenzel and Ross, 2003; Moghadam and
Livernois, 2010). Simply stated, prevailing theory dictates that as a vehicle ages in terms of both
years and miles accumulated, the emissions levels will increase; yet, Washburn, Seet, and
Mannering produced results stating the opposite.
Additionally, Wenzel and Ross (2003) published a comment on the Washburn et al.
study and suggested that in the development of their model, Washburn may have focused on the
wrong vehicle parameters by utilizing emissions values at an idle and misrepresented certain
principles of emissions testing by utilizing certain measured exhaust gasses as explanatory
variables for other exhaust gasses. Wenzel and Ross (2003) were correct in their observation that
Washburn, Seet, and Mannering’s use of emissions values at an idle was incorrect because of the
fact that vehicle emissions are nearly always lower at an idle because of the relatively small load
(amount of work the engine must perform) placed upon the engine. In addition, Washburn, Seet,
and Mannering (2001) utilize certain emissions values as explanatory variables for other
emissions values labeled as dependent variables. As an example, Washburn, Seet, and Mattering
list carbon dioxide as an explanatory variable for carbon monoxide. Both gasses are products of
the combustion process and not causal of one-another. This specification error is problematic in
that although the five emissions values measured during an emissions inspection are interrelated
19
and will increase or decrease in expected patterns, they are all the result of vehicle combustion
and not causal of one-another. The two errors discussed above likely resulted in the inconsistent
results obtained from the study.
The published studies present effective models at predicting which vehicles are likely to
fail emissions tests consisting of measurements of vehicle emissions; however, the fact that
California will no longer be utilizing tailpipe emissions to certify vehicles as emissions compliant
necessitates a different model for predicting failures. The studies above are useful in
demonstrating that regression analysis is the appropriate tool for identifying these vehicles,
provided the researchers select the appropriate variables. This supports the research conducted in
this analysis.
On-Board Diagnostics, Generation Two (OBD II) Systems
The current smog check inspection program consists of three parts: a visual inspection of
the emissions components on the vehicle, an emissions measurement of the exhaust gasses of the
vehicle, and a functional test of certain vehicle components (Bureau of Automotive Repair,
2009). One of the functional tests performed during a smog check inspection on model-year 1996
and newer vehicles is a scan of the on-board diagnostics, generation two (OBD II) system
(Bureau of Automotive Repair, 2009). The OBD II system is a computer based electronic control
system that controls engine, transmission, and emissions control system functions to maintain
proper engine efficiency and achieve required emissions reductions (Sosnowski and Gardetto,
2001). Currently, 33 states or municipalities utilize an emissions inspection program consisting
wholly or partly of an OBD II inspection (Environmental Protection Agency, 2013).
By the 1990’s, state and federal regulators had begun enacting regulations requiring
automobile manufacturers to utilize on-board diagnostic systems to control the function of the
vehicle’s engine. California implemented such a requirement by mandating that beginning with
20
model-year 1991 vehicles, all new cars sold in the State would have on-board diagnostics, first
generation, (OBD I) systems (Air Resources Board, 2009). At the federal level, the Clean Air Act
Amendments of 1990 mandated improvements to on-board diagnostic systems, including
standardization between manufacturers, and that these improvements be incorporated into all
passenger cars and light trucks beginning with model-year 1996 (Sosnowski and Gardetto, 2001).
Eventually, experts attached the label of "OBD II" systems to these improved, standardized onboard diagnostic systems (Air Resources Board, 2009 and Sosnowski and Gardetto, 2001).
California followed the federal lead and also implemented a requirement for OBD II systems
beginning with model-year 1996 (Air Resources Board, 2009). OBD II systems are still in use
today and have undergone numerous improvements in functionality (Lyons and McCarthy, 2009).
This supports the use of the system in evaluating the performance of vehicle emissions control
systems.
OBD II systems control the operation of the vehicle’s engine, transmission, and
emissions control systems to obtain the greatest efficiency while producing the lowest possible
emissions levels. The system utilizes a series of sensors to monitor numerous engine parameters.
The readings from these sensors are transmitted to the electronic control unit which processes the
information and determines how much fuel to supply to the engine and at what point during the
engine rotation to ignite the fuel by creating a spark in the ignition system which travels through a
spark plug in the engine cylinder, thereby igniting the fuel in the cylinder which generates heat
and expansion, creating power. Additionally, the OBD II system determines when to activate
various emissions components present on a given vehicle. (Sosnowski and Gardetto, 2001)
Another function of the OBD II system is to perform self-tests of the system and related
components. These self-tests are referred to as monitors and are performed automatically during
vehicle operation when certain conditions are met. If a failure is detected during the completion
21
of these monitors, the OBD II system may generate a diagnostic trouble code (DTC) and may
turn on the malfunction indicator lamp, commonly referred to as the ‘check engine’ light, on the
vehicle dash display. By regulation, the OBD II system is required to generate a DTC and
illuminate the malfunction indicator lamp if a fault is detected that may cause the vehicle to
produce 150% of the specified emissions level for that vehicle.(Sosnowski and Gardetto, 2001)
The OBD II functional test portion of the California smog check inspection consists of
scanning the OBD II computer to determine if all of the monitors are complete. If the monitors
are complete, the assumption is that the values generated during the functional test are an accurate
representation of the engine and emissions system’s current state of health. In addition to
determining monitor status, the functional test also requires the smog check technician to
determine if the malfunction indicator lamp illuminates when the key is turned on and turns off
when the engine is started. This is another indicator of the proper performance of the system.
Finally, the functional test scans the OBD II computer to determine if any DTCs are stored in the
system. If DTCs are stored in the computer, this is an indicator of a recent or pending
malfunction. Failure of any of these three steps will result in the vehicle failing the smog check
inspection. (Bureau of Automotive Repair, 2009)
Empirical Studies of OBD II Systems
Researchers have conducted empirical analysis of the functionality of OBD II systems.
Some of these studies sought to identify a qualitative method for determining the appropriate
design and components for the OBD II system. Others have sought to determine the long-term
accuracy of the system at identifying emissions failures. These studies demonstrate the validity of
applying statistical models to these systems.
The first study carried the goal of developing a set of models for use by engineers to
formulate the decision algorithms for OBD II systems to identify failures (Cascio, Console,
22
Guagliumi, Osella, Panati, Sottano, and Dupre, 1999). At the time of the research, engineers
manually created the algorithms for each application, which countered the EPA’s stated goal of
standardization. The researchers concluded they were able to develop specific algorithms for
OBD II failure determinations. The principles proposed by the researchers are valid because
although each vehicle manufacturer may use different structural components within their
emissions and engine control systems, the principles of operation are the same. As an example, an
ignition system failure in one cylinder of an engine will result in elevated emissions, regardless of
the type of ignition system employed. The specific algorithms developed by the researchers will
be useful in developing standardized models for the design of the diagnostic and control
structures of the systems in place on automobiles such as the fuel control system or the emissions
control systems, and identifying specific areas of fault within those systems (Cascio et al, 1999).
Because of the need to use the available computing power for the operation of the system, OBD II
systems in use in 1999 were only capable of identifying a specific portion of system operating
outside of expected parameters or a generalized fault. Research such as Cascio et al has led to
OBD II systems that are much more precise in identifying faults than earlier systems and are
improved in overall functionality (Lyons and McCarthy, 2009). This is important because
variations in technological advancements between automotive manufacturers will likely lead to
variation between manufacturers in failure rates of the OBD II functional portion of the smog
check inspection.
Barone conducted additional research at streamlining the design process of OBD II
systems in 2006. Barone’s stated goal was to develop a statistical method for determining which
subsystems to include in the OBD II system. His purpose for the research was to assist
automobile manufacturers in achieving the balance between government regulation and consumer
satisfaction (2006). Barone pointed out that an OBD II system that was overly sensitive at
23
identifying potential faults would frequently illuminate the malfunction indicator lamp, creating
customer satisfaction issues; while a system that was not sensitive enough would fail to meet the
requirements of government regulations (2006). Barone’s research provides a fundamental
statistical basis for the development of OBD II monitors, which are evaluated as a portion of the
OBD II functional portion of the smog check inspection. This is related to the current research
because of the fact that monitors are evaluated as part of the inspection and variations in the
function of the monitors may lead to certain vehicles exhibiting a higher likelihood of failure of
the inspection.
In 2011, Supnithadnaporn, Noonan, Samoylov, and Rodgers published a study that
detailed the results of an analysis they had conducted to determine the reliability of the emissions
inspection program in Atlanta, which utilizes only the on-board diagnostic scan for 1996 and
newer light duty vehicles. Their analysis compared the results of on-road emissions
measurements gathered for analytical purposes to the results of official inspections to predict the
probability of a vehicle passing the official inspection while in actuality the vehicle was emitting
excessive emissions. Supnithadnaporn et al. utilized data from an on-going Georgia Tech
Research Institute study of actual in-use vehicle emissions and compared that data to the results
from the on-board diagnostic scan inspection method. The in-use vehicle emissions data are
gathered through the use of remote sensing devices, which measure the emissions of vehicles
while the vehicles are in operation (Supnithadnaporn, Noonan, Samoylov, and Rodgers, 2011).
The study found that as vehicles age, they are 3.3% per year more likely to generate elevated
vehicle emissions while in use in spite of having achieved passing results during an on-board
diagnostic scan inspection (Supnithadnaporn, Noonan, Samoylov, and Rodgers, 2011). This
means vehicles that successfully pass an inspection consisting of only a scan of the OBD II
system may in fact still be generating excessive emissions; especially as those vehicles age. These
24
findings certainly raise concerns about the effectiveness of this methodology at identifying
excessively polluting vehicles.
The findings of Supnithadnaporn et al. are statistically significant at the p<0.001 level
(2011), which indicates a 99.9% probability that actual test results will match the results of their
findings. In spite of this statistical significance, there are concerns with the methodology of the
study. The major concern is the fact that the regression model compared the on-board diagnostic
scan results to the remote sensing data utilizing inspection results obtained after the collection of
the remote sensing data. This creates a specification error in the model in that dependent upon the
period between the events; it is possible for the vehicle to have been repaired, resulting in an
omitted variable.
Another concern with the Supnithadnaporn analysis is the model-year distribution of the
sample. The study, although published in 2011, only included 1996 through 2002 model-year
vehicles. Lyons & McCarthy (2009) indicate that the on-board diagnostic system on 1996 through
1999 model-year vehicles was not fully functional. This creates an additional error in that the
regression model does not account for this lack of functionality in the system. In addition to these
issues, Supnithadnaporn et al. (2011) incorrectly identify supporting information such as listing
that the on-board diagnostic systems were mandated to be installed beginning in the 1994 modelyear and that the system is capable of identifying if the fuel cap has been left off. Lyons &
McCarthy (2009) correctly indicate the systems were mandated beginning in 1996 and that
although the system is able to identify a vapor leak in the fuel system; it is not able to pinpoint a
cause, such as a missing fuel cap. Neither of these errors is as significant as the specification
errors and is not likely to affect the regression model.
The problems identified with the Supnithadnaporn study do not diminish its usefulness
for this research. The Bureau of Automotive Repair mitigated much of this concern by specifying
25
the new smog check inspection program applies to 2000 and newer vehicles only, likely in
response to the Lyons and McCarthy study. Consequently, the research conducted in this study is
relevant and necessary for identifying model-year 2000 and newer vehicles that are highly likely
to fail a smog check inspection consisting of only a visual inspection of the emissions
components and a functional test of the OBD II system.
Conclusion
The literature clearly demonstrates a need for a modern smog check inspection program
that takes advantage of the technology employed on modern vehicles. Additionally, the literature
outlines the statutory requirement of the Bureau of Automotive Repair to identify a certain
portion of the vehicles subject to smog check certification and direct those vehicles to specific
facilities for testing. The combination of these two factors identifies the need for a new model of
identifying and classifying vehicles as directed vehicles that accounts for a technology based
testing program.
The literature demonstrates that over the years, various research projects have sought to
create an effective model for identifying these directed vehicles. However, previous research fails
to identify a model capable of identifying vehicles likely to fail a smog check inspection that
consists of only a visual inspection and scan of the OBD II diagnostic system. The functionality
and reliability of the OBD II system as described in the literature supports the development of a
model that evaluates the data collected from the OBD II functional test portion of the current
smog check inspection procedure and operationalizing that data for the purposes of identifying
vehicles highly likely to fail the imminent new smog check inspection procedure.
26
Chapter 3
METHODOLOGY
To answer the research question of which vehicles are highly likely to fail the OBD II
functional portion and the visual inspection portion of the current smog check inspection
procedure, I used binomial logistic regression. Binomial logistic regression was the best method
to answer this question because of the fact that both dependent variables are binary variables.
Both OBD II functional test failure and visual inspection failure are yes (1) or no (0) possibilities.
Specifically, I identified the dependent variables as follows: the dependent variable for the
regression to identify vehicles highly likely to fail the OBD II functional test portion of the
current smog check inspection procedure was “FailOBD” and the dependent variable for the
regression model employed to identify vehicles highly likely to fail the visual inspection portion
of the current smog check inspection procedure was “FailVisual”. Table 2 below details the
number and percentage of vehicles failing each of these two portions of the current procedure.
Table 2
Distribution of Failing Vehicles in Current Inspection Procedure
Dependent Variable
(Component of Current
Smog Check Inspection
Procedure)
Number of
Vehicles Failing
Number of
Vehicles Passing
Total Observations
OBD II Functional Test
Visual Inspection
389,398
54,950
6,894,829
7,229,277
7,284,227
7,284,277
At first glance, the percentages of vehicles failing the OBD II functional test and visual
inspection, 5.6% and 0.75% respectively, were lower than desirable for statistical validity.
However, evaluating the validity of these values requires consideration of two important factors.
First, the overall failure rate for the smog check inspection program in the 2012 calendar year was
27
13.1% (Bureau of Automotive Repair, 2013). As such, the values are consistent with overall
program results. Second, and most importantly, the purpose of this thesis is to identify which
vehicles are most likely to fail when compared to other vehicles subject to inspection.
Consequently, the regression model compared each of the failure groups and identified the
vehicles within each group that are most likely to fail the test or inspection. These facts supported
the use of the binomial regression analysis with the current dataset to answer the primary thesis
question.
The remainder of this chapter discusses the data used in the analysis, the coding of the
data, and the process utilized in selecting variables. Finally, I conclude with a description of the
regression models used to answer the research questions.
Smog Check Test Data
The primary data source for this thesis is the Smog Check Test Record data. These data
are required to be maintained by the California State Bureau of Automotive Repair pursuant to
section 44024.5 of the California Health and Safety Code. I requested and obtained the test data
for the nearly 13.3 million smog check inspections performed in the state during the 2012
calendar year. Each inspection is considered one observation and each observation consists of 152
categories of information related to vehicle characteristics such as engine size and manufacturer;
inspection characteristics such as the technician performing the inspection and the date and time
of the inspection; weather conditions at the time of inspection; the results of the inspection; and
other information related to the inspection performed on each vehicle.
Pursuant to public disclosure laws prohibiting the release of personal identifying
information, the Bureau redacted the vehicle license plate numbers and vehicle identification
numbers from the dataset prior to providing me with the data. This limited my ability to identify
and remove duplicate tests performed on the same vehicle in the dataset. A vehicle receives
28
multiple tests in the same calendar year under two circumstances. First, the vehicle fails the initial
test, is repaired, and then retested for certification. In limited instances, the vehicle may receive
multiple retests. The second circumstance is change of ownership during the same calendar year
the vehicle was tested for biennial registration purposes. Both of these circumstances potentially
may skew the results, with the first increasing the number of test failures for vehicles in that
category and the second decreasing the number of failures for vehicles in that category. However,
unless only a select group of vehicles fall within these circumstances, the size of the dataset
overcomes any skewed values caused by these circumstances.
The 2012 Smog Check Test Record contains data for all inspections performed, including
those not subjected to the new inspection procedures (model-year 1999 and older vehicles).
Consequently, I deleted all vehicles not subject to the new inspection procedures from the dataset,
leaving approximately 7.9 million observations relevant to only model-year 2000 and newer
vehicles. From the 152 categories of information and based on the reviewed literature, I selected
only those items capable of identifying those vehicles highly likely to fail the OBD II functional
test portion or the visual inspection portion of the current smog check inspection procedure.
Regression Models: OBD II Test Failure
The first set of regression models sought to predict which vehicles or vehicle
combinations of make, model-year, model, engine size, and transmission type are most likely to
fail the OBD II test. Problems such as manufacturing defects, computer diagnostic strategies, and
defective components can all lead to an increase in the likelihood of a vehicle in any or all of the
categories above failing the OBD II functional test.
Additionally, although the federal and state requirements for OBD II system functionality
vary somewhat, both require that the system identify faults that may cause an increase in vehicle
emissions. Consequently, and as the body of literature indicates, each vehicle manufacturer is free
29
to develop specific criteria and algorithms for the OBD II system to follow when performing selfdiagnostic tests aimed at identifying potential emissions failures. This leads to a variation
between manufacturers in the functionality and reliability of the OBD II system. Often, there will
even be differences in functionality and reliability between model-years of the same
manufacturer, as exhibited by the OBD II reference guide the Bureau provides to industry
(Bureau of Automotive Repair, 2010b).
This variation in design and functionality creates the potential that certain vehicles will
demonstrate a higher likelihood of failing the OBD II portion of the smog check. In order to
address this issue, it was necessary for me to not only employ categorical variables identifying
each vehicle manufacturer as an individual categorical variable, I also needed to include
individual categorical variables that identify each vehicle in detail; specifically model-year,
manufacturer (make), and engine size, among other characteristics. This led to a large number of
explanatory variables; however, this was necessary to specifically identify, with statistical
validity, vehicles likely to fail the OBD II functional test. Consideration of these factors led to the
following functional form for the first set of regression models:
Failure of the OBD II Functional Test (FailOBD) = f (Vehicle Design
Characteristics, Vehicle Manufacturer, Year of Manufacture)
Previously, I discussed the fact that the dataset contains 152 different categories of
information gathered during the smog check inspection procedure. Of these, three were necessary
to create the explanatory variables needed to operationalize the three causal categories above.
These three categories were VLT REC NO, recoded to evaluate Vehicle Design Characteristics;
VEH MODEL YR, recoded into categorical variables to quantify the manufacturer-designated
vehicle Model-Year; and VEH MAKE, which I separated into individual categorical variables to
describe Vehicle Manufacturer. I discuss the process used to create these variables below.
30
The most important of the explanatory variables from the base dataset was the “VLT
REC NO”, which is used to describe Vehicle Design Characteristics. The Vehicle Look-Up Table
(VLT) Record (Row) Number (“VLT ID”) identifies the manufacturer of the vehicle (vehicle
make), the model-year, the vehicle model, engine size, transmission type, and other factors
specific to that vehicle and provides the greatest level of detail about each vehicle.. The Vehicle
Look-Up Table is a database programmed into each emissions inspection system throughout the
state and is maintained and updated by the Bureau on a regular basis (Bureau of Automotive
Repair, 2009). Each specific vehicle configuration, with some exceptions, receives a unique VLT
ID used to specifically identify that vehicle. By using each VLT ID as a unique and separate
explanatory variable, the analysis will identify specific vehicle configurations likely to fail the
new smog check inspection procedures.
The specific numerical VLT IDs within the table extend from 00001 to 53084, indicating
that there are potentially 53084 specific vehicle configurations identified in the VLT. However,
after removing incomplete and invalid observations, the current dataset only contained 2925 VLT
ID numbers accounting for approximately 7.5 million observations. There are three probable
reasons for only 2925 of 53084 VLT ID numbers remaining in the dataset. First, since the
inception of the smog check program in the 1980s, this is the method utilized for identifying
vehicles in the program. Consequently, not all vehicle configurations remain in the program.
Second, the Bureau frequently updates these identification numbers to identify new vehicle
configurations; however, some of the 53084 potential numbers are not in use as of yet. Finally,
filtering and cleaning of the dataset likely resulted in the deletion of some VLT ID numbers.
In addition to removing observations that were incomplete or invalid, I also deleted
observations accounting for incomplete tests, aborted tests, or other irregularities causing a
flawed record. The final number of observations for the dataset is 7,284,277. The most prevalent
31
of the vehicle specific ID numbers was VLT ID 29434, which is a 2000 Honda with a 1.6 liter,
four-cylinder engine. This vehicle accounts for nearly 50,000 of the smog check inspections
performed in 2012. I used this dataset for all regression models in the current study. Appendix B
contains a detail of the distribution of the VLT ID numbers in the dataset.
As stated above, nearly all of the VLT ID numbers apply to a specific vehicle
configuration. However, there are some exceptions to this statement in that in certain situations,
vehicles manufactured by a different corporation but with similar engine configurations, vehicle
weight, identical model-years, and other categories were grouped together under one VLT ID
number. An example of this is VLT ID 2270, which accounts for 51,434 observations. This VLT
ID number encompasses model-year 2008 compact vehicles with four cylinder engines and
manufactured by several different companies; including Honda, Toyota, Chevrolet, Nissan, and
others. This grouping within the same VLT ID number of vehicles manufactured by different
companies accounts for approximately 900,000 observations. I have accounted for this
circumstance by running a second binomial logistic regression analyses aimed at determining
which manufacturers within grouped VLT ID numbers are more likely than the other
manufacturers in that group to fail the OBD II test portion or visual inspection portion of the
current smog check inspection procedure.
In the first iterations of the regression analyses I conducted, I included vehicle make,
vehicle model-year, and VLT ID number as the explanatory variables. However, because vehicle
make is one of the factors identified in the VLT ID number, these iterations of the regression
model resulted in significant multicollinearity. Studenmund (2011) defines multicollinearity in
regression formulas as two or more variables that behave identical to one-another in the formula.
In essence, the variables are one in the same and are incapable of individual evaluation as to their
effect on the dependent variable. This is to be expected in the current analysis because of the fact
32
that vehicle make is a component of VLT ID numbers. As an example, it is quite possible for the
aforementioned VLT ID number 29434 as an explanatory variable to exert the same effect on the
dependent variable as Honda as an explanatory variable. To remedy this situation and to account
for grouped VLT ID numbers, I conducted three separate regression analyses on each dependent
variable.
The first regression analysis was a binomial regression with failure of the OBD II
functional test as the dependent variable and 2925 VLT ID numbers as the explanatory variables.
To account for the VLT ID numbers containing multiple manufacturers and to determine if a
specific vehicle manufacturer within a combined VLT ID is more likely than others to fail the
OBD II functional test; I completed a second regression analysis with individual vehicle makes as
the explanatory variables. Conducting the binomial regression with vehicle manufacturer as the
explanatory variable sorts the manufacturers by likelihood of failure and created the ability to
accurately classify each manufacturer within combined VLT ID numbers. There were 48
individual categorical variables created from the unique identifiers within the VEH MAKE
category from the primary dataset. Table 3 below details the distribution of the vehicle
manufacturers within the dataset.
Table 3
Distribution of Vehicle Manufacturers within the dataset
Manufacturer
Number of Observations
Percentage of Total Observations
ACURA
134,221
1.84
ASTON-MARTIN
114
0
AUDI
39,829
0.55
BENTLEY
92
0
BMW
230,074
3.16
BUICK
53,544
0.74
CADILLAC
76,206
1.05
CHEVROLET
819,246
11.25
CHRYSLER
181,412
2.49
33
DAEWOO
3,879
0.05
DODGE
363,085
4.98
FERRARI
325
0
FIAT
669
0.01
FORD
963,066
13.22
GMC
201,102
2.76
HONDA
796,354
10.93
HUMMER
12,500
0.17
HYUNDAI
122,244
1.68
INFINITI
80,954
1.11
ISUZU
9,034
0.12
JAGUAR
18,313
0.25
JEEP
134,006
1.84
KIA
81,529
1.12
LAMBORGHINI
48
0
LAND ROVER
24,374
0.33
LEXUS
197,881
2.72
LINCOLN
54,229
0.74
LOTUS
61
0
MASERATI
425
0.01
MAZDA
109,333
1.5
MERCEDES
189,849
2.61
MERCURY
34,258
0.47
MINI
23,060
0.32
MITSUBISHI
105,790
1.45
NISSAN
470,520
6.46
OLDSMOBILE
12,183
0.17
PLYMOUTH
5,633
0.08
PONTIAC
71,791
0.99
PORSCHE
20,590
0.28
SAAB
7,967
0.11
SATURN
87,776
1.21
SCION
69,453
0.95
SMART
512
0.01
SUBARU
59,938
0.82
SUZUKI
15,406
0.21
TOYOTA
1,159,923
15.92
VOLKSWAGEN
177,448
2.44
VOLVO
63,981
Total Observations
0.88
7,284,227
34
Finally, I conducted a third binomial regression analysis with the explanatory variables
generated from the VEH MODEL YR category of information from the 2012 smog check dataset.
Completion of this third regression analysis was necessary to provide further detail as to which
vehicles will likely fail the OBD II functional test portion of the current smog check inspection
procedure. The additional detail is necessary to clarify any discrepancies in the results of the
primary regression model conducted using VLT ID Numbers. From the VEH MODEL YR
category of information, I created 12 individual categorical variables, one for each model-year
between 2000 and 2012. Table 4 below details the distribution of vehicle model-years within the
primary dataset.
Table 4
Distribution of Vehicle Model-Years within the Dataset
Vehicle Model Year
Number of Observations
Percentage of Total Observations
2000
1,122,911
15.42
2001
556,844
7.64
2002
1,246,385
17.11
2003
508,263
6.98
2004
1,353,929
18.59
2005
417,171
5.73
2006
1,432,545
19.67
2007
210,009
2.88
2008
190,003
2.61
2009
63,743
0.88
2010
71,235
0.98
2011
73,767
1.01
2012
37,422
0.51
Total Observations
7,284,227
Regression Models: Visual Inspection Failure
The second dependent variable was “FailVisual”, which is the overall pass or fail result
for the visual inspection portion of the current smog check inspection procedure. The inclusion of
35
a regression model to identify vehicles likely to fail to the visual inspection portion of the smog
check inspection procedure was aimed at determining whether certain vehicles are likely to
undergo modifications or incur deterioration of components that will cause failure of this
inspection. The predominant reason for a failure of the visual inspection is modification of
emission components with the goal of improving the performance, often associated with ‘street
racing’, the illegal modification and racing of vehicles operated on public roadways. If certain
vehicles are more prone to modification, it is appropriate to direct those vehicles for specialized
testing procedures. The functional form for this regression model was as follows:
Visual Inspection Failure (FailVisual) = f (Vehicle Design Characteristics, Vehicle
Manufacturer, Year of Manufacture)
For the second regression analysis evaluating the likelihood of a vehicle failing the visual
inspection portion of the current smog check inspection procedure, I utilized the identical final
dataset and sets of explanatory variables (See Tables 2, 3 and Appendix B). Additionally, I
performed additional binomial logistic regression analyses on the dependent variable of failure of
the visual inspection portion of the current smog check inspection procedures. This was for the
same reason as performing additional regressions on the failure of the OBD II functional test, to
determine if specific manufactures within the VLT ID numbers that contain multiple
manufacturers result in a higher likelihood of failure of the visual inspection and to further define
the results of the analyses.
The methodology and categorization described was the best approach for answering the
research question. In the next chapter, I detail the results of the regression analyses just described.
36
Chapter 4
ANALYSIS
Completion of the regression analyses as described in Chapter 3 provided significant and
meaningful results. The results accurately identify a specific group of vehicles significantly more
likely to fail a smog check inspection consisting of a scan of the vehicle’s OBD II diagnostic
system and a visual inspection of the vehicle emission control devices. In review, I designed two
separate but similar groups of regression models to identify which vehicles are highly likely to
fail the revised smog check inspection procedure scheduled for implementation by the Bureau of
Automotive Repair in 2014. One group of regression models focuses upon identifying vehicles
likely to fail a scan of the OBD II diagnostic system present on nearly all passenger cars and light
trucks produced for sale in the United States since 1996. The second group of regression models
seeks to identify vehicles likely to fail the visual inspection portion of the smog check inspection.
I conducted binomial logistic regressions with the results reported as “odds ratios”. This
reporting method lists the coefficient for each explanatory variable as a prediction of the
likelihood of an occurrence. In this reporting method, a coefficient of 1.000 is the base value,
meaning the event or effect described by a variable with this value is equal to the event or effect
described by the control variable. A variable with a coefficient value of 0.500 is 50% less likely
to take place or exerts 50% less of an effect on the dependent variable as compared to the control
variable. Finally, a variable with a coefficient value of 2.000 is 100% more likely to take place or
exerts 100% greater of an effect on the dependent variable as compared to the control variable.
Although this reporting method enables me to predict the likelihood of a vehicle failing
the inspection as compared to a reference vehicle, the goal of this thesis is to predict which
vehicles are most likely to fail. Consequently, the result tables included later in this thesis and in
37
the appendices attached display likelihood of failure expressed as odds ratios with the higher odds
ratio values indicating a higher likelihood of failure.
It is important to note that I originally completed a fourth regression model wherein I
combined the vehicle make with the vehicle model year, 2002 Chevrolet as an example.
However, this methodology resulted in substantial collinearity, resulting in the omission of a
large number of variables by the statistical software and compromising the validity of such an
approach. Additionally, the Vehicle Look-Up Table identification numbers previously described
provide far greater detail for analysis. The next paragraphs detail the results of the regression
analyses.
Predicting OBD II Test Failures
I.
First Regression: Fail OBD = f (VLT ID Numbers)
The three regressions I conducted to identify vehicles highly likely to fail the OBD II
functional test portion of the current smog check inspection procedure, and subsequently the
revised smog check inspection procedure consisting of a scan of the OBD II system and visual
inspection, utilized failure of the OBD II functional test as the dependent variable. For the first of
the three, the explanatory variables were the Vehicle Look-Up Table identification numbers. The
final cleaned version of the VLT ID numbers consisted of 2925 individual identifiers (Appendix
B). During previous iterations of the regression model I conducted in determining the preferred
method for conducting this analysis; I discovered Stata, the statistical software used in the
analysis, automatically selected the last explanatory variable as the control variable. This takes
place regardless of whether I independently selected a control variable. Consequently, I allowed
Stata to select the control variable, which in this regression model was VLT ID number 53079.
This VLT ID number applies to a 2011 Chevrolet Silverado 2500 pick-up truck with a 6.6-liter
diesel engine and either a manual or an automatic transmission. There were 657 observations
38
associated with this vehicle, accounting for 0.01% of the nearly 7.3 million observations. I
compared the remaining 2924 VLT ID numbers to this VLT ID.
Completion of the regression model provided meaningful results. Exactly 116 of the VLT
ID numbers resulted in a lower likelihood of failing the OBD II functional test as compared to
VLT ID 53079, a 2011 Chevrolet Silverado. Of the remaining group, 18 of VLT ID numbers
demonstrated zero failures, meaning none of these vehicles resulted in a failure of the OBD II
functional test. Finally, 2790 of the VLT ID numbers demonstrated a higher likelihood of failing
the OBD II functional test as compared to a 2011 Chevrolet Silverado pick-up truck.
The 18 VLT ID numbers that demonstrated no failures of the OBD II functional test are
all 2005 and newer vehicles. In theory, these vehicles will likely demonstrate future failures as
they age. It is interesting to note that of the vehicles demonstrating zero failures the Toyota
Corporation (Toyota and Lexus) manufactures 10 of the 18 vehicles, or 56%, while at the same
time these makes only account for 18.7% of the dataset. Table 5 below provides information for
these vehicles.
Table 5
List of Vehicles Demonstrating Zero OBD II Functional Test Failures
VLT ID
2487
33036
33235
33696
33768
34025
34081
34351
34359
34366
34370
34549
Model-Year
2012
2005
2005
2006
2006
2007
2007
2007
2007
2007
2007
2007
Make
Volvo
Lexus
Toyota
Lexus
Mercury
Cadillac
Chevrolet
Land Rover
Lexus
Lexus
Lexus
Toyota
Model
S 60
LS 430
Sequoia 4WD
SC 430
Mariner
DTS
K1500 Suburban
Range Rover
GS 350
LS 460
RX 350 4WD
4 Runner 4WD
Engine Size
2.5 liter
4.3 liter
4.7 liter
4.3 liter
3.0 liter
4.6 liter
5.3 liter
4.4 liter
3.5 liter
4.6 liter
3.5 liter
4.0 liter
39
34570
34580
34583
34597
52982
53038
2007
2007
2007
2007
2010
2010
Toyota
Toyota
Toyota
Volkswagen
BMW
Ford
Rav 4 2WD
Tacoma 4WD
Tundra 2WD
Rabbit
X5
F350 Diesel
2.4 liter
4.0 liter
4.7 liter
2.5 liter
3.0 liter
6.4 liter
Beyond the vehicles that demonstrated zero failures of the OBD II functional test, the
actual odds ratios for the remaining variables, while significant and meaningful, are less
important than the rank of each VLT ID number as compared to the remaining ID numbers. As
stated throughout this thesis, the goal of the current research is to predict which vehicles are most
likely to fail the OBD II functional test portion of the current smog check inspection procedure.
Consequently, the following discussion and analysis of the regression results focuses on the
distribution of the VLT ID numbers within the regression model.
The mean odds ratio value for the regression analysis was 10.981. In determining which
vehicles are more likely to fail the OBD II functional test, any VLT ID number with an odds ratio
above this value has a greater than average likelihood of failing the test. Reviewing the results of
the analysis as detailed in Appendix C shows that 1636 VLT ID numbers were below this mean
value, leaving 1288 VLT ID numbers with odds ratio values above the mean value. These 1288
VLT ID numbers account for 36.92% of the total observations in the dataset. The first of the
vehicles identified by this group of VLT ID numbers range is a 2004 Toyota Tundra with a 3.4
liter engine and two-wheel drive. The vehicle identified by this VLT ID number is only slightly
above average in terms of likelihood of failing the OBD II functional test with an odds ratio of
10.986. The vehicle identified by the VLT ID number most likely to fail the OBD II functional
test is a 2001 Mazda MPV mini-van with a 2.5-liter engine. This vehicle is very highly likely to
fail the OBD II functional test with a reported odds ratio for this VLT ID number of 43.542.
40
Appendix C contains a detailed list of the VLT ID numbers and associated vehicles with odds
ratio values greater than the mean.
The binomial logistic regression with VLT ID numbers as the explanatory variables
predicting which vehicles are highly likely to fail the OBD II functional test was very effective at
identifying a group of vehicles likely to cause a failure of the test. This group of VLT ID
numbers, especially those with the highest odds ratio values, identifies a very specific group of
vehicles that are appropriate for identification as directed vehicles.
II.
Second Regression: Fail OBD = f (Vehicle Make)
The second of the three regressions utilizing failure of the OBD II functional test portion
of the current smog check inspection procedure encompassed vehicle make (manufacturer) as the
explanatory variables. The regression model with VLT ID numbers as the explanatory variables
resulted in the identification of approximately 65 VLT ID numbers that are attributable to
multiple vehicle manufacturers. This made it necessary to run this second regression to determine
if any manufacturer, or group of manufacturers, is more likely than the others to fail the OBD II
functional test. These data enabled me to determine if specific manufacturers within a combined
VLT ID number are more likely than the other manufacturers to fail. The next paragraphs discuss
the results of this regression.
There were 48 individual vehicle makes remaining in the filtered and cleaned dataset.
Completion of the regression resulted in all tests performed on vehicles manufactured by
Lamborghini passing the OBD II functional test. Every other manufacturer had at least one
observation resulting in failure of the OBD II functional test. Additionally, the regression
software automatically selected a comparison variable for the regression model. In this model,
Stata omitted the manufacturer Volvo as the comparison variable. . In this regression, the mean
odds ratio was 1.062, meaning those manufacturers with a higher value than 1.062 have a higher
41
likelihood of failure of the OBD II functional test as compared to the others. Table 6 below
details the results of this regression model.
Table 6
Results of Regression Model with Vehicle Make as the Explanatory Variables
Vehicle Make
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
FIAT
ASTON-MARTIN
LOTUS
SMART
LEXUS
MINI
MASERATI
PORSCHE
ACURA
INFINITI
HONDA
SCION
SAAB
TOYOTA
BENTLEY
MERCEDES
SUBARU
LAMBORGHINI
BMW
FORD
NISSAN
LINCOLN
AUDI
JAGUAR
Odds
Ratio
0.054
0.159
0.300
0.359
0.441
0.444
0.568
0.578
0.582
0.636
0.641
0.645
0.685
0.796
0.819
0.881
0.986
Omitted
1.019
1.027
1.039
1.044
1.089
1.096
Vehicle Make
25
CADILLAC
26
BUICK
27
MERCURY
28
JEEP
29
GMC
30
FERRARI
31
HUMMER
32
HYUNDAI
33
MAZDA
34 CHEVROLET
35 LAND ROVER
36
SATURN
37 VOLKSWAGEN
38
CHRYSLER
39
DODGE
40
PONTIAC
41
KIA
42 MITSUBISHI
43
PLYMOUTH
44
SUZUKI
45
ISUZU
46 OLDSMOBILE
47
DAEWOO
Mean Odds Ratio
Odds
Ratio
1.103
1.121
1.123
1.155
1.168
1.181
1.222
1.235
1.265
1.295
1.326
1.331
1.351
1.394
1.430
1.521
1.556
1.631
1.762
1.795
1.847
1.872
3.329
1.062
As Table 6 shows, the results of this regression model are evenly distributed. The mean
odds ratio value of 1.062 is very close to the base value of 1.000 for binomial regression.
Additionally, the reported odds ratio values for all but one of the manufacturers range from 0.054,
42
which is 95% less likely than a Volvo to fail the OBD II functional test to a high of 1.872, which
is 87% more likely than a Volvo to fail the OBD II functional test. The only exception to this is
the manufacturer Daewoo, which is 3.3 times more likely than the Volvo to fail the OBD II
functional test.
This regression model is very effective at showing that, with the exception of Daewoo;
the vehicles are nearly identical in terms of likelihood of failure of the OBD II functional test.
Consequently, only Daewoo vehicles contained within any of the combined VLT ID numbers
have a greater likelihood of failure of the OBD II functional test. The remaining manufacturers,
while showing noticeable differences in the results tables, cannot be labeled as “highly likely” to
fail the OBD II functional test portion of the current smog check inspection procedure.
III.
Third Regression: Fail OBD = f (Model-Year)
Finally, the third regression with failure of the OBD II functional portion of the current
smog check inspection procedure as the dependent variable encompasses model-year as the
explanatory variables. As a reminder, the model-year is identified by the manufacturer and is not
necessarily the year of production. As an example, many 2012 model-year vehicles were
manufactured in July or August, and occasionally earlier months, of 2011. This circumstance has
always existed in automobile manufacturing.
Given the current study evaluated model-year 2000 and newer vehicles, the regression
model consisted of 13 explanatory variables; model-year 2000 through model-year 2012. As
always, Stata dropped the last explanatory variable as the control variable. In this instance, this
was model-year 2012. The distribution of odds ratios details the expected results. The older a
vehicle is, the more likely it is to fail the OBD II functional test. Model-years 2007 through 2011
are least likely to fail the test, while model-years 2000 through 2006 are clearly more likely to fail
43
the test. For this regression, I did not include the mean odds ratio of 4.459 as the results in Table
7 clearly demonstrate the model-years most likely to fail the OBD II functional test.
Table 7
Odds Ratios for
Vehicle Model-Years
ModelYear
MY2011
MY2010
MY2009
MY2008
MY2007
MY2006
MY2004
MY2005
MY2002
MY2003
MY2000
MY2001
Odds
Ratio
0.948
1.023
1.550
1.640
1.818
2.908
4.376
4.610
7.415
7.549
7.551
12.120
It is important to note that in the first six model-years, the
odd year is more likely to fail than the even year; despite the fact the
even year vehicles are one year older. This is likely because the
smog check is required every other year once a vehicle is six years
old or when a transfer of ownership occurs. Consequently, in even
calendar years, even model-year vehicles are undergoing
inspection as a requirement for renewal of registration while odd
model-year vehicles are undergoing an inspection primarily for
change of ownership.
The dataset in this analysis is for the 2012 calendar year.
Consequently, it is reasonable to assume that a substantial portion
of the model-year 2001 vehicles received inspections for change of ownership purposes, and as
Table 7 shows, these vehicles are likely to fail the OBD II test at a significantly higher rate than
other model-years. The fact that vehicles inspected for change of ownership purposes
demonstrate a higher likelihood of failure is significant for the purposes of identifying directed
vehicles.
Predicting Visual Inspection Failures
I.
First Regression: Fail Visual = f (VLT ID Numbers)
As stated previously, I conducted two sets of regression analyses in determining which
vehicles are highly likely to fail the new smog check inspection procedure scheduled for
implementation in 2014. The first set of analyses sought to predict vehicles highly likely to fail
the OBD II functional test portion of the current smog check inspection procedure. The new
44
inspection procedure mirrors this test. The second set of three regression analyses seek to predict
which vehicles will fail the visual inspection portion of the current smog check inspection
procedure, which will mirror the new inspection procedure as well. All three of the regression
models in this group utilize “Fail Visual” as the dependent variable. Each of the three models
applies a set of explanatory variables against this variable to determine which of the explanatory
variables identifies vehicles highly likely to result in a visual inspection failure.
The first regression analysis in this group utilizes the same set of VLT ID numbers as
explanatory variables as utilized in the first OBD II regression model. Stata once again selected
VLT ID number 53079 as the control variable. As stated earlier, this VLT ID number applies to a
2011 Chevrolet Silverado 2500 pick-up truck with a 6.6-liter diesel engine and either a manual or
an automatic transmission. However, there are little similarities beyond the same control variable
for the two regression models utilizing VLT ID numbers as explanatory variables. Although the
comparison of the explanatory variables to one-another is again more important than the actual
odds ratio values, there is a substantial difference between the two models as far as this
comparison is concerned.
This regression model identified 2435 VLT ID numbers less likely to fail the visual
inspection as compared to the control variable, which is nearly the exact opposite of the results
with OBD II failure as the dependent variable. Additionally, there were 211 VLT ID numbers
demonstrating zero failures of the visual inspection as compared to only 18 demonstrating zero
failures in the OBD II functional test. Appendix D lists the VLT ID numbers and associated
vehicles that demonstrated zero failures of the visual inspection portion of the smog check
inspection in the 2012 calendar year.
As with the regression model with OBD II failure as the dependent variable, the actual
values of the odds ratios reported in this regression model are less important that the distribution
45
of the VLT ID numbers in the list. In the visual inspection failure model, the mean value of the
odds ratios was 0.624. The fact that the mean value is noticeably below the base value of 1.000
for logistic regression is likely because of the fact that, as discussed in Chapter 3 of this thesis,
only 0.75% of vehicles failed the visual inspection in the 2012 calendar year. This low failure rate
leads to the fact that the substantial majority of vehicles are unlikely to fail the visual inspection,
in turn causing the low mean value for the odds ratios of visual inspection failure.
The aforementioned circumstances led me to select only the VLT ID numbers that
demonstrated statistical significance in the regression results as those VLT ID numbers with a
higher likelihood of failure of the visual inspection portion of the smog check inspection
procedure. In the current analysis, there were 153 VLT ID numbers with statistically significant
results demonstrating a higher likelihood of failure of the visual inspection. These VLT ID
numbers account for 3.71% of the inspections performed in 2012. The lowest of these, but
measurably higher than the mean, was VLT ID number 50022 with a reported odds ratio of 1.605,
indicating this vehicle is nearly 100% more likely than the mean and 60% more likely than the
control variable to fail the visual inspection. VLT ID number 50022 is a grouped VLT ID number
that applies to heavy duty light trucks with diesel engines and built by domestic manufacturers
(Chevrolet, Dodge, GM, and Ford). The VLT ID number demonstrating the highest likelihood of
failure of the visual inspection is VLT ID 31158, with a reported odds ratio of 4.331. This VLT
ID number applies to a 2002 Volkswagen Cabrio with a 2.0-liter engine and manual transmission.
Appendix E provides details for the 153 VLT ID numbers demonstrating a significant likelihood
of failure of the visual inspection portion of the current smog check inspection procedure.
II.
Second Regression: Fail Visual = f (Vehicle Make)
As with the second regression model to identify vehicles highly likely to fail the OBD II
functional test, the second regression model to identify vehicles highly likely to fail the visual
46
inspection portion of the smog check inspection procedure utilizes vehicle make as the
explanatory variables. The set of explanatory variables consisted of the same 48 individual
vehicle manufacturers as the OBD II set, and Stata again dropped Volvo as the control variable.
Although there are similarities between the two regression models with different
dependent variables, there are also noticeable differences in the results of the regression analysis.
As with the fail OBD regression model, the manufacturer Lamborghini demonstrated zero
failures of the visual inspection. However, in this model, the manufacturers Aston Martin,
Bentley, Ferrari, Lotus, and Maserati also demonstrated zero failures of the visual inspection. The
fact that these are all high-end luxury car manufacturers is worth noting.
Another notable difference between the two models is that this regression model
demonstrated a greater number of manufacturers with a higher likelihood of failure of the visual
inspection. In spite of this greater number of manufacturers demonstrating a higher likelihood of
failure, the manufacturer Daewoo is once again the most likely manufacturer to fail the visual
inspection portion of the current smog check inspection procedure. However, the manufacturers
Pontiac, Mitsubishi, Dodge, Oldsmobile, Volkswagen, and Plymouth also demonstrate higher
odds ratio values, thus indicating a higher likelihood of failure. Table 8 below details the odds
ratio values generated in the regression analysis.
Table 8
Results of Regression Model with Vehicle Make as the Explanatory Variables
1
2
3
4
5
6
7
Vehicle Make
FIAT
PORSCHE
SMART
MERCEDES
LEXUS
LAND ROVER
MINI
Odds Ratio
0.237
0.285
0.310
0.409
0.413
0.581
0.677
25
26
27
28
29
30
31
Vehicle Make
HYUNDAI
GMC
ISUZU
SAAB
SCION
SATURN
BUICK
Odds Ratio
1.080
1.091
1.096
1.122
1.150
1.232
1.347
47
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
TOYOTA
BMW
HONDA
INFINITI
CADILLAC
JAGUAR
LINCOLN
ACURA
ASTONMARTIN
BENTLEY
FERARI
LAMBORGHINI
LOTUS
MASERATI
NISSAN
KIA
HUMMER
0.683
0.700
0.746
0.773
0.801
0.844
0.920
0.996
Omitted
Omitted
Omitted
Omitted
Omitted
Omitted
1.030
1.069
1.073
32
33
34
35
36
37
38
39
SUZUKI
AUDI
CHEVROLET
SUBARU
FORD
MAZDA
MERCURY
CHRYSLER
JEEP
40
PONTIAC
41
42 MITSUBISHI
DODGE
43
44 OLDSMOBILE
45 VOLKSWAGEN
PLYMOUTH
46
DAEWOO
47
Mean Odds Ratio
1.360
1.369
1.438
1.487
1.561
1.580
1.637
1.790
1.855
2.201
2.325
2.351
2.606
2.651
4.041
4.714
1.305
The results of the binomial logistic regression analysis clearly identify a set of vehicle
manufacturers with a higher likelihood of failing the visual inspection portion of the current smog
check inspection procedure. This information will again prove useful at identifying specific
vehicle makes within combined VLT ID numbers that are more likely to fail the visual inspection.
III.
Third Regression: Fail Visual = f (Model-Year)
The final regression model in the second group encompassed vehicle model-year as the
set of explanatory variables utilized to describe which vehicles are highly likely to fail the visual
inspection portion of the smog check inspection. Stata again dropped model-year 2012 from the
dataset as the control variable. Another similarity to the regression conducted to predict which
vehicles are highly likely to fail the OBD II inspection as explained by vehicle model-years is that
older vehicles are substantially more likely to fail the visual inspection portion of the smog check
48
inspection procedure. A final similarity is that over the first six model-years in the dataset, the
Table 9
Results of Regression
Model with ModelYear as the
Explanatory
Variables
ModelYear
MY2011
MY2010
MY2009
MY2008
MY2007
MY2006
MY2004
MY2005
MY2002
MY2003
MY2000
MY2001
Odds
Ratio
1.488311
3.400279
3.643585
4.523026
6.906138
9.775916
14.06943
15.0705
23.86911
25.09562
31.61048
32.40856
odd model-year vehicles exhibit a higher likelihood of failure
than the even model-year in spite of the fact that the even
model-year vehicles are one year older. Again, this supports the
theory that vehicles inspected for the purposes of change of ownership
are more likely to fail the inspection. One difference between these
regression results and those for the OBD II failure regression with
model-years as the dependent variable is the odds ratio values are
substantially higher in the visual inspection regression model. This
indicates that vehicle model-year is quite significant in terms of
predicting which vehicles are highly likely to fail the visual inspection
portion of the smog check inspection procedure. Table 9 details the
results of the regression analysis.
Reliability and Validity
In any study, the implementation of measures to insure the reliability and validity of the
conclusions drawn from the analysis is an important step. In order to draw valid conclusions from
a regression analysis, reliability of measurement is essential. According to Singleton and Straits
(2010), reliability of a measurement means whatever is under measurement in the analysis is
consistently and dependably measured. Validity means the predictions made by the analysis are a
true reflection of the potential actual outcome of the event (Singleton and Strait, 2010).
Reliability of measurement in regression analysis is generally a concern because most
regression analyses analyze a sample of data from a much larger dataset. However, this is not a
concern in this analysis because the model did not examine a sample of the smog check
49
inspection data. I specifically decided to apply the regression model to the entire dataset of smog
check inspections to overcome reliability concerns. Additionally, the large number of potential
explanatory variables required inclusion of all valid and applicable observations to ensure
sufficient data for each variable. Finally, because of the fact that a small portion of the vehicles in
the dataset received multiple inspections in the reporting period, inclusion of all possible
observations was necessary to overcome any skewed values created by the multiple inspections.
These factors cause me to have a high confidence level in the reliability of the dataset and
subsequent regression analyses.
The question asked in evaluating the validity of this model is whether it accurately
identified those vehicles with the highest likelihood of failing either the OBD II functional test or
the visual inspection portions of the current smog check inspection procedure. A potential cause
for failure of the model to accurately identify these vehicles is the omission of an important
variable or variables. In the current study, two potential omitted variables were vehicle condition
and testing behaviors. A poorly maintained vehicle may cause an unexpected failure of either
portion of the inspection procedure while similar vehicles are less likely to fail the inspection. At
the same time, a less-than-scrupulous smog check technician or station may, for whatever reason,
take steps to allow a vehicle to pass the inspection when it should not. Both of these conditions
can skew the results of the analysis but were addressed by the methodology selected.
In regard to the issue of an unscrupulous technician or station, the volume of observations
in the dataset sufficiently minimizes any risk to validity. There were 7,962 active smog check
stations in 2012 and 11,921 active technicians (Bureau of Automotive Repair, 2013). The average
percentage of tests in the dataset of 7,284,277 tests performed by each station was approximately
0.11% while the average percentage of tests performed by each technician was 0.16%.
Consequently, the likelihood of any technician or station performing a large number of tests on a
50
specific category of vehicle within the dataset is very minimal and constitutes an acceptable level
of risk.
Maintenance of vehicles is not a risk to the validity of the analysis simply because a
major goal of the smog check program is to encourage the improved maintenance of vehicles,
resulting in lower emissions (Eisinger, 2010). Smog check programs are often referred to as
“I/M” programs, meaning “inspection and maintenance” programs. The theory is that if
consumers are aware of the maintenance requirements that must be met for their vehicle to pass
the inspection, they are more likely to have the necessary repairs performed. Consequently, if a
vehicle is less likely to fail the inspection as a result the maintenance practices of the owner, that
is precisely the type of vehicle this study seeks to identify. Again, the risk to the validity of the
study created by this potential omitted variable is minimal.
Another common test for validity is hypothesis testing. Although none of the models
were seeking to prove or disprove a hypothesis, the principles of hypothesis testing still apply to
the models. Hypothesis testing employs the use of a calculated “t-statistic” or “z-test”, which are
compared to a chart containing a critical value for the t-statistic or a normal distribution chart for
the z-test. If the calculated value exceeds the critical value, the researcher must accept the
hypothesis proposed by the variable. For logistic regression, determination of validity of the
hypothesis encompasses the z-test. Using the regression model with OBD II failure as the
dependent variable and VLT ID numbers as the explanatory variables as an example, all 1288
VLT ID numbers above the mean value reported a z-value of greater than 4.30 while the normal
distribution value for the z-value is 3.013 at a significance of less than .01. This means there is a
99% probability that the actual failure rate of the vehicles will be within the range specified in the
analysis. In five of the six regression models conducted as part of this study, the explanatory
variables demonstrating the greatest likelihood of failing the OBD II functional test or the visual
51
inspection all demonstrated a significance level of greater than 99%. The only exception was the
model predicting failure of the visual inspection based upon VLT ID number.
As discussed above, there were 153 explanatory variables demonstrating statistical
significance in the model utilizing failure of the visual inspection as the dependent variable. Of
these, 50 were significant at the 99% level, 57 were significant at the 95% level, and 46% were
significant at the 90% level. This means that there is a 99%, 95%, and 90% probability,
respectively, that the actual failure rates of these vehicles fall within the range specified in the
analysis. Again, the low number of failures as a portion of the overall dataset leads to the low
number of statistically significant explanatory variables.
A final test for validity conducted on the regression models was the chi-squared test. This
test assesses the validity of the entire regression model by calculating the likelihood of results
generated by random chance. The lower the likelihood of random chance, the greater the validity
of the regression model as a whole (Adcock, 2010). The chi-squared calculation is similar to the
t-statistic and z-test in that the regression model calculates a chi-squared value to compare to a
value contained in the “Critical Values of Chi-squared” table. Obtaining the critical chi-squared
value requires identifying the intercept of rows labeled “degrees of freedom” and columns labeled
“probability”. The degrees of freedom equates to the number of valid explanatory variables. The
probability ranges from .001 (99.9% likelihood that the results are not random chance) to .10
(90% likelihood that the results are not random chance). For the current regression analyses, the
calculated chi-squared values ranged from 11569.80 to 183315.20, well above the critical chisquared values and indicating a 99.9% likelihood the results of each regression model were not a
result of random chance. Table 8 below details the chi-squared values for the current analysis.
52
Table 10
Detail of Chi-Squared Values for Regression Model
Dependent Variable
Explanatory
Variables
Degrees of Freedom
Critical Chi-squared
Value
Calculated Chisquared Value
Likelihood of Random
Chance
Fail OBD II Test
Vehicle
ModelVLT ID
Make
Year
2906
46
12
Fail Visual Inspection
Vehicle
ModelVLT ID
Make
Year
2173
41
12
149.45
86.66
32.91
149.45
80.08
32.91
183315.20
31335.43
93242.97
43818.82
11569.80
14038.44
< .001%
< .001%
< .001%
< .001%
< .001%
< .001%
As stated in the introduction to this section, the results of the regression models were
significant and meaningful. In the next chapter, I will detail the conclusions drawn from the
analysis and make recommendations for future policy actions.
53
Chapter 5
CONCLUSION
The regression analyses conducted in this thesis provide insight into alternative methods
of identifying vehicles highly likely to fail the California smog check inspection procedure. The
required inspection procedure is undergoing substantial change, with the elimination of the
emissions measurement portion of the inspection procedure for a large portion of the fleet being
the most significant change. One of the reasons for this thesis is the fact that the current model for
identifying vehicles, referred to as directed vehicles, highly likely to fail the smog check
inspection is predominately predicated on the emissions measurements obtained during the
inspection. Consequently, transition to a different inspection procedure requires transition to a
different model for identifying directed vehicles.
Historically, the Bureau of Automotive Repair has directed, for specialized testing,
approximately 30% of the entire fleet of vehicles subject to inspection. All vehicles, regardless of
model-year, were subject to evaluation for directed vehicle status. However, beginning in 2014,
vehicles model-year 2000 and newer will no longer be subject to an emissions measurement as
part of their smog check inspection. The inspection procedure for these vehicles will consist
solely of a scan of the vehicle’s on-board diagnostic (OBD II) system and a visual inspection of
the emission control devices on the vehicle. Consequently, the existing method for identifying
directed vehicles will no longer apply to this group of vehicles, which comprises nearly 60% of
the vehicles inspected each year, and this percentage grows each year. To say a system of
identifying vehicles for specialized testing by applying the model to less than half of the vehicles
subject to inspection lacks validity is an understatement at best.
In 2011, 1999 and older vehicles accounted for approximately 41% of all inspections and
in 2012, that number dropped to 35%. Based upon this pattern, in the very near future, the
54
number of these vehicles subject to testing will be less than 30%. Although the Bureau has
indicated the intention to identify all model-year 1999 and older vehicles as directed vehicles, the
progressively decreasing volume of these vehicles necessitates the development of a model
capable of identifying directed vehicles within the model-year 2000 and newer group
The regression models developed as part of this thesis proved highly effective at
identifying vehicles in the model-year 2000 and newer group with a higher propensity for failure
of the revised smog check inspection procedure scheduled for implementation in 2014. In the
remainder of the chapter, I discuss the findings of the analysis, opportunities for improvement to
the model, and recommendations for continued effectiveness of the practice of identifying
directed vehicles.
Predicting Likelihood of Failure of the OBD II Functional Test
The first set of regression models completed in support of this thesis sought to identify
vehicles highly likely to fail the OBD II functional test portion of the current smog check
inspection procedure. This test is equivalent to the diagnostic scan portion of the new smog check
inspection procedure; as such, vehicles highly likely to fail this portion of the current inspection
procedure are also likely to fail the new inspection procedure. The three regression models
separately utilized vehicle look-up table identification (VLT ID) numbers, vehicle make
(manufacturer), and vehicle model-year as explanatory variables with failure of the OBD II (Fail
OBD) as the dependent variable in each. VLT ID numbers provide the most precise detail about
the vehicles subject to inspection because, with a few exceptions, each VLT ID number is
assigned to vehicles of the same make, model, model-year, engine size, transmission type, and
other characteristics. The exceptions are a group of VLT ID numbers that group together vehicles
of a similar configuration but manufactured by different companies.
55
The model utilizing VLT ID numbers as the explanatory variables effectively identified a
group of 1288 vehicles that account for approximately 37% of the inspection volume and are
more likely to fail the OBD II functional test. The vehicles identified range from only slightly
more likely than other vehicles to fail the inspection to substantially more likely to fail the
inspection. Given the progressively decreasing volume of vehicles in the model-year 1999 and
older set, labeling the vehicles in the 2000 and newer set with the highest likelihood of failure as
directed vehicles creates the ability to maintain the level of directed vehicles at approximately
30%.
The results of the regression with vehicle-model year as the explanatory variables yielded
expected results in that older vehicles were significantly more likely to fail the OBD II functional
test. This is expected because as vehicles age and are driven more miles, they develop
malfunctions leading to emissions failures. A notable finding in this regression model is the
observation that odd model-year (2001, 2003, etc.) vehicles are apparently more likely to fail the
functional test than the next older even model-year vehicles. This is notable because vehicles that
are more than six model-years old are required to undergo biennial inspection. This is based upon
actual model-year, meaning that most model-year 2006 vehicles were required to undergo their
first smog check inspection in 2012. This pattern would continue with 2004, 2002, 2000 and soon model-year vehicles requiring a biennial inspection in 2012. Odd model-year vehicles, for the
most part, likely underwent inspections in 2012 for change of ownership purposes. This suggests
that vehicles undergoing a change of ownership are significantly more likely to fail an OBD II
functional test.
The regression model based on vehicle make yielded expected results as well. Only one
vehicle manufacturer, Daewoo, demonstrated a significantly higher likelihood of failing the OBD
56
II functional test. The other manufacturers represented in the dataset were all relatively close to
one another in likelihood of failure.
As stated in the introduction to this chapter, these results provide substantial foundation
for identifying which vehicles, when compared to the other vehicles subject to inspection, have a
higher likelihood of failing the OBD II functional test of the current inspection procedure and,
subsequently, the revised smog check inspection procedure predicated upon a diagnostic scan of
the OBD II system.
Predicting Likelihood of Failure of the Visual Inspection
The set of regression models to identify which vehicles are highly likely to fail the visual
inspection portion of the current smog check inspection procedure utilized the same sets of
explanatory variables as the previous set of models: VLT ID numbers, vehicle make, and vehicle
model-year. The dependent variable for this set of models was failure of the visual inspection
(Fail Visual) portion of the current smog check inspection procedure.
The model with VLT ID numbers as the explanatory variables produced meaningful
results as well. This model identified 153 VLT ID numbers with a higher likelihood of failing the
visual inspection portion of the current smog check inspection procedure. The vehicles identified
by this group account for nearly 4% of the volume of inspections in the dataset.
The regression model for visual failures with vehicle make as the explanatory variables
produced results similar to the same model for OBD II failure, with the exception that the
distribution of odds ratio values for the visual inspection was greater than that of the OBD II test
failure. This increased variance in the results points to variances of consistency in vehicle
durability or maintenance practices. Additionally, the manufacturer Daewoo again demonstrated
the highest likelihood of failure of the visual inspection.
57
Finally, the regression with vehicle model-years as the explanatory variables for visual
inspection failure also produced results similar to the OBD II regression models. Older vehicles
are significantly more likely to fail the visual inspection portion of the current smog check
inspection procedure, indicating likely failure of the visual inspection portion of the new smog
check inspection procedure.
Significance of Findings
It is important to point out that this thesis focused upon identifying vehicles highly likely
to fail the smog check inspection as compared to all other vehicles in the fleet and subject to
inspection. I did not seek to predict failure rates for individual VLT ID numbers, specific makes,
or individual calendar years. As such, although each of the regression models produced variables
with a significantly higher likelihood of failure, my focus was on the overall ranking as compared
to the other variables.
The binomial logistic regressions performed in this thesis clearly identified a valid
method for ranking vehicles based upon their likelihood of failing either the OBD II functional
test or visual inspection portion of the smog check inspection. Those vehicles identified by the
VLT ID numbers with the highest likelihood of failure are appropriate for designation as directed
vehicles to supplement the model-year 1999 and older group and to maintain a directed vehicle
set of approximately 30% of the vehicles subject to inspection.
There is less support in the data to employ vehicle manufacturer as a basis for
identification as directed vehicles. Although some manufacturers demonstrated a higher
likelihood of failure, the results are not significant enough to identify vehicles from any single
manufacturer as directed vehicles.
The data related to vehicle model-year clearly demonstrates that older vehicles are
significantly more likely to fail either test or inspection. However, the ability to identify a
58
sufficient number of specific vehicles through the use of VLT ID numbers is far more equitable
and appropriate than a blanket policy of directing all older vehicles for specialized testing.
Additionally, as the body of literature indicated, older vehicles are often driven less miles,
resulting in less of an emissions impact (Moghadam & Livernois, 2010). A more significant
result arising from the regression models utilizing vehicle model-year as the as explanatory
variables is the fact that it appears vehicles undergoing inspection for change of ownership
purposes are substantially more likely to fail the smog check inspection.
Recommendations for Identifying Directed Vehicles
The process of directing vehicles for specialized testing based upon the likelihood of the
vehicle failing the inspection is a necessary and important component of the smog check
inspection program. Directing vehicles based upon broad categories such as age or manufacturer
while appropriate in some circumstances, compromises the legitimacy of the program by failing
to effectively scrutinize the potentially highest polluting vehicles. The binomial logistic
regression analysis employed in this thesis creates a methodology for continued identification of
the potentially highest polluting vehicles. This model, or a similar model, should be employed by
the Bureau to continue the process of selecting only the appropriate vehicles as directed vehicles.
Consideration of the following recommendations would support that endeavor.
1. Improve the identification of like vehicles by assigning VLT ID numbers through a
process of decoding the vehicle identification number (VIN) to ascertain the correct
make, model, model-year, engine size, transmission type, and other factors. It may be
possible to eliminate VLT ID numbers entirely by creating tables based upon the first 11
characters in the VIN, which are the characters used to specifically identify a vehicle.
2. On an annual basis, conduct binomial logistic regression analyses of the initial
inspections performed on all vehicles over the preceding two years. Analyzing the initial
59
inspections only will overcome any skewness caused by multiple tests on the same
vehicle. Additionally, the initial inspection evaluates the vehicle based upon its most
recent operating condition.
3. Failure of the OBD II functional test and failure of the visual inspection are the
appropriate dependent variables for the regression analyses described in item two above.
The VLT ID numbers or VIN assigned identification categories are the appropriate
explanatory variables.
4. Also run binomial logistic regressions with inspection reason (change of ownership or
biennial) as the explanatory variables and failure of the OBD II test and visual inspection
as the dependent variables. These regressions, also performed on an annual basis, should
only evaluate the preceding one year of data, than compared to the prior year’s results to
identify variances in likelihood of failure of the inspection based upon inspection reason.
5. At the same time of the regression models described in item four, conduct regression
models with model-year as the explanatory variables to evaluate the accuracy of the
technician entries of inspection reason. As described previously, it can be reasonably
assumed that odd model-year vehicles tested in even calendar years are undergoing
inspection for change of ownership purposes, and vice-versa.
6. If the determination is that vehicles undergoing inspection for change of ownership have
a significantly higher likelihood of failure of the inspection, those vehicles would be
appropriate as directed vehicles. Implementation of this policy may require amendments
to the California Health and Safety Code.
7. The data does not support the use of vehicle make (manufacturer) as a criterion for
identification of directed vehicles. Although some manufacturers stood out in the
regression models, only one, Daewoo, did so with any significance. That level of
60
significance is insufficient to support the use of manufacturer as a criterion. As the body
of literature indicates, all manufacturers experience design flaws or failures that lead to
failure of one of the tests or inspections.
Suggestions for Improved Statistical Modeling
In totality, the binomial logistic regression model was quite effective at ranking each set
of explanatory variables in order of likelihood of failure. Additionally, performing the regression
on the entire dataset rather than a sample of the data produced meaningful and valid results.
However, an opportunity for improvement of the model exists. I recommend future research
include measures to address the following item.
An area for improvement is the assignment of VLT ID numbers. According to the Smog
Check Inspection Procedures Manual (Bureau of Automotive Repair, 2009) technicians are
required to verify and correct data entries such as the vehicle make, model, engine size, etc.
Based upon these entries, the appropriate VLT ID number is assigned to the vehicle under
inspection (Bureau of Automotive Repair, 2009). Also during the inspection procedure, the
technicians are required to enter the vehicle identification number (VIN), a 17-digit number
equating to the serial number of the vehicle. Numerous VIN ‘decoders’ exist on the internet and
certainly the software is available for purchase. Programming the vehicle identification database
to utilize VIN decoding software and assigning the VLT ID number based upon the VIN would
remove the opportunity for human error in data entry. Such a process would ensure the validity of
the inspection record for future analysis.
Conclusion
In this thesis, I sought to develop a methodology for the continued identification of
directed vehicles based upon their likelihood of failure as compared to all other vehicles subject
to inspection. The six regression models conducted in support of this analysis clearly support the
61
use of binomial logistic regression in the evaluation of the fleet of vehicles. Using failure of the
OBD II functional test and visual inspection portion of the current smog check inspection
procedure as the dependent variables and VLT ID numbers as the explanatory variables enabled
me to identify a highly effective model for identifying directed vehicles without relying upon
emissions measurements. Implementation of this model would enable officials to continue
directing vehicles based upon the actual likelihood of failure, rather than based upon broad,
general categories such as age or manufacturer.
Calendar Year
Model Year
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Up to 1976
122,494
103,658
60,561
32,728
19,417
20,887
18,047
17,738
14,427
12,692
11,383
10,359
1977
74,131
62,615
51,622
42,934
35,167
29,512
25,410
21,721
19,862
17,723
15,647
13,908
1978
87,837
73,387
60,139
49,903
40,132
34,196
28,692
24,888
22,636
20,203
17,834
15,816
1979
99,475
83,117
67,845
55,867
45,174
37,637
32,655
27,219
25,408
22,017
19,718
17,416
1980
74,269
59,088
46,495
37,610
29,335
24,067
20,111
17,265
15,342
13,330
11,871
10,422
1981
89,647
71,615
56,079
45,150
35,242
28,677
23,747
20,285
18,008
15,533
13,584
12,120
1982
112,385
89,679
70,099
55,597
42,097
34,588
28,237
23,733
20,776
18,085
15,687
13,788
1983
155,947
124,505
97,110
77,455
58,484
47,579
38,373
32,575
28,076
24,187
20,906
18,138
1984
275,873
225,153
178,454
142,733 108,973
87,417
71,236
59,519
51,447
44,241
37,761
32,610
1985
356,604
296,758
238,247
192,885 148,126 119,140
96,967
80,991
69,618
59,702
50,836
44,045
1986
454,687
385,767
316,763
261,552 205,539 168,222 138,276 115,660
99,174
84,947
72,238
63,072
1987
516,031
445,684
372,070
311,037 244,867 201,643 165,370 137,478 118,046
100,004
84,267
72,691
1988
554,593
491,953
417,575
357,337 285,785 239,788 197,940 166,109 142,785
122,332
102,690
88,396
1989
643,548
585,923
510,698
447,567 367,390 313,440 263,551 223,210 194,515
167,146
141,579
122,694
1990
626,273
586,711
525,079
475,250 403,116 352,860 304,411 262,996 233,044
205,538
175,110
153,305
1991
651,496
618,129
562,694
518,859 449,202 398,576 349,721 308,498 275,209
245,795
211,920
186,416
1992
586,843
563,485
523,753
488,960 432,451 387,325 342,621 304,411 275,520
248,567
215,275
191,201
1993
676,034
639,814
612,988
569,064 515,600 457,472 413,751 364,315 336,360
302,914
267,107
236,487
1994
629,098
769,550
606,536
687,338 539,705 561,923 450,534 455,387 380,486
386,791
312,719
311,089
1995
1,079,486
572,717
956,236
589,536 814,703 524,224 661,575 448,549 551,981
401,845
453,245
333,467
1996
399,765
996,133
452,187
871,049 464,151 714,842 421,534 591,217 386,543
513,482
341,603
426,162
1997
1,191,184
547,867 1,057,711
588,941 932,452 567,689 791,433 512,028 689,331
485,418
594,431
432,426
1998
381,038 1,208,695
554,818 1,088,934 580,712 953,315 568,786 814,094 532,383
740,376
497,446
647,217
1999
278,018
417,603 1,298,216
601,860 1,166,874 615,687 1,030,559 592,924 914,148
591,268
819,753
559,600
Pre 2000 MY Tests 10,116,756 10,019,606 9,693,975 8,590,146 7,964,694 6,920,706 6,483,537 5,622,810 5,415,125 4,844,136 4,504,610 4,012,845
Total Annual Tests 10,441,518 10,597,594 10,737,282 11,050,541 9,018,521 9,360,649 9,319,693 9,554,028 9,825,967 10,571,730 10,776,597 11,290,816
Percentage of Fleet
96.89%
94.55%
90.28%
77.74%
88.31%
73.93%
69.57%
58.85%
55.11%
45.82%
41.80%
35.54%
62
Appendix A: Test Volume for Pre-Model-Year 2000 Vehicles
63
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
1838
1840
1841
1847
1849
1856
3,242
4,693
1,209
1,115
744
1,497
0.04
0.06
0.02
0.02
0.01
0.02
1955
1956
1957
1964
1966
1967
1,828
773
1,520
995
6,218
20,887
0.03
0.01
0.02
0.01
0.09
0.29
1858
1859
1860
1865
1867
1868
6,726
14,344
2,618
5,320
4,446
4,594
0.09
0.2
0.04
0.07
0.06
0.06
1968
1973
1974
1975
1976
1977
2,224
8,422
753
7,591
8,668
392
0.03
0.12
0.01
0.1
0.12
0.01
1869
1876
1885
1886
1887
1892
1,038
2,079
1,323
6,124
729
3,179
0.01
0.03
0.02
0.08
0.01
0.04
1984
1993
1994
1995
2000
2002
3,823
811
5,252
731
4,947
3,485
0.05
0.01
0.07
0.01
0.07
0.05
1894
1895
1901
1903
1910
1912
1913
2,838
704
634
512
774
2,881
9,338
0.04
0.01
0.01
0.01
0.01
0.04
0.13
2003
2009
2011
2018
2020
2021
2022
1,066
1,035
464
430
2,413
9,062
990
0.01
0.01
0.01
0.01
0.03
0.12
0.01
1914
1919
1920
1921
1922
1930
1,675
2,634
591
2,598
3,217
1,790
0.02
0.04
0.01
0.04
0.04
0.02
2027
2029
2030
2038
2047
2048
2,565
4,582
5,763
712
477
2,955
0.04
0.06
0.08
0.01
0.01
0.04
1939
1940
1941
1946
1947
1948
546
3,065
444
7,010
1,027
5,924
0.01
0.04
0.01
0.1
0.01
0.08
2049
2054
2055
2056
2057
2063
434
9,328
959
21,681
3,948
2,602
0.01
0.13
0.01
0.3
0.05
0.04
1949
2,622
0.04
2064
1,030
0.01
64
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
2065
2072
2073
2074
2075
2076
1,160
1,359
406
4,178
27,056
1,737
0.02
0.02
0.01
0.06
0.37
0.02
2192
2198
2200
2209
2210
2211
8,405
581
3,008
397
5,573
628
0.12
0.01
0.04
0.01
0.08
0.01
2081
2083
2084
2092
2101
2102
10,180
14,302
13,783
3,396
759
6,047
0.14
0.2
0.19
0.05
0.01
0.08
2216
2218
2219
2225
2236
2237
2,409
2,990
1,270
1,261
527
3,594
0.03
0.04
0.02
0.02
0.01
0.05
2103
2108
2110
2111
2117
2119
696
4,960
4,955
1,943
763
1,271
0.01
0.07
0.07
0.03
0.01
0.02
2243
2245
2246
2254
2264
2269
829
1,851
1,689
459
786
537
0.01
0.03
0.02
0.01
0.01
0.01
2128
2129
2130
2135
2137
2138
2146
1,108
6,110
384
893
4,448
3,255
1,599
0.02
0.08
0.01
0.01
0.06
0.04
0.02
2270
2271
2272
2273
2274
2279
2281
51,464
2,932
38,394
9,503
397
4,340
1,070
0.71
0.04
0.53
0.13
0.01
0.06
0.01
2156
2162
2163
2164
2165
2171
2,207
10,361
437
16,712
4,551
4,734
0.03
0.14
0.01
0.23
0.06
0.06
2288
2289
2290
2291
2297
2299
1,878
517
5,038
12,154
8,096
21,845
0.03
0.01
0.07
0.17
0.11
0.3
2173
2180
2182
2183
2184
2189
1,230
881
2,885
18,027
683
3,398
0.02
0.01
0.04
0.25
0.01
0.05
2300
2308
2318
2324
2325
2326
9,732
5,333
2,042
23,586
1,408
12,106
0.13
0.07
0.03
0.32
0.02
0.17
2191
7,667
0.11
2327
2,396
0.03
65
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
2333
2335
2342
2344
2345
2351
1,878
389
587
996
2,303
3,164
0.03
0.01
0.01
0.01
0.03
0.04
2488
2489
2495
2506
2507
2513
4,331
969
820
708
1,600
2,152
0.06
0.01
0.01
0.01
0.02
0.03
2353
2354
2362
2372
2378
2379
6,203
1,696
1,582
504
27,491
925
0.09
0.02
0.02
0.01
0.38
0.01
2515
2516
2524
2534
29100
29101
4,977
1,505
2,165
1,333
7,962
1,739
0.07
0.02
0.03
0.02
0.11
0.02
2380
2381
2387
2396
2398
2399
9,674
1,891
1,779
507
907
2,716
0.13
0.03
0.02
0.01
0.01
0.04
29102
29117
29118
29123
29124
29137
4,504
409
405
570
393
2,484
0.06
0.01
0.01
0.01
0.01
0.03
2405
2407
2408
2416
2426
2432
2433
4,415
6,318
1,411
2,578
577
27,244
699
0.06
0.09
0.02
0.04
0.01
0.37
0.01
29138
29141
29142
29143
29144
29145
29146
6,497
1,204
2,847
4,888
415
466
1,402
0.09
0.02
0.04
0.07
0.01
0.01
0.02
2434
2435
2441
2450
2452
2453
9,235
2,061
2,017
364
1,190
2,714
0.13
0.03
0.03
0
0.02
0.04
29148
29149
29152
29155
29158
29159
869
1,505
471
574
642
752
0.01
0.02
0.01
0.01
0.01
0.01
2459
2461
2462
2470
2480
2486
5,756
8,170
2,000
3,137
1,076
13,784
0.08
0.11
0.03
0.04
0.01
0.19
29161
29162
29163
29164
29165
29166
4,070
4,807
1,251
2,340
699
3,897
0.06
0.07
0.02
0.03
0.01
0.05
2487
741
0.01
29167
1,023
0.01
66
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
29168
29169
29170
29171
29173
29174
2,886
1,633
2,474
8,535
483
4,670
0.04
0.02
0.03
0.12
0.01
0.06
29236
29238
29239
29240
29241
29243
1,989
735
674
671
634
678
0.03
0.01
0.01
0.01
0.01
0.01
29175
29179
29180
29181
29182
29184
2,666
3,752
7,197
13,520
527
1,815
0.04
0.05
0.1
0.19
0.01
0.02
29244
29245
29246
29249
29250
29252
2,191
1,935
1,766
1,029
987
756
0.03
0.03
0.02
0.01
0.01
0.01
29188
29189
29190
29191
29192
29193
1,652
1,590
1,507
4,650
499
2,137
0.02
0.02
0.02
0.06
0.01
0.03
29256
29267
29268
29269
29270
29273
509
2,735
487
10,906
806
3,170
0.01
0.04
0.01
0.15
0.01
0.04
29194
29196
29200
29201
29206
29207
29209
607
700
2,356
3,451
709
5,964
2,453
0.01
0.01
0.03
0.05
0.01
0.08
0.03
29274
29277
29281
29282
29283
29285
29286
1,728
831
3,874
2,029
3,394
2,301
2,852
0.02
0.01
0.05
0.03
0.05
0.03
0.04
29211
29212
29213
29214
29215
29217
2,264
1,167
575
814
701
6,194
0.03
0.02
0.01
0.01
0.01
0.09
29287
29289
29290
29291
29292
29294
897
5,552
664
1,608
1,495
611
0.01
0.08
0.01
0.02
0.02
0.01
29219
29220
29222
29223
29224
29229
1,258
1,408
4,124
1,938
461
520
0.02
0.02
0.06
0.03
0.01
0.01
29297
29298
29299
29305
29306
29307
570
2,117
1,003
624
917
1,089
0.01
0.03
0.01
0.01
0.01
0.01
29235
2,425
0.03
29309
1,532
0.02
67
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
29313
29314
29315
29317
29318
29321
1,082
630
524
716
596
750
0.01
0.01
0.01
0.01
0.01
0.01
29366
29367
29370
29372
29373
29375
927
2,198
443
1,097
406
1,171
0.01
0.03
0.01
0.02
0.01
0.02
29322
29323
29324
29325
29326
29327
2,675
822
9,581
14,781
2,776
12,681
0.04
0.01
0.13
0.2
0.04
0.17
29377
29378
29379
29381
29383
29390
509
10,346
820
958
7,013
882
0.01
0.14
0.01
0.01
0.1
0.01
29328
29330
29331
29332
29333
29334
2,588
507
1,702
1,272
902
1,718
0.04
0.01
0.02
0.02
0.01
0.02
29391
29392
29398
29407
29408
29411
2,490
3,998
657
1,429
1,024
1,971
0.03
0.05
0.01
0.02
0.01
0.03
29335
29336
29337
29338
29340
29341
29342
739
1,148
860
553
4,184
3,902
6,991
0.01
0.02
0.01
0.01
0.06
0.05
0.1
29417
29418
29419
29420
29421
29424
29425
1,161
882
1,204
715
2,646
1,216
396
0.02
0.01
0.02
0.01
0.04
0.02
0.01
29345
29346
29347
29349
29350
29352
971
8,659
1,724
2,917
1,333
3,515
0.01
0.12
0.02
0.04
0.02
0.05
29430
29431
29433
29434
29437
29438
37,528
18,244
843
48,573
2,841
4,261
0.52
0.25
0.01
0.67
0.04
0.06
29353
29354
29355
29356
29357
29358
826
1,863
5,470
2,380
1,113
1,184
0.01
0.03
0.08
0.03
0.02
0.02
29439
29440
29441
29442
29443
29444
15,515
1,287
566
2,035
1,165
1,500
0.21
0.02
0.01
0.03
0.02
0.02
29363
884
0.01
29445
666
0.01
68
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
29446
29447
29449
29450
29453
29454
3,292
537
422
1,438
1,567
4,755
0.05
0.01
0.01
0.02
0.02
0.07
29524
29526
29533
29543
29544
29545
2,510
1,093
423
4,945
1,644
3,588
0.03
0.02
0.01
0.07
0.02
0.05
29455
29456
29465
29466
29467
29470
476
779
1,956
679
643
1,664
0.01
0.01
0.03
0.01
0.01
0.02
29546
29547
29548
29551
29553
29555
1,261
3,442
1,182
1,232
1,011
5,550
0.02
0.05
0.02
0.02
0.01
0.08
29471
29477
29485
29487
29488
29489
1,277
730
1,750
2,341
7,033
1,605
0.02
0.01
0.02
0.03
0.1
0.02
29558
29559
29562
29563
29566
29567
598
1,410
5,276
1,661
2,662
3,386
0.01
0.02
0.07
0.02
0.04
0.05
29490
29491
29493
29494
29495
29497
29499
2,375
3,994
707
3,291
3,457
2,302
1,602
0.03
0.05
0.01
0.05
0.05
0.03
0.02
29568
29570
29573
29574
29575
29576
29580
761
753
1,545
3,340
504
1,250
2,167
0.01
0.01
0.02
0.05
0.01
0.02
0.03
29503
29504
29505
29508
29509
29510
459
6,607
3,587
696
1,458
1,616
0.01
0.09
0.05
0.01
0.02
0.02
29582
29583
29584
29585
29586
29587
637
2,267
2,795
4,704
1,002
845
0.01
0.03
0.04
0.06
0.01
0.01
29511
29512
29517
29518
29519
29520
4,951
3,804
815
1,139
2,407
3,897
0.07
0.05
0.01
0.02
0.03
0.05
29588
29591
29592
29593
29598
29599
3,268
4,246
739
1,305
10,944
7,295
0.04
0.06
0.01
0.02
0.15
0.1
29521
2,808
0.04
29600
6,120
0.08
69
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
29602
29603
29604
29605
29606
29607
1,320
13,118
1,904
1,006
6,095
5,442
0.02
0.18
0.03
0.01
0.08
0.07
29661
29662
29663
29664
29670
29671
413
480
538
3,567
452
737
0.01
0.01
0.01
0.05
0.01
0.01
29608
29610
29611
29612
29613
29614
743
5,549
3,097
1,154
1,398
478
0.01
0.08
0.04
0.02
0.02
0.01
29683
29684
29686
29687
29688
29689
4,185
8,602
5,424
12,119
35,579
12,158
0.06
0.12
0.07
0.17
0.49
0.17
29615
29616
29620
29622
29624
29627
1,717
825
2,507
878
1,253
1,146
0.02
0.01
0.03
0.01
0.02
0.02
29691
29692
29695
29696
29697
29698
2,635
3,731
6,190
25,961
6,017
2,081
0.04
0.05
0.08
0.36
0.08
0.03
29628
29629
29630
29631
29632
29633
29634
564
677
1,840
2,757
809
2,291
1,229
0.01
0.01
0.03
0.04
0.01
0.03
0.02
29699
29700
29701
29704
29705
29706
29707
720
4,211
876
21,166
9,365
2,533
6,361
0.01
0.06
0.01
0.29
0.13
0.03
0.09
29635
29637
29639
29641
29645
29646
1,189
1,251
810
580
554
449
0.02
0.02
0.01
0.01
0.01
0.01
29708
29709
29710
29711
29713
29714
997
2,678
1,557
11,120
3,862
512
0.01
0.04
0.02
0.15
0.05
0.01
29653
29654
29656
29657
29658
29659
2,958
1,913
617
1,227
6,958
733
0.04
0.03
0.01
0.02
0.1
0.01
29719
29723
29724
29725
29726
29727
1,825
409
831
8,765
3,101
1,209
0.03
0.01
0.01
0.12
0.04
0.02
29660
1,688
0.02
29728
1,178
0.02
70
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
29729
29730
29731
29732
29733
29734
3,614
1,770
2,094
735
1,449
573
0.05
0.02
0.03
0.01
0.02
0.01
29835
29836
29837
29839
29842
29843
1,719
414
1,278
1,990
650
820
0.02
0.01
0.02
0.03
0.01
0.01
29735
29736
29737
29746
29747
29749
378
432
442
2,016
621
2,702
0.01
0.01
0.01
0.03
0.01
0.04
29844
29847
29848
29852
29853
29854
1,712
1,766
969
2,389
5,358
5,178
0.02
0.02
0.01
0.03
0.07
0.07
29752
29753
29754
29758
29759
29760
1,206
1,658
806
1,389
1,054
2,617
0.02
0.02
0.01
0.02
0.01
0.04
29855
29857
29858
29859
29860
29863
3,765
5,917
876
582
2,683
1,249
0.05
0.08
0.01
0.01
0.04
0.02
29762
29764
29765
29774
29776
29780
29794
2,137
1,297
2,204
597
375
554
600
0.03
0.02
0.03
0.01
0.01
0.01
0.01
29870
29871
29878
29880
29882
29883
29884
1,711
1,588
2,277
1,961
3,612
513
3,829
0.02
0.02
0.03
0.03
0.05
0.01
0.05
29802
29803
29804
29810
29812
29814
1,054
615
4,305
1,532
2,057
1,332
0.01
0.01
0.06
0.02
0.03
0.02
29885
29886
29887
29888
29890
29891
800
663
841
1,049
2,365
644
0.01
0.01
0.01
0.01
0.03
0.01
29816
29817
29819
29820
29827
29828
1,484
603
883
1,313
1,754
1,219
0.02
0.01
0.01
0.02
0.02
0.02
29903
29904
29909
29910
29912
29914
1,301
1,012
4,027
611
394
498
0.02
0.01
0.06
0.01
0.01
0.01
29834
1,854
0.03
29915
1,237
0.02
71
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
29918
29921
29922
29926
29927
29928
457
598
429
586
3,710
690
0.01
0.01
0.01
0.01
0.05
0.01
30010
30011
30012
30013
30014
30015
475
412
1,730
536
2,018
2,948
0.01
0.01
0.02
0.01
0.03
0.04
29931
29932
29935
29937
29939
29940
1,232
802
387
2,146
1,361
675
0.02
0.01
0.01
0.03
0.02
0.01
30016
30017
30018
30019
30021
30022
1,190
3,030
5,324
915
2,017
814
0.02
0.04
0.07
0.01
0.03
0.01
29941
29946
29947
29948
29949
29951
1,890
2,741
541
2,371
1,458
902
0.03
0.04
0.01
0.03
0.02
0.01
30024
30026
30027
30028
30029
30030
1,476
458
430
2,595
1,059
1,296
0.02
0.01
0.01
0.04
0.01
0.02
29952
29965
29966
29968
29974
29979
29985
749
909
367
477
1,563
614
755
0.01
0.01
0.01
0.01
0.02
0.01
0.01
30038
30039
30045
30047
30049
30051
30052
468
854
1,041
507
6,133
449
1,735
0.01
0.01
0.01
0.01
0.08
0.01
0.02
29990
29991
29992
29993
29994
29995
3,799
1,533
411
4,239
5,399
3,726
0.05
0.02
0.01
0.06
0.07
0.05
30053
30059
30060
30063
30064
30072
1,722
1,747
1,487
1,657
952
572
0.02
0.02
0.02
0.02
0.01
0.01
29996
29998
30002
30003
30004
30005
2,209
2,224
953
558
504
667
0.03
0.03
0.01
0.01
0.01
0.01
30080
30084
30085
30086
30087
30089
670
1,172
716
819
701
760
0.01
0.02
0.01
0.01
0.01
0.01
30008
477
0.01
30092
656
0.01
72
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
30096
30097
30099
30101
30102
30103
16,771
7,187
24,924
1,323
1,624
6,203
0.23
0.1
0.34
0.02
0.02
0.09
30168
30169
30172
30173
30175
30176
454
1,806
3,165
1,538
2,127
1,259
0.01
0.02
0.04
0.02
0.03
0.02
30104
30106
30107
30108
30109
30110
662
1,327
615
534
1,447
2,676
0.01
0.02
0.01
0.01
0.02
0.04
30178
30180
30181
30182
30184
30211
513
1,008
1,816
1,204
1,222
1,532
0.01
0.01
0.02
0.02
0.02
0.02
30111
30112
30113
30114
30117
30118
467
873
446
586
704
495
0.01
0.01
0.01
0.01
0.01
0.01
30212
30213
30215
30217
30218
30220
926
741
797
1,363
1,128
1,517
0.01
0.01
0.01
0.02
0.02
0.02
30119
30121
30122
30127
30128
30136
30137
2,031
737
665
927
396
552
502
0.03
0.01
0.01
0.01
0.01
0.01
0.01
30224
30226
30229
30233
30236
30237
30239
587
487
2,690
806
2,065
581
1,200
0.01
0.01
0.04
0.01
0.03
0.01
0.02
30148
30149
30150
30151
30152
30155
1,124
2,283
607
1,253
1,770
1,425
0.02
0.03
0.01
0.02
0.02
0.02
30240
30245
30247
30250
30251
30254
1,319
443
647
552
1,129
1,179
0.02
0.01
0.01
0.01
0.02
0.02
30156
30158
30159
30160
30161
30163
526
1,534
1,583
885
419
886
0.01
0.02
0.02
0.01
0.01
0.01
30256
30258
30259
30260
30262
30263
405
422
1,448
492
794
2,879
0.01
0.01
0.02
0.01
0.01
0.04
30166
1,851
0.03
30264
876
0.01
73
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
30265
30266
30268
30271
30272
30274
449
1,425
776
1,319
415
506
0.01
0.02
0.01
0.02
0.01
0.01
30348
30355
30376
30377
30378
30379
1,475
1,396
2,888
1,576
3,013
11,511
0.02
0.02
0.04
0.02
0.04
0.16
30278
30280
30281
30282
30283
30284
5,874
1,304
1,384
1,996
432
3,568
0.08
0.02
0.02
0.03
0.01
0.05
30380
30382
30383
30386
30388
30389
2,909
1,211
994
2,449
14,314
2,538
0.04
0.02
0.01
0.03
0.2
0.03
30287
30288
30290
30291
30292
30294
1,803
1,288
2,224
4,382
1,053
2,317
0.02
0.02
0.03
0.06
0.01
0.03
30390
30391
30393
30396
30397
30398
514
1,238
711
588
2,316
977
0.01
0.02
0.01
0.01
0.03
0.01
30295
30296
30298
30299
30302
30304
30311
1,067
674
674
450
465
948
539
0.01
0.01
0.01
0.01
0.01
0.01
0.01
30399
30400
30401
30402
30403
30404
30407
1,577
1,129
5,913
3,524
1,701
2,962
1,417
0.02
0.02
0.08
0.05
0.02
0.04
0.02
30312
30315
30317
30318
30319
30320
410
1,091
1,446
717
1,037
607
0.01
0.01
0.02
0.01
0.01
0.01
30408
30409
30411
30412
30417
30418
1,519
2,938
958
603
615
498
0.02
0.04
0.01
0.01
0.01
0.01
30321
30323
30325
30341
30342
30345
775
728
369
1,193
902
1,472
0.01
0.01
0.01
0.02
0.01
0.02
30419
30421
30422
30423
30424
30427
427
1,805
5,161
1,579
615
1,376
0.01
0.02
0.07
0.02
0.01
0.02
30346
3,105
0.04
30428
1,985
0.03
74
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
30429
30430
30431
30440
30442
30445
871
1,559
1,863
1,851
1,576
529
0.01
0.02
0.03
0.03
0.02
0.01
30531
30532
30533
30534
30535
30536
3,420
880
2,103
898
1,100
4,040
0.05
0.01
0.03
0.01
0.02
0.06
30446
30447
30448
30452
30454
30455
384
451
1,056
666
7,936
1,495
0.01
0.01
0.01
0.01
0.11
0.02
30537
30538
30539
30540
30542
30543
591
1,976
5,235
1,131
1,359
1,565
0.01
0.03
0.07
0.02
0.02
0.02
30456
30460
30463
30469
30470
30471
6,679
3,916
698
947
626
1,010
0.09
0.05
0.01
0.01
0.01
0.01
30544
30547
30548
30549
30550
30554
3,534
2,021
1,636
2,286
1,049
603
0.05
0.03
0.02
0.03
0.01
0.01
30472
30475
30482
30490
30494
30495
30496
498
418
452
693
2,049
1,229
1,213
0.01
0.01
0.01
0.01
0.03
0.02
0.02
30555
30556
30557
30558
30559
30560
30561
3,571
13,595
9,196
8,950
3,884
11,174
1,278
0.05
0.19
0.13
0.12
0.05
0.15
0.02
30500
30501
30503
30507
30509
30511
1,698
1,010
2,362
3,596
931
1,885
0.02
0.01
0.03
0.05
0.01
0.03
30562
30563
30565
30566
30569
30573
2,046
6,328
1,646
1,265
567
3,279
0.03
0.09
0.02
0.02
0.01
0.05
30512
30516
30517
30518
30520
30522
1,154
945
432
461
904
1,065
0.02
0.01
0.01
0.01
0.01
0.01
30574
30578
30580
30581
30583
30584
2,596
387
612
4,300
4,368
425
0.04
0.01
0.01
0.06
0.06
0.01
30530
4,244
0.06
30585
6,873
0.09
75
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
30586
30587
30588
30589
30590
30592
5,018
807
1,513
964
993
3,318
0.07
0.01
0.02
0.01
0.01
0.05
30661
30668
30669
30671
30677
30684
415
1,331
471
781
1,422
1,100
0.01
0.02
0.01
0.01
0.02
0.02
30593
30594
30605
30606
30610
30611
693
528
6,563
3,097
2,223
1,543
0.01
0.01
0.09
0.04
0.03
0.02
30686
30689
30693
30694
30695
30696
409
1,392
389
8,351
2,756
781
0.01
0.02
0.01
0.11
0.04
0.01
30614
30617
30618
30620
30622
30624
685
12,066
2,174
377
817
2,145
0.01
0.17
0.03
0.01
0.01
0.03
30697
30698
30699
30700
30701
30703
4,710
5,714
4,320
19,103
2,784
5,310
0.06
0.08
0.06
0.26
0.04
0.07
30626
30627
30628
30635
30636
30637
30642
469
850
395
1,388
7,675
1,777
2,416
0.01
0.01
0.01
0.02
0.11
0.02
0.03
30704
30706
30707
30708
30709
30710
30711
2,211
3,014
618
450
1,538
972
685
0.03
0.04
0.01
0.01
0.02
0.01
0.01
30644
30647
30649
30650
30651
30652
1,596
656
3,242
418
1,583
631
0.02
0.01
0.04
0.01
0.02
0.01
30712
30713
30714
30715
30719
30720
615
743
886
815
1,138
934
0.01
0.01
0.01
0.01
0.02
0.01
30653
30654
30655
30657
30658
30659
2,832
722
4,916
6,164
2,317
680
0.04
0.01
0.07
0.08
0.03
0.01
30721
30722
30723
30724
30725
30726
3,756
1,141
4,584
5,317
3,088
5,201
0.05
0.02
0.06
0.07
0.04
0.07
30660
930
0.01
30727
10,074
0.14
76
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
30728
30729
30730
30731
30734
30735
2,159
468
732
643
2,701
1,018
0.03
0.01
0.01
0.01
0.04
0.01
30803
30807
30810
30811
30813
30814
933
1,157
39,144
16,337
44,043
1,632
0.01
0.02
0.54
0.22
0.6
0.02
30736
30737
30738
30739
30742
30743
718
4,789
1,468
2,642
417
1,467
0.01
0.07
0.02
0.04
0.01
0.02
30817
30818
30819
30820
30823
30825
3,263
5,017
18,001
454
1,348
998
0.04
0.07
0.25
0.01
0.02
0.01
30749
30751
30753
30754
30755
30757
2,769
566
850
612
1,231
568
0.04
0.01
0.01
0.01
0.02
0.01
30826
30827
30828
30829
30830
30831
3,045
6,263
1,102
3,091
1,081
1,408
0.04
0.09
0.02
0.04
0.01
0.02
30759
30760
30761
30762
30763
30764
30768
9,748
572
2,395
2,003
548
3,155
788
0.13
0.01
0.03
0.03
0.01
0.04
0.01
30832
30835
30836
30837
30838
30840
30841
3,209
1,166
1,985
3,738
1,020
802
721
0.04
0.02
0.03
0.05
0.01
0.01
0.01
30770
30771
30773
30774
30775
30776
3,818
3,070
1,325
3,502
2,268
3,174
0.05
0.04
0.02
0.05
0.03
0.04
30842
30846
30854
30855
30858
30860
693
1,113
691
550
1,543
1,360
0.01
0.02
0.01
0.01
0.02
0.02
30777
30793
30794
30797
30798
30799
1,899
1,406
489
2,061
3,399
917
0.03
0.02
0.01
0.03
0.05
0.01
30869
30870
30871
30872
30874
30876
3,842
1,342
1,630
2,985
3,656
3,751
0.05
0.02
0.02
0.04
0.05
0.05
30800
1,137
0.02
30880
555
0.01
77
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
30882
30883
30884
30885
30886
30887
2,639
650
872
2,823
3,092
3,551
0.04
0.01
0.01
0.04
0.04
0.05
30955
30959
30963
30967
30968
30970
1,186
3,355
964
3,894
1,389
2,694
0.02
0.05
0.01
0.05
0.02
0.04
30888
30889
30892
30893
30894
30895
1,247
472
1,228
854
476
9,743
0.02
0.01
0.02
0.01
0.01
0.13
30971
30974
30976
30979
30982
30983
2,132
435
413
549
868
1,879
0.03
0.01
0.01
0.01
0.01
0.03
30897
30898
30901
30902
30903
30904
553
2,677
5,352
3,750
770
5,260
0.01
0.04
0.07
0.05
0.01
0.07
30984
30985
30986
30987
30989
30991
1,382
662
503
767
2,712
656
0.02
0.01
0.01
0.01
0.04
0.01
30905
30908
30909
30910
30911
30912
30918
2,366
522
1,299
2,178
2,445
1,415
1,855
0.03
0.01
0.02
0.03
0.03
0.02
0.03
30992
30993
30994
30996
30997
30998
31000
1,229
1,427
377
455
5,079
949
7,070
0.02
0.02
0.01
0.01
0.07
0.01
0.1
30920
30926
30938
30939
30940
30941
559
560
404
2,565
1,210
6,658
0.01
0.01
0.01
0.04
0.02
0.09
31001
31004
31005
31006
31008
31013
496
1,173
2,154
1,107
1,368
14,679
0.01
0.02
0.03
0.02
0.02
0.2
30943
30944
30945
30948
30949
30950
2,797
3,029
5,324
2,627
676
800
0.04
0.04
0.07
0.04
0.01
0.01
31014
31015
31016
31017
31018
31019
3,707
2,659
2,397
5,266
1,143
6,086
0.05
0.04
0.03
0.07
0.02
0.08
30953
1,562
0.02
31021
2,091
0.03
78
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
31022
31024
31025
31026
31027
31028
1,553
2,979
9,104
702
717
627
0.02
0.04
0.12
0.01
0.01
0.01
31103
31104
31105
31106
31115
31116
1,729
564
794
445
1,192
774
0.02
0.01
0.01
0.01
0.02
0.01
31029
31030
31031
31032
31036
31048
5,098
1,974
678
705
394
655
0.07
0.03
0.01
0.01
0.01
0.01
31123
31124
31125
31126
31127
31128
8,003
3,522
7,149
36,482
14,215
1,973
0.11
0.05
0.1
0.5
0.2
0.03
31049
31050
31051
31052
31053
31054
435
715
1,600
2,087
1,610
1,625
0.01
0.01
0.02
0.03
0.02
0.02
31129
31131
31132
31134
31135
31136
1,874
812
3,374
19,336
3,511
3,261
0.03
0.01
0.05
0.27
0.05
0.04
31056
31057
31060
31067
31068
31073
31074
797
1,098
601
514
464
718
427
0.01
0.02
0.01
0.01
0.01
0.01
0.01
31137
31138
31139
31141
31142
31143
31144
7,882
718
1,565
665
722
8,545
1,439
0.11
0.01
0.02
0.01
0.01
0.12
0.02
31077
31082
31083
31084
31085
31087
452
3,351
1,707
438
531
2,997
0.01
0.05
0.02
0.01
0.01
0.04
31145
31146
31147
31148
31149
31150
8,620
3,914
13,966
8,938
5,887
10,219
0.12
0.05
0.19
0.12
0.08
0.14
31088
31090
31092
31093
31095
31098
8,600
1,356
638
3,081
2,096
1,213
0.12
0.02
0.01
0.04
0.03
0.02
31152
31153
31154
31155
31157
31158
822
4,044
5,221
10,701
3,093
1,042
0.01
0.06
0.07
0.15
0.04
0.01
31099
1,001
0.01
31160
798
0.01
79
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
31162
31163
31165
31166
31167
31169
1,625
1,020
674
7,661
1,041
720
0.02
0.01
0.01
0.11
0.01
0.01
31291
31293
31294
31295
31299
31300
1,411
722
1,762
536
622
1,050
0.02
0.01
0.02
0.01
0.01
0.01
31171
31172
31173
31174
31177
31185
681
3,451
689
4,050
398
1,685
0.01
0.05
0.01
0.06
0.01
0.02
31303
31304
31306
31309
31310
31311
673
388
505
1,820
5,515
3,174
0.01
0.01
0.01
0.02
0.08
0.04
31189
31190
31192
31199
31201
31205
612
1,383
499
3,917
2,382
1,076
0.01
0.02
0.01
0.05
0.03
0.01
31312
31313
31314
31317
31319
31320
2,920
2,476
3,960
2,886
609
549
0.04
0.03
0.05
0.04
0.01
0.01
31211
31213
31217
31219
31246
31248
31252
997
416
709
474
664
499
502
0.01
0.01
0.01
0.01
0.01
0.01
0.01
31336
31337
31345
31347
31349
31350
31351
2,788
1,341
1,673
1,387
2,532
2,927
519
0.04
0.02
0.02
0.02
0.03
0.04
0.01
31255
31259
31264
31265
31268
31272
1,071
1,483
803
1,276
434
539
0.01
0.02
0.01
0.02
0.01
0.01
31352
31356
31370
31372
31377
31386
726
2,190
2,838
1,068
1,280
3,578
0.01
0.03
0.04
0.01
0.02
0.05
31275
31282
31284
31285
31287
31289
717
566
2,018
1,306
694
592
0.01
0.01
0.03
0.02
0.01
0.01
31387
31394
31397
31401
31402
31411
699
408
469
595
525
3,323
0.01
0.01
0.01
0.01
0.01
0.05
31290
3,046
0.04
31412
616
0.01
80
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
31416
31418
31423
31425
31428
31430
733
471
1,585
729
796
3,190
0.01
0.01
0.02
0.01
0.01
0.04
31526
31527
31528
31529
31539
31548
1,962
606
1,540
454
763
5,253
0.03
0.01
0.02
0.01
0.01
0.07
31432
31434
31435
31450
31452
31461
1,349
444
2,459
718
394
2,451
0.02
0.01
0.03
0.01
0.01
0.03
31550
31551
31552
31553
31555
31556
400
525
2,695
455
1,349
1,078
0.01
0.01
0.04
0.01
0.02
0.01
31468
31476
31480
31481
31483
31484
579
3,087
4,738
5,518
798
2,586
0.01
0.04
0.07
0.08
0.01
0.04
31558
31559
31560
31561
31563
31565
736
1,618
979
1,493
540
440
0.01
0.02
0.01
0.02
0.01
0.01
31485
31486
31488
31489
31491
31493
31494
421
1,478
1,328
660
885
805
692
0.01
0.02
0.02
0.01
0.01
0.01
0.01
31586
31591
31592
31593
31594
31596
31600
524
883
1,634
454
478
417
624
0.01
0.01
0.02
0.01
0.01
0.01
0.01
31495
31501
31511
31513
31514
31515
631
526
1,004
1,530
1,455
1,147
0.01
0.01
0.01
0.02
0.02
0.02
31605
31606
31607
31609
31610
31613
2,144
6,428
866
14,328
558
1,033
0.03
0.09
0.01
0.2
0.01
0.01
31516
31517
31519
31520
31522
31523
1,587
2,250
4,161
763
462
1,070
0.02
0.03
0.06
0.01
0.01
0.01
31614
31615
31616
31620
31621
31624
1,973
2,659
6,238
3,703
550
1,442
0.03
0.04
0.09
0.05
0.01
0.02
31525
421
0.01
31625
3,658
0.05
81
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
31626
31627
31630
31631
31633
31639
516
1,010
728
1,113
1,222
6,350
0.01
0.01
0.01
0.02
0.02
0.09
31766
31767
31769
31773
31781
31788
826
1,146
1,779
778
524
2,654
0.01
0.02
0.02
0.01
0.01
0.04
31640
31641
31669
31688
31690
31691
917
415
751
1,370
612
658
0.01
0.01
0.01
0.02
0.01
0.01
31794
31798
31799
31800
31802
31803
1,702
470
998
445
864
677
0.02
0.01
0.01
0.01
0.01
0.01
31693
31695
31698
31702
31703
31704
1,322
1,308
1,520
1,233
1,376
695
0.02
0.02
0.02
0.02
0.02
0.01
31808
31818
31825
31830
31831
31833
1,480
835
814
1,655
429
592
0.02
0.01
0.01
0.02
0.01
0.01
31706
31711
31713
31714
31716
31719
31722
1,904
844
437
1,003
2,634
867
684
0.03
0.01
0.01
0.01
0.04
0.01
0.01
31834
31835
31837
31844
31851
31853
31854
3,306
489
3,329
667
1,181
2,658
9,319
0.05
0.01
0.05
0.01
0.02
0.04
0.13
31723
31724
31727
31728
31731
31736
1,629
968
1,058
463
563
540
0.02
0.01
0.01
0.01
0.01
0.01
31855
31856
31857
31858
31860
31861
2,153
725
714
1,675
1,704
1,018
0.03
0.01
0.01
0.02
0.02
0.01
31737
31738
31743
31744
31762
31763
2,553
1,495
922
696
1,237
491
0.04
0.02
0.01
0.01
0.02
0.01
31862
31864
31865
31868
31872
31873
712
667
406
3,321
1,665
555
0.01
0.01
0.01
0.05
0.02
0.01
31764
2,454
0.03
31874
723
0.01
82
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
31875
31888
31889
31890
31894
31896
390
521
1,375
665
404
965
0.01
0.01
0.02
0.01
0.01
0.01
32004
32005
32006
32009
32010
32011
3,175
1,353
3,339
1,225
2,309
2,978
0.04
0.02
0.05
0.02
0.03
0.04
31917
31924
31925
31926
31927
31934
444
2,298
539
1,233
679
692
0.01
0.03
0.01
0.02
0.01
0.01
32013
32018
32019
32021
32022
32023
847
860
795
754
4,557
641
0.01
0.01
0.01
0.01
0.06
0.01
31935
31936
31937
31939
31943
31956
885
651
628
1,064
686
628
0.01
0.01
0.01
0.01
0.01
0.01
32028
32029
32030
32041
32057
32058
1,716
484
2,544
643
818
8,617
0.02
0.01
0.03
0.01
0.01
0.12
31958
31976
31977
31979
31980
31981
31982
703
1,345
864
573
752
1,860
12,110
0.01
0.02
0.01
0.01
0.01
0.03
0.17
32061
32062
32063
32065
32066
32069
32070
3,615
9,544
7,234
2,300
510
1,337
694
0.05
0.13
0.1
0.03
0.01
0.02
0.01
31983
31988
31989
31990
31991
31992
2,324
1,144
15,001
1,382
1,191
2,293
0.03
0.02
0.21
0.02
0.02
0.03
32073
32079
32090
32091
32094
32099
427
524
1,563
749
587
1,682
0.01
0.01
0.02
0.01
0.01
0.02
31994
31997
31999
32000
32001
32002
880
4,177
2,268
455
2,798
982
0.01
0.06
0.03
0.01
0.04
0.01
32100
32101
32102
32103
32105
32106
2,238
2,132
2,887
1,144
734
669
0.03
0.03
0.04
0.02
0.01
0.01
32003
2,623
0.04
32107
1,767
0.02
83
Appendix B: VLT ID Number Distribution
VLT ID
32108
Frequency
3,090
% of Observations
0.04
VLT ID
32174
Frequency
654
% of Observations
0.01
32109
32110
32114
32116
32119
32121
429
388
915
1,208
4,781
2,985
0.01
0.01
0.01
0.02
0.07
0.04
32175
32176
32177
32179
32181
32183
7,726
12,924
1,844
6,715
2,349
1,870
0.11
0.18
0.03
0.09
0.03
0.03
32122
32123
32127
32128
32130
32132
1,467
4,762
1,106
875
2,368
3,709
0.02
0.07
0.02
0.01
0.03
0.05
32185
32187
32188
32191
32192
32193
817
421
1,552
2,300
1,822
11,190
0.01
0.01
0.02
0.03
0.03
0.15
32133
32135
32136
32137
32138
32139
1,090
2,825
1,021
710
3,210
10,468
0.01
0.04
0.01
0.01
0.04
0.14
32194
32195
32196
32198
32199
32208
2,639
4,087
436
559
775
1,651
0.04
0.06
0.01
0.01
0.01
0.02
32140
32141
32142
32143
32144
32145
32147
12,394
6,217
5,163
1,270
741
3,641
484
0.17
0.09
0.07
0.02
0.01
0.05
0.01
32209
32211
32212
32214
32215
32216
32217
859
2,255
1,871
667
536
1,551
2,090
0.01
0.03
0.03
0.01
0.01
0.02
0.03
32148
32152
32160
32161
32163
32164
3,160
557
5,793
2,726
449
4,390
0.04
0.01
0.08
0.04
0.01
0.06
32218
32219
32220
32221
32222
32224
1,110
1,733
1,674
427
5,578
987
0.02
0.02
0.02
0.01
0.08
0.01
32165
32166
32167
32168
32169
32170
456
3,868
3,374
4,341
1,375
978
0.01
0.05
0.05
0.06
0.02
0.01
32225
32226
32228
32229
32232
32233
5,834
10,881
534
2,169
2,698
1,435
0.08
0.15
0.01
0.03
0.04
0.02
84
Appendix B: VLT ID Number Distribution
VLT ID
32240
32242
32243
Frequency
2,437
995
395
% of Observations
0.03
0.01
0.01
VLT ID
32292
32295
32297
Frequency
1,148
4,512
1,523
% of Observations
0.02
0.06
0.02
32244
32246
32248
32249
32251
32252
719
4,883
3,816
7,852
3,385
9,665
0.01
0.07
0.05
0.11
0.05
0.13
32311
32314
32316
32317
32318
32319
1,936
643
3,493
7,468
1,524
12,960
0.03
0.01
0.05
0.1
0.02
0.18
32253
32254
32255
32256
32257
32258
1,359
2,763
1,108
2,310
451
3,191
0.02
0.04
0.02
0.03
0.01
0.04
32320
32321
32322
32324
32325
32326
2,387
36,796
828
3,804
1,710
15,569
0.03
0.51
0.01
0.05
0.02
0.21
32259
32260
32261
32264
32265
32267
433
10,759
13,442
611
6,718
1,007
0.01
0.15
0.18
0.01
0.09
0.01
32327
32328
32330
32333
32334
32335
14,913
1,107
3,419
5,927
1,154
1,704
0.2
0.02
0.05
0.08
0.02
0.02
32268
32270
32271
32272
32273
32274
32275
7,126
1,112
506
2,340
824
2,603
5,821
0.1
0.02
0.01
0.03
0.01
0.04
0.08
32336
32337
32338
32340
32341
32342
32343
1,818
2,168
4,655
822
1,618
1,299
1,392
0.02
0.03
0.06
0.01
0.02
0.02
0.02
32276
32277
32278
32279
32281
32283
3,261
1,959
6,009
1,415
775
6,407
0.04
0.03
0.08
0.02
0.01
0.09
32345
32346
32347
32350
32351
32362
14,473
465
1,440
486
591
503
0.2
0.01
0.02
0.01
0.01
0.01
32285
32287
32288
32289
1,222
983
2,871
4,219
0.02
0.01
0.04
0.06
32366
32367
32375
32376
1,718
398
5,340
1,817
0.02
0.01
0.07
0.02
85
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
32377
32378
32380
32382
32384
32385
2,552
2,673
3,650
2,565
4,532
1,627
0.04
0.04
0.05
0.04
0.06
0.02
32434
32436
32437
32438
32441
32446
3,531
1,253
4,487
2,861
1,380
2,035
0.05
0.02
0.06
0.04
0.02
0.03
32386
32387
32388
32389
32390
32391
1,595
1,087
3,272
2,906
1,916
831
0.02
0.01
0.04
0.04
0.03
0.01
32447
32449
32450
32451
32452
32455
1,417
5,204
780
2,200
416
2,169
0.02
0.07
0.01
0.03
0.01
0.03
32392
32393
32396
32398
32399
32400
1,110
3,284
1,514
2,154
10,135
1,723
0.02
0.05
0.02
0.03
0.14
0.02
32456
32457
32458
32461
32467
32469
472
1,230
1,209
1,484
1,780
706
0.01
0.02
0.02
0.02
0.02
0.01
32403
32404
32405
32406
32407
32408
32409
437
4,907
4,467
1,037
12,113
4,831
1,581
0.01
0.07
0.06
0.01
0.17
0.07
0.02
32470
32474
32477
32478
32479
32480
32481
586
1,189
945
718
3,637
2,979
923
0.01
0.02
0.01
0.01
0.05
0.04
0.01
32410
32411
32412
32413
32414
32415
1,625
507
1,263
998
4,952
2,395
0.02
0.01
0.02
0.01
0.07
0.03
32482
32483
32485
32486
32488
32489
3,526
1,025
348
482
1,371
5,956
0.05
0.01
0
0.01
0.02
0.08
32426
32427
32428
32429
32431
32432
1,513
4,468
1,998
2,127
799
806
0.02
0.06
0.03
0.03
0.01
0.01
32490
32491
32492
32493
32494
32495
8,238
1,364
5,728
5,282
722
8,834
0.11
0.02
0.08
0.07
0.01
0.12
32433
2,323
0.03
32496
3,543
0.05
86
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
32497
32498
32499
32501
32502
32503
1,842
1,073
394
2,879
1,153
6,180
0.03
0.01
0.01
0.04
0.02
0.08
32559
32560
32565
32566
32567
32568
1,296
838
1,388
825
819
1,731
0.02
0.01
0.02
0.01
0.01
0.02
32504
32505
32506
32507
32508
32509
13,540
1,174
7,154
2,279
680
6,071
0.19
0.02
0.1
0.03
0.01
0.08
32571
32574
32575
32576
32577
32578
1,014
6,227
1,584
2,953
1,938
4,843
0.01
0.09
0.02
0.04
0.03
0.07
32510
32511
32512
32520
32521
32522
1,213
760
545
895
3,305
4,496
0.02
0.01
0.01
0.01
0.05
0.06
32579
32580
32581
32582
32583
32584
32,402
6,946
2,188
2,445
3,070
921
0.44
0.1
0.03
0.03
0.04
0.01
32523
32525
32526
32527
32528
32530
32532
988
602
734
2,246
720
630
383
0.01
0.01
0.01
0.03
0.01
0.01
0.01
32585
32586
32588
32589
32591
32592
32593
1,179
37,032
3,587
7,725
2,857
3,335
577
0.02
0.51
0.05
0.11
0.04
0.05
0.01
32534
32535
32537
32541
32542
32543
394
1,998
1,817
7,547
485
1,022
0.01
0.03
0.02
0.1
0.01
0.01
32594
32597
32598
32599
32600
32601
8,342
6,332
1,271
7,102
2,323
31,860
0.11
0.09
0.02
0.1
0.03
0.44
32547
32548
32549
32550
32551
32553
1,700
1,875
2,554
3,844
6,654
2,608
0.02
0.03
0.04
0.05
0.09
0.04
32602
32603
32604
32605
32606
32607
1,536
10,112
3,617
11,466
626
3,731
0.02
0.14
0.05
0.16
0.01
0.05
32558
556
0.01
32608
3,520
0.05
87
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
32609
32611
32613
32614
32616
32617
11,631
3,374
1,593
873
2,751
5,561
0.16
0.05
0.02
0.01
0.04
0.08
32707
32708
32709
32717
32718
32732
927
942
562
657
752
613
0.01
0.01
0.01
0.01
0.01
0.01
32621
32622
32623
32624
32625
32633
1,604
3,807
3,643
1,076
787
633
0.02
0.05
0.05
0.01
0.01
0.01
32734
32737
32739
32740
32742
32744
1,302
392
419
473
1,035
916
0.02
0.01
0.01
0.01
0.01
0.01
32634
32635
32636
32638
32639
32640
1,310
1,270
499
1,535
867
553
0.02
0.02
0.01
0.02
0.01
0.01
32746
32749
32750
32751
32752
32753
2,148
1,120
2,059
3,714
998
540
0.03
0.02
0.03
0.05
0.01
0.01
32642
32644
32645
32646
32647
32648
32652
404
1,224
2,165
590
1,137
617
525
0.01
0.02
0.03
0.01
0.02
0.01
0.01
32754
32755
32757
32759
32760
32763
32765
1,239
571
1,730
884
1,840
1,197
2,519
0.02
0.01
0.02
0.01
0.03
0.02
0.03
32653
32654
32655
32656
32658
32661
782
1,132
2,539
2,831
1,835
694
0.01
0.02
0.03
0.04
0.03
0.01
32777
32778
32781
32783
32785
32786
2,476
660
1,475
631
1,429
2,068
0.03
0.01
0.02
0.01
0.02
0.03
32662
32663
32664
32667
32675
32696
1,331
3,284
1,552
869
412
555
0.02
0.05
0.02
0.01
0.01
0.01
32787
32792
32793
32795
32797
32799
508
1,604
2,583
1,635
1,194
636
0.01
0.02
0.04
0.02
0.02
0.01
32705
467
0.01
32801
1,106
0.02
88
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
32802
32803
32812
32813
32815
32816
1,853
1,515
575
1,062
521
4,387
0.03
0.02
0.01
0.01
0.01
0.06
32887
32888
32890
32895
32896
32898
2,962
4,271
2,011
1,355
5,271
640
0.04
0.06
0.03
0.02
0.07
0.01
32817
32818
32823
32824
32826
32827
785
910
1,383
1,174
2,615
1,394
0.01
0.01
0.02
0.02
0.04
0.02
32904
32906
32907
32908
32909
32913
1,011
3,383
1,518
630
1,552
2,613
0.01
0.05
0.02
0.01
0.02
0.04
32830
32831
32835
32836
32837
32838
732
649
505
646
402
518
0.01
0.01
0.01
0.01
0.01
0.01
32918
32919
32923
32926
32940
32946
536
1,442
530
797
531
692
0.01
0.02
0.01
0.01
0.01
0.01
32842
32843
32846
32848
32849
32850
32853
1,362
1,084
3,543
409
1,561
2,724
604
0.02
0.01
0.05
0.01
0.02
0.04
0.01
32947
32949
32950
32951
32952
32955
32956
1,645
1,209
3,747
1,049
11,811
3,050
1,528
0.02
0.02
0.05
0.01
0.16
0.04
0.02
32855
32856
32865
32866
32873
32874
1,230
674
1,881
510
791
1,803
0.02
0.01
0.03
0.01
0.01
0.02
32957
32958
32959
32961
32964
32966
4,462
3,456
582
1,415
2,657
385
0.06
0.05
0.01
0.02
0.04
0.01
32876
32878
32879
32880
32882
32884
967
1,935
868
1,525
1,026
563
0.01
0.03
0.01
0.02
0.01
0.01
32968
32969
32973
32975
32976
32978
492
939
731
404
542
4,049
0.01
0.01
0.01
0.01
0.01
0.06
32886
935
0.01
33001
1,003
0.01
89
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
33002
33003
33004
33005
33007
33009
678
554
525
737
997
1,082
0.01
0.01
0.01
0.01
0.01
0.01
33102
33116
33117
33119
33124
33125
372
394
543
790
1,836
12,237
0.01
0.01
0.01
0.01
0.03
0.17
33011
33012
33013
33015
33016
33017
1,762
640
478
1,030
1,819
833
0.02
0.01
0.01
0.01
0.02
0.01
33126
33127
33130
33131
33132
33133
1,969
1,204
1,810
496
2,671
604
0.03
0.02
0.02
0.01
0.04
0.01
33019
33026
33028
33029
33034
33036
1,387
551
705
2,474
768
947
0.02
0.01
0.01
0.03
0.01
0.01
33134
33135
33136
33138
33139
33141
932
1,395
609
1,426
5,161
2,095
0.01
0.02
0.01
0.02
0.07
0.03
33038
33039
33044
33045
33057
33058
33059
1,538
430
887
799
704
1,553
913
0.02
0.01
0.01
0.01
0.01
0.02
0.01
33142
33143
33144
33148
33150
33151
33157
584
1,399
552
742
696
1,218
434
0.01
0.02
0.01
0.01
0.01
0.02
0.01
33060
33064
33065
33067
33069
33070
463
605
871
462
2,579
492
0.01
0.01
0.01
0.01
0.04
0.01
33158
33166
33170
33175
33179
33182
842
482
476
1,795
389
621
0.01
0.01
0.01
0.02
0.01
0.01
33079
33080
33082
33084
33089
33095
747
508
1,123
601
702
661
0.01
0.01
0.02
0.01
0.01
0.01
33183
33184
33185
33186
33188
33193
514
3,879
1,346
3,301
664
407
0.01
0.05
0.02
0.05
0.01
0.01
33099
514
0.01
33196
431
0.01
90
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
33199
33202
33209
33211
33212
33214
1,189
801
1,387
781
402
1,180
0.02
0.01
0.02
0.01
0.01
0.02
33282
33284
33285
33286
33287
33288
444
6,725
1,422
4,837
12,265
6,453
0.01
0.09
0.02
0.07
0.17
0.09
33215
33216
33217
33218
33219
33222
9,144
2,080
414
444
479
13,238
0.13
0.03
0.01
0.01
0.01
0.18
33293
33295
33296
33300
33303
33305
2,036
717
2,927
1,635
509
622
0.03
0.01
0.04
0.02
0.01
0.01
33224
33225
33227
33228
33230
33232
744
2,002
645
624
2,860
1,472
0.01
0.03
0.01
0.01
0.04
0.02
33321
33322
33324
33328
33332
33333
662
2,763
696
592
886
738
0.01
0.04
0.01
0.01
0.01
0.01
33233
33234
33235
33236
33238
33240
33242
411
1,641
527
5,779
2,115
3,605
1,050
0.01
0.02
0.01
0.08
0.03
0.05
0.01
33335
33336
33337
33339
33340
33341
33342
3,383
3,197
1,271
460
661
403
2,399
0.05
0.04
0.02
0.01
0.01
0.01
0.03
33243
33244
33245
33247
33249
33250
957
3,311
825
471
541
1,239
0.01
0.05
0.01
0.01
0.01
0.02
33343
33344
33351
33357
33360
33361
1,703
856
877
1,318
2,645
580
0.02
0.01
0.01
0.02
0.04
0.01
33251
33255
33256
33264
33266
33271
2,399
1,204
600
388
789
453
0.03
0.02
0.01
0.01
0.01
0.01
33362
33365
33367
33369
33374
33375
1,656
1,053
929
573
1,896
413
0.02
0.01
0.01
0.01
0.03
0.01
33281
769
0.01
33376
3,676
0.05
91
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
33377
33378
33379
33380
33381
33382
3,348
9,072
12,400
1,324
1,406
427
0.05
0.12
0.17
0.02
0.02
0.01
33447
33448
33449
33453
33454
33455
14,877
1,479
2,378
458
4,898
2,484
0.2
0.02
0.03
0.01
0.07
0.03
33383
33385
33386
33388
33390
33391
7,842
2,960
3,354
607
2,837
583
0.11
0.04
0.05
0.01
0.04
0.01
33456
33457
33458
33459
33460
33461
831
4,422
2,750
2,837
4,791
5,440
0.01
0.06
0.04
0.04
0.07
0.07
33393
33394
33398
33406
33407
33408
4,299
1,018
1,447
2,655
3,516
6,638
0.06
0.01
0.02
0.04
0.05
0.09
33462
33463
33464
33467
33468
33469
564
1,929
1,689
629
1,298
1,216
0.01
0.03
0.02
0.01
0.02
0.02
33409
33410
33412
33413
33416
33417
33418
1,941
932
468
4,078
976
3,002
4,261
0.03
0.01
0.01
0.06
0.01
0.04
0.06
33470
33471
33472
33473
33474
33475
33478
808
585
2,411
1,867
1,391
542
1,289
0.01
0.01
0.03
0.03
0.02
0.01
0.02
33420
33421
33425
33426
33428
33429
781
490
4,457
3,596
430
2,402
0.01
0.01
0.06
0.05
0.01
0.03
33479
33480
33482
33483
33484
33485
3,307
5,992
402
1,980
2,414
578
0.05
0.08
0.01
0.03
0.03
0.01
33432
33435
33436
33437
33438
33444
2,442
1,719
2,643
8,389
762
3,324
0.03
0.02
0.04
0.12
0.01
0.05
33490
33493
33494
33500
33501
33503
2,101
2,747
568
2,045
3,820
1,438
0.03
0.04
0.01
0.03
0.05
0.02
33446
774
0.01
33504
5,177
0.07
92
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
33505
33506
33507
33508
33509
33510
2,210
5,076
1,045
2,220
949
2,013
0.03
0.07
0.01
0.03
0.01
0.03
33575
33576
33577
33578
33579
33580
2,049
655
2,451
31,069
11,786
40,408
0.03
0.01
0.03
0.43
0.16
0.55
33511
33512
33513
33514
33515
33520
3,573
7,177
13,129
566
5,707
2,597
0.05
0.1
0.18
0.01
0.08
0.04
33581
33585
33586
33587
33588
33589
2,035
14,865
5,601
23,144
11,636
6,621
0.03
0.2
0.08
0.32
0.16
0.09
33522
33524
33526
33528
33529
33530
8,534
965
401
1,042
462
1,642
0.12
0.01
0.01
0.01
0.01
0.02
33590
33591
33592
33593
33594
33595
8,582
1,192
5,068
1,565
1,545
5,876
0.12
0.02
0.07
0.02
0.02
0.08
33531
33532
33533
33535
33536
33537
33538
425
3,260
3,966
8,113
6,531
2,334
3,608
0.01
0.04
0.05
0.11
0.09
0.03
0.05
33597
33598
33601
33602
33603
33604
33605
614
421
2,712
6,349
558
865
770
0.01
0.01
0.04
0.09
0.01
0.01
0.01
33539
33541
33545
33547
33548
33549
696
404
5,142
1,149
2,568
5,436
0.01
0.01
0.07
0.02
0.04
0.07
33606
33608
33609
33610
33612
33613
1,534
434
1,328
637
9,549
3,673
0.02
0.01
0.02
0.01
0.13
0.05
33552
33553
33557
33558
33559
33560
836
1,102
1,976
669
640
462
0.01
0.02
0.03
0.01
0.01
0.01
33614
33616
33617
33624
33627
33629
1,545
739
486
607
430
573
0.02
0.01
0.01
0.01
0.01
0.01
33570
1,880
0.03
33638
1,534
0.02
93
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
33639
33640
33641
33642
33643
33644
1,391
418
489
909
1,304
2,140
0.02
0.01
0.01
0.01
0.02
0.03
33695
33696
33697
33698
33700
33701
1,689
1,321
553
1,376
2,763
1,935
0.02
0.02
0.01
0.02
0.04
0.03
33645
33646
33647
33648
33649
33651
849
500
960
1,002
897
3,599
0.01
0.01
0.01
0.01
0.01
0.05
33711
33712
33713
33714
33715
33719
3,936
5,386
1,428
2,527
1,781
1,487
0.05
0.07
0.02
0.03
0.02
0.02
33653
33655
33656
33657
33658
33659
2,756
4,520
381
1,025
875
2,812
0.04
0.06
0.01
0.01
0.01
0.04
33720
33722
33723
33724
33728
33729
1,420
739
1,115
464
8,515
1,462
0.02
0.01
0.02
0.01
0.12
0.02
33660
33661
33663
33664
33665
33670
33671
2,192
2,484
4,711
477
1,426
524
1,786
0.03
0.03
0.06
0.01
0.02
0.01
0.02
33730
33736
33737
33739
33740
33741
33743
738
2,743
979
3,238
478
10,742
863
0.01
0.04
0.01
0.04
0.01
0.15
0.01
33674
33680
33681
33682
33684
33686
4,835
647
4,676
7,768
531
1,799
0.07
0.01
0.06
0.11
0.01
0.02
33748
33749
33750
33751
33752
33753
4,650
1,450
1,597
658
616
1,051
0.06
0.02
0.02
0.01
0.01
0.01
33687
33688
33689
33690
33691
33692
3,535
6,185
408
4,664
2,826
952
0.05
0.08
0.01
0.06
0.04
0.01
33754
33758
33762
33763
33766
33768
1,111
883
1,138
954
952
452
0.02
0.01
0.02
0.01
0.01
0.01
33694
3,551
0.05
33771
465
0.01
94
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
33772
33777
33784
33785
33786
33787
802
321
3,551
4,442
1,377
1,054
0.01
0
0.05
0.06
0.02
0.01
33832
33834
33837
33839
33840
33841
972
438
477
512
1,467
1,028
0.01
0.01
0.01
0.01
0.02
0.01
33788
33790
33792
33794
33796
33799
653
1,382
2,506
1,580
698
4,980
0.01
0.02
0.03
0.02
0.01
0.07
33844
33845
33852
33853
33856
33857
1,121
483
5,825
773
377
1,347
0.02
0.01
0.08
0.01
0.01
0.02
33800
33801
33802
33803
33804
33805
28,476
3,300
5,339
2,974
7,103
1,219
0.39
0.05
0.07
0.04
0.1
0.02
33858
33859
33861
33862
33863
33865
2,721
3,141
15,222
8,134
15,721
2,560
0.04
0.04
0.21
0.11
0.22
0.04
33806
33807
33808
33809
33810
33811
33812
5,252
4,439
5,718
1,238
3,077
15,749
852
0.07
0.06
0.08
0.02
0.04
0.22
0.01
33866
33867
33868
33872
33873
33875
33876
2,478
1,208
645
3,238
543
1,338
618
0.03
0.02
0.01
0.04
0.01
0.02
0.01
33813
33814
33815
33816
33817
33818
7,460
1,774
4,774
1,575
973
4,031
0.1
0.02
0.07
0.02
0.01
0.06
33877
33878
33881
33884
33885
33886
2,131
971
402
6,663
783
3,043
0.03
0.01
0.01
0.09
0.01
0.04
33819
33820
33822
33826
33828
33829
783
2,951
1,361
1,264
1,156
2,071
0.01
0.04
0.02
0.02
0.02
0.03
33887
33888
33889
33890
33891
33892
1,253
9,614
20,230
3,940
1,007
2,227
0.02
0.13
0.28
0.05
0.01
0.03
33831
590
0.01
33893
2,007
0.03
95
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
33894
33895
33897
33898
33900
33901
2,378
47,651
2,889
7,157
2,365
453
0.03
0.65
0.04
0.1
0.03
0.01
33955
33956
33958
33959
33960
33965
1,014
465
718
429
574
924
0.01
0.01
0.01
0.01
0.01
0.01
33902
33903
33904
33905
33906
33907
9,048
6,164
2,016
2,194
1,312
4,733
0.12
0.08
0.03
0.03
0.02
0.06
33969
33989
34005
34025
34027
34039
409
605
700
386
1,448
789
0.01
0.01
0.01
0.01
0.02
0.01
33908
33909
33910
33911
33913
33914
1,750
27,986
11,854
20,365
5,201
6,784
0.02
0.38
0.16
0.28
0.07
0.09
34043
34044
34045
34047
34050
34052
895
551
1,714
616
1,530
464
0.01
0.01
0.02
0.01
0.02
0.01
33915
33916
33921
33923
33924
33926
33927
16,352
4,228
1,278
1,061
7,395
4,206
4,683
0.22
0.06
0.02
0.01
0.1
0.06
0.06
34053
34057
34059
34070
34075
34081
34083
414
606
527
1,667
805
431
806
0.01
0.01
0.01
0.02
0.01
0.01
0.01
33928
33932
33933
33936
33937
33941
684
518
793
2,455
528
1,769
0.01
0.01
0.01
0.03
0.01
0.02
34084
34091
34092
34093
34099
34104
444
602
3,540
387
516
415
0.01
0.01
0.05
0.01
0.01
0.01
33942
33944
33950
33951
33952
33953
524
416
513
477
1,335
822
0.01
0.01
0.01
0.01
0.02
0.01
34116
34117
34123
34132
34133
34134
1,779
772
476
663
416
1,048
0.02
0.01
0.01
0.01
0.01
0.01
33954
1,486
0.02
34135
871
0.01
96
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
34136
34154
34159
34160
34166
34168
442
792
487
784
649
1,048
0.01
0.01
0.01
0.01
0.01
0.01
34279
34280
34290
34291
34318
34326
615
420
1,653
385
625
391
0.01
0.01
0.02
0.01
0.01
0.01
34173
34174
34183
34184
34185
34187
703
494
600
1,097
1,395
852
0.01
0.01
0.01
0.02
0.02
0.01
34330
34331
34341
34351
34358
34359
522
1,202
1,028
409
1,981
625
0.01
0.02
0.01
0.01
0.03
0.01
34188
34190
34199
34200
34202
34203
399
2,650
944
535
1,572
975
0.01
0.04
0.01
0.01
0.02
0.01
34363
34365
34366
34369
34370
34374
1,688
408
848
1,070
493
380
0.02
0.01
0.01
0.01
0.01
0.01
34205
34206
34211
34218
34254
34258
34259
518
644
788
523
1,339
1,067
5,515
0.01
0.01
0.01
0.01
0.02
0.01
0.08
34375
34385
34386
34387
34389
34399
34403
478
870
715
1,018
553
1,590
401
0.01
0.01
0.01
0.01
0.01
0.02
0.01
34260
34261
34262
34263
34264
34265
2,317
6,439
945
3,480
700
1,223
0.03
0.09
0.01
0.05
0.01
0.02
34408
34413
34414
34420
34445
34452
1,359
697
432
1,147
431
371
0.02
0.01
0.01
0.02
0.01
0.01
34266
34267
34268
34269
34271
34272
2,919
1,259
664
639
480
455
0.04
0.02
0.01
0.01
0.01
0.01
34455
34456
34457
34459
34461
34462
3,253
541
489
815
929
725
0.04
0.01
0.01
0.01
0.01
0.01
34274
924
0.01
34463
485
0.01
97
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
34465
34466
34467
34469
34470
34475
776
2,014
628
1,647
531
681
0.01
0.03
0.01
0.02
0.01
0.01
34737
34738
34739
34741
34742
34743
2,095
970
5,880
656
1,557
2,429
0.03
0.01
0.08
0.01
0.02
0.03
34478
34509
34512
34523
34525
34531
522
554
985
2,765
532
401
0.01
0.01
0.01
0.04
0.01
0.01
34744
34745
34746
34747
34748
34749
904
761
893
1,082
434
1,705
0.01
0.01
0.01
0.01
0.01
0.02
34547
34549
34551
34553
34554
34558
722
415
925
7,769
2,490
6,355
0.01
0.01
0.01
0.11
0.03
0.09
34751
34752
34753
34756
34758
34759
471
1,346
828
2,453
3,462
1,690
0.01
0.02
0.01
0.03
0.05
0.02
34559
34560
34563
34568
34570
34571
34573
747
1,400
490
884
1,218
480
388
0.01
0.02
0.01
0.01
0.02
0.01
0.01
34760
34761
34762
34763
34764
34765
34766
777
486
692
1,823
457
565
2,645
0.01
0.01
0.01
0.03
0.01
0.01
0.04
34574
34576
34577
34578
34580
34583
454
1,494
1,370
2,156
682
541
0.01
0.02
0.02
0.03
0.01
0.01
34767
34768
34769
34770
34772
34775
769
1,434
2,287
1,077
1,530
387
0.01
0.02
0.03
0.01
0.02
0.01
34585
34588
34590
34591
34592
34597
3,441
483
1,932
630
495
646
0.05
0.01
0.03
0.01
0.01
0.01
34776
34778
34779
34780
34781
34782
1,331
624
1,060
627
9,287
668
0.02
0.01
0.01
0.01
0.13
0.01
34736
3,220
0.04
34783
1,403
0.02
98
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
34784
34785
34789
34790
34791
34792
550
1,621
702
407
940
540
0.01
0.02
0.01
0.01
0.01
0.01
34848
34863
34865
34869
34877
34878
449
1,108
427
498
648
760
0.01
0.02
0.01
0.01
0.01
0.01
34793
34794
34795
34796
34799
34800
562
4,483
20,995
10,261
6,084
2,003
0.01
0.06
0.29
0.14
0.08
0.03
34888
34889
34894
34908
34915
34916
685
580
475
489
2,233
5,511
0.01
0.01
0.01
0.01
0.03
0.08
34801
34803
34807
34808
34810
34811
1,478
539
1,597
918
643
1,685
0.02
0.01
0.02
0.01
0.01
0.02
34922
34924
34926
34928
34931
34932
456
1,201
941
686
913
1,431
0.01
0.02
0.01
0.01
0.01
0.02
34815
34816
34817
34818
34819
34820
34824
3,031
2,493
432
1,451
1,043
395
478
0.04
0.03
0.01
0.02
0.01
0.01
0.01
34941
34942
34944
34958
34959
34962
34972
1,267
1,066
452
779
727
439
2,624
0.02
0.01
0.01
0.01
0.01
0.01
0.04
34825
34828
34829
34830
34834
34837
389
1,083
1,306
538
677
605
0.01
0.01
0.02
0.01
0.01
0.01
34976
34977
34978
34989
34991
34992
390
613
400
463
671
505
0.01
0.01
0.01
0.01
0.01
0.01
34838
34839
34842
34843
34845
34846
419
512
1,101
1,209
715
638
0.01
0.01
0.02
0.02
0.01
0.01
34993
34994
35006
35011
35023
35039
623
3,333
762
805
486
459
0.01
0.05
0.01
0.01
0.01
0.01
34847
2,356
0.03
35041
390
0.01
99
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
35045
35053
35055
35058
35059
35061
548
660
417
1,017
2,336
849
0.01
0.01
0.01
0.01
0.03
0.01
35191
35194
35197
35198
35199
35201
584
1,607
4,293
6,982
2,447
979
0.01
0.02
0.06
0.1
0.03
0.01
35062
35063
35065
35068
35072
35074
975
1,184
897
483
2,561
1,236
0.01
0.02
0.01
0.01
0.04
0.02
35203
35204
35206
35209
35211
35218
1,358
1,654
3,469
888
417
1,049
0.02
0.02
0.05
0.01
0.01
0.01
35077
35078
35079
35080
35087
35090
669
1,392
677
474
971
674
0.01
0.02
0.01
0.01
0.01
0.01
35221
35222
35226
35228
35229
35233
503
1,333
1,059
2,434
390
1,284
0.01
0.02
0.01
0.03
0.01
0.02
35092
35101
35103
35104
35105
35113
35126
447
549
1,346
1,109
1,095
1,086
666
0.01
0.01
0.02
0.02
0.02
0.01
0.01
35237
35243
35245
35248
35249
35251
35257
443
452
439
815
548
1,296
436
0.01
0.01
0.01
0.01
0.01
0.02
0.01
35134
35142
35157
35158
35162
35168
476
441
4,438
1,797
545
692
0.01
0.01
0.06
0.02
0.01
0.01
35276
35277
35278
35279
35290
35291
464
601
431
400
2,413
428
0.01
0.01
0.01
0.01
0.03
0.01
35169
35177
35178
35179
35181
35183
717
631
492
9,674
735
6,711
0.01
0.01
0.01
0.13
0.01
0.09
35292
35293
35294
35295
35296
35297
2,879
916
439
577
546
515
0.04
0.01
0.01
0.01
0.01
0.01
35187
2,704
0.04
35298
1,096
0.02
100
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
35299
35301
50009
50010
50013
50014
639
1,213
394
2,509
400
3,212
0.01
0.02
0.01
0.03
0.01
0.04
51899
51943
51959
51967
51985
52045
1,083
558
3,521
2,415
597
693
0.01
0.01
0.05
0.03
0.01
0.01
50017
50018
50021
50022
50025
50026
577
3,782
690
4,962
659
6,106
0.01
0.05
0.01
0.07
0.01
0.08
52059
52063
52087
52095
52111
52135
2,295
449
1,536
718
398
487
0.03
0.01
0.02
0.01
0.01
0.01
50029
50030
50033
50034
50035
50036
780
5,318
896
8,261
556
4,531
0.01
0.07
0.01
0.11
0.01
0.06
52171
52175
52197
52199
52209
52255
2,390
1,369
1,736
853
1,001
940
0.03
0.02
0.02
0.01
0.01
0.01
50038
50039
50040
50041
50042
50043
50044
3,901
457
606
899
545
4,139
3,308
0.05
0.01
0.01
0.01
0.01
0.06
0.05
52265
52269
52293
52305
52341
52361
52369
2,756
608
2,514
929
570
4,645
2,739
0.04
0.01
0.03
0.01
0.01
0.06
0.04
50045
50046
51497
51549
51565
51581
940
1,397
981
509
3,183
2,358
0.01
0.02
0.01
0.01
0.04
0.03
52387
52461
52469
52493
52505
52517
1,065
3,251
663
2,078
925
487
0.01
0.04
0.01
0.03
0.01
0.01
51687
51703
51715
51751
51767
51775
550
2,407
576
538
2,722
2,208
0.01
0.03
0.01
0.01
0.04
0.03
52545
52561
52565
52585
52645
52649
660
3,697
1,648
1,091
3,571
834
0.01
0.05
0.02
0.01
0.05
0.01
51859
1,695
0.02
52669
3,682
0.05
101
Appendix B: VLT ID Number Distribution
VLT ID
Frequency
% of Observations
VLT ID
Frequency
% of Observations
52677
52685
52693
52701
52705
52719
1,381
682
576
5,065
3,728
1,339
0.02
0.01
0.01
0.07
0.05
0.02
52891
52895
52961
52982
52998
53016
2,235
648
524
389
795
587
0.03
0.01
0.01
0.01
0.01
0.01
52759
52765
52767
52783
52791
52807
641
2,757
613
1,308
839
1,603
0.01
0.04
0.01
0.02
0.01
0.02
53022
53024
53031
53037
53038
53060
1,191
592
599
1,130
540
2,765
0.02
0.01
0.01
0.02
0.01
0.04
52809
52817
52853
52855
52871
52875
1,220
938
1,644
508
1,407
673
0.02
0.01
0.02
0.01
0.02
0.01
53061
53062
53064
53067
53069
53079
1,447
548
1,901
970
856
657
0.02
0.01
0.03
0.01
0.01
0.01
52887
3,191
0.04
Total Observations
7,284,227
102
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
32608
35126
30700
30047
32750
32330
30739
33150
32521
32243
31704
31856
30797
31523
32380
31494
30940
30583
32827
31476
29701
29660
35296
Odds
Ratio
10.986
10.988
10.997
10.998
11.002
11.002
11.011
11.017
11.020
11.030
11.034
11.060
11.069
11.079
11.081
11.085
11.095
11.096
11.125
11.128
11.149
11.159
11.182
Model
Year
2004
2005
2002
2001
2005
2004
2002
2005
2004
2004
2003
2003
2002
2003
2004
2003
2002
2002
2005
2003
2000
2000
2004
50017
11.186
2002
Multiple Light Truck Makes and Models
35059
30617
33303
29421
51715
33417
11.187
11.187
11.200
11.204
11.206
11.209
2004
2002
2006
2000
2001
2006
50030
11.218
2005
32812
11.227
2005
BMW
325CI COUPE
CHRYSLER
PT CRUISER (Auto)
AUDI
A6
GMC
SAFARI 2WD PASSENGER
DODGE
RAM 3500 DIESEL
CHEVROLET
MALIBU
Multiple Heavy Duty Light Truck Makes and
Models
CHRYSLER
CROSSFIRE
1957
11.235
2002
Multiple Station Wagon Makes and Models
30941
30560
30341
11.238
11.242
11.254
2002
2002
2001
Make
Model (Transmission Type)
Engine Size
TOYOTA
CHEVROLET
FORD
FORD
CHEVROLET
HYUNDAI
FORD
PONTIAC
PONTIAC
FORD
KIA
NISSAN
GMC
FORD
JEEP
FORD
MAZDA
CHEVROLET
DODGE
FORD
TOYOTA
SUBARU
BMW
TUNDRA 2WD
SILVERADO 2500 2WD
EXPLORER 4DR
RANGER SUPER CAB 4DR
SILVERADO 1500 2WD
ACCENT
MUSTANG COUPE
GRAND AM
GRAND AM
E150
SORENTO 2WD
FRONTIER 2WD
K1500 YUKON 4WD
MUSTANG CONVERTIBLE
LIBERTY 2WD
F150 2WD
MX-5 MIATA
K1500 SUBURBAN 4WD
CARAVAN 2WD
ESCAPE
RAV4 4WD
FORESTER AWD
645CI
003.4
006.0
004.0
004.0
004.8
001.6
004.6
003.4
003.4
005.4
003.5
002.4
005.3
003.8
003.7
004.6
001.8
005.3
003.8
003.0
002.0
002.5
004.4
8 Cylinder
Diesel
002.5
002.4
003.1
004.3
005.9
002.2
8 Cylinder
Diesel
003.2
6 Cylinder
Gasoline
002.0
005.3
002.2
MAZDA
CHEVROLET
SATURN
PROTEGE/PROTEGE 5
C1500 TAHOE 2WD
L100/200
103
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
29327
32148
29445
30539
32543
30581
Odds
Ratio
11.254
11.266
11.268
11.268
11.270
11.275
Model
Year
2000
2004
2000
2002
2004
2002
2003
11.276
2003
29273
30831
33453
29736
32389
11.276
11.280
11.285
11.286
11.292
2000
2002
2006
2000
2004
50033
11.299
2006
32824
31143
34924
32787
29612
29285
33382
31515
33810
32259
30557
32367
31303
30698
29294
29707
29317
31319
32199
30584
33119
30103
11.300
11.337
11.339
11.342
11.347
11.374
11.399
11.400
11.405
11.408
11.409
11.410
11.420
11.431
11.436
11.451
11.469
11.476
11.484
11.485
11.487
11.492
2005
2002
2002
2005
2000
2000
2006
2003
2006
2004
2002
2004
2003
2002
2000
2000
2000
2003
2004
2002
2005
2001
CHRYSLER
TOYOTA
CHEVROLET
CHEVROLET
OLDSMOBILE
DODGE
CHEVROLET
FORD
NISSAN
FORD
CHEVROLET
JAGUAR
CHEVROLET
FORD
DODGE
TOYOTA
FORD
CHEVROLET
CHRYSLER
CHEVROLET
MITSUBISHI
HONDA
TOWN & COUNTRY 2WD
RAV4 2WD
SILVERADO 2500 2WD
MONTE CARLO
ALERO
DURANGO 4WD
COBALT
F150 SUPER CREWCAB
QUEST
F150 2WD
SILVERADO 1500 2WD
XJR
AVALANCHE 1500 2WD
EXPEDITION
RAM 1500 4WD
TACOMA 2WD
E250 ECONOLINE
CORVETTE (Auto)
TOWN & COUNTRY
K1500 TAHOE 4WD
LANCER
ODYSSEY
30798
11.510
2002
GMC
K1500 YUKON DENALI
AWD
Make
Model (Transmission Type)
FORD
EXPLORER 4DR
CHEVROLET
CORVETTE
HYUNDAI
ACCENT (Manual)
CADILLAC
ESCALADE AWD
SATURN
L300
CHEVROLET
SILVERADO 1500 4WD
Multiple Full-Size Passenger Car Makes and
Models
DODGE
DAKOTA 2WD
HYUNDAI
SONATA
CHRYSLER
SRT-8
VOLKSWAGEN
PASSAT WAGON
KIA
SEDONA
Multiple Light Truck Makes and Models
Engine Size
004.0
005.7
001.5
006.0
003.0
005.3
8 Cylinder
Gasoline
003.9
002.4
006.1
001.8
003.5
8 Cylinder
Diesel
003.8
002.0
006.0
003.4
002.4
005.9
002.0
004.6
003.5
004.6
005.3
004.2
005.3
005.4
005.9
003.4
005.4
005.7
003.8
004.8
002.0
003.5
006.0
104
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
32101
30537
31144
30793
35078
34742
33804
31084
Odds
Ratio
11.532
11.534
11.539
11.545
11.553
11.564
11.566
11.575
Model
Year
2004
2002
2002
2002
2004
2000
2006
2002
1947
11.576
2002
30618
31014
29281
32209
32398
31005
29729
11.580
11.593
11.599
11.603
11.609
11.609
11.609
2002
2002
2000
2004
2004
2002
2000
BMW
X3
CADILLAC
ELDORADO
TOYOTA
RAV4 AWD
GMC
SIERRA 1500 4WD
FORD
F250 SUPER DUTY
FORD
RANGER REG CAB SHORT
NISSAN
FRONTIER 2WD
SATURN
LW200
Multiple European Compact Passenger Car Makes
and Models
CHRYSLER
PT CRUISER (Manual)
NISSAN
ALTIMA
DODGE
DURANGO 2WD
DODGE
CARAVAN 2WD
LAND ROVER
RANGE ROVER
MITSUBISHI
MONTERO SPORT 2WD
VOLKSWAGEN
NEW BEETLE (Auto)
1976
11.619
2002
Multiple Full-Size SUV Makes and Models
30901
29699
29219
31594
34926
11.628
11.661
11.671
11.687
11.687
2002
2000
2000
2003
2002
2092
11.690
2004
31528
30710
29270
30389
30874
11.693
11.699
11.707
11.720
11.722
2003
2002
2000
2001
2002
2101
11.726
2004
Multiple Mid-Size Van Makes and Models
6 Cylinder
Gasoline
1995
11.729
2002
Multiple Ford Vans and Light Truck Models
006.8
31854
32803
29706
31312
30742
11.732
11.734
11.735
11.742
11.751
2003
2005
2000
2003
2002
Make
LEXUS
TOYOTA
CHEVROLET
GMC
CHEVROLET
Model (Transmission Type)
IS 300
MR2
METRO
K1500 YUKON XL 4WD
SILVERADO 2500 4WD
Multiple Mini-Van Makes and Models
FORD
FORD
DODGE
TOYOTA
JEEP
NISSAN
CHRYSLER
TOYOTA
CHEVROLET
FORD
MUSTANG COUPE
F150 REG CAB LONG
CARAVAN 2WD
ECHO
LIBERTY 2WD
ALTIMA
300C
TACOMA 2WD
C1500 SUBURBAN 2WD
RANGER PICKUP 4WD
Engine Size
003.0
004.6
002.0
005.3
005.4
003.0
004.0
002.2
4 Cylinder
Gasoline
002.4
003.5
005.2
003.8
004.4
003.0
002.0
8 Cylinder
Gasoline
003.0
001.8
001.3
005.3
006.0
6 Cylinder
Gasoline
004.6
004.6
003.8
001.5
003.7
002.5
005.7
002.7
005.3
004.0
105
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
29349
30518
51687
31693
33820
34959
31844
29370
34915
32549
32121
30968
32377
29671
30920
30926
Odds
Ratio
11.759
11.769
11.774
11.784
11.796
11.798
11.811
11.830
11.840
11.847
11.849
11.850
11.871
11.881
11.884
11.884
Model
Year
2000
2002
2001
2003
2006
2002
2003
2000
2002
2004
2004
2002
2004
2000
2002
2002
1867
11.891
2000
33247
30011
32165
31155
31794
29292
32152
33300
31830
11.900
11.904
11.908
11.912
11.933
11.945
11.953
11.955
11.963
2005
2001
2004
2002
2003
2000
2004
2006
2003
VOLKSWAGEN
FORD
CHEVROLET
TOYOTA
MERCEDES
DODGE
CHEVROLET
AUDI
MITSUBISHI
GTI
F150 SUPER CAB LONG
SILVERADO 1500 AWD
TUNDRA 2WD
E500
RAM 1500 2WD
EXPRESS 1500
A4 QUATTRO
ECLIPSE
001.8
005.4
006.0
004.7
005.0
005.9
004.3
002.0
002.4
30775
11.970
2002
GMC
C1500 YUKON XL 2WD
005.3
29188
33256
34807
32338
29277
29670
30647
31190
31556
31019
11.971
11.974
11.996
12.005
12.007
12.021
12.023
12.032
12.032
12.035
2000
2005
2005
2004
2000
2000
2002
2002
2003
2002
CHEVROLET
VOLKSWAGEN
CADILLAC
HYUNDAI
DODGE
SUZUKI
DODGE
VOLVO
GMC
NISSAN
C2500 SILVERADO 2WD
PASSAT
CTS
SONATA
DAKOTA 4WD
GRAND VITARA 2WD
DAKOTA 4WD
S80/S80 EXECUTIVE
SIERRA 1500 2WD
MAXIMA
006.0
001.8
003.6
002.7
004.7
002.5
004.7
002.9
005.3
003.5
Make
Model (Transmission Type)
Engine Size
FORD
BMW
CHEVROLET
JEEP
PONTIAC
FORD
MITSUBISHI
FORD
ACURA
SATURN
CADILLAC
MERCEDES
JEEP
SUZUKI
MAZDA
MAZDA
FOCUS ZX3 3DR
M5
SILVERADO 2500 DIESEL
LIBERTY 2WD
GRAND PRIX
E350 ECONOLINE
MONTERO SPORT 2WD
RANGER SUPER CAB 2DR
RSX
VUE FWD
DEVILLE
ML500
GRAND CHEROKEE 4WD
GRAND VITARA 4WD
626
B2300 REG CAB SHORT
002.0
004.9
006.6
003.7
003.8
005.4
003.0
004.0
002.0
003.5
004.6
005.0
004.0
002.5
002.5
002.3
Multiple Compact SUV Makes and Models
6 Cylinder
Gasoline
106
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
30097
29663
30774
34846
29335
Odds
Ratio
12.044
12.056
12.061
12.072
12.084
Model
Year
2001
2000
2002
2000
2000
2129
12.088
2005
Multiple Full-Size Light Truck Makes and Models
32090
32918
32163
30696
33730
32273
31606
29309
29749
12.101
12.105
12.107
12.126
12.127
12.132
12.134
12.135
12.138
2004
2005
2004
2002
2006
2004
2003
2000
2000
BMW
GMC
CHEVROLET
FORD
MERCEDES
FORD
HONDA
FORD
VOLVO
30753
12.146
2002
FORD
30538
30064
32843
32823
32099
12.155
12.161
12.176
12.177
12.181
2002
2001
2005
2005
2004
CADILLAC
GMC
DODGE
CHRYSLER
BMW
1919
12.189
2001
31304
33264
31495
29656
32138
29149
34093
33255
12.210
12.210
12.216
12.230
12.242
12.244
12.244
12.244
2003
2005
2003
2000
2004
2000
2007
2005
1859
12.252
2000
33117
33815
29487
31630
12.260
12.281
12.292
12.298
2005
2006
2000
2003
Make
Model (Transmission Type)
Engine Size
HONDA
SUBARU
GMC
FORD
FORD
ACCORD
LEGACY AWD
C1500 YUKON 2WD
EXCURSION
F150 REG CAB SHORT
003.0
002.5
005.3
005.4
004.6
8 Cylinder
Gasoline
002.5
004.8
004.8
002.0
003.5
004.2
003.0
004.6
002.4
3-SERIES
SIERRA 1500 2WD
SILVERADO 1500 4WD
ESCORT ZX2
C350
FREESTAR WAGON FWD
ACCORD
CROWN VICTORIA POLICE
S70
RANGER SUPER CAB 2DR
SH
ESCALADE 2WD
C1500 YUKON XL 2WD
MAGNUM
TOWN & COUNTRY 2WD
M3
Multiple Compact SUV Makes and Models
CHEVROLET
VOLKSWAGEN
FORD
SATURN
CHEVROLET
BMW
CHEVROLET
VOLKSWAGEN
AVALANCHE 1500 4WD
TOUAREG
F150 2WD
LW
SILVERADO 1500 2WD
740IL
TRAILBLAZER 2WD
NEW BEETLE
Multiple Full-Size Light Truck Makes and Models
MITSUBISHI
NISSAN
JEEP
HYUNDAI
GALANT
XTERRA 2WD
CHEROKEE 4WD
SONATA
003.0
005.3
005.3
005.7
003.3
003.2
4 Cylinder
Gasoline
005.3
003.2
005.4
003.0
004.3
004.4
004.2
002.0
8 Cylinder
Gasoline
002.4
004.0
004.0
002.4
107
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
32130
31162
32634
31738
29313
30548
31486
29645
32585
31555
30532
29719
29447
34944
29664
30530
30803
Odds
Ratio
12.301
12.317
12.321
12.332
12.333
12.343
12.348
12.363
12.363
12.367
12.373
12.402
12.407
12.409
12.410
12.418
12.427
Model
Year
2004
2002
2004
2003
2000
2002
2003
2000
2004
2003
2002
2000
2000
2002
2000
2002
2002
2156
12.430
2005
32469
33592
31099
30108
29998
33653
12.443
12.446
12.463
12.482
12.485
12.485
2004
2006
2002
2001
2001
2006
50009
12.489
2000
32853
12.491
2005
DODGE
2002
12.492
2003
Multiple Mid-Size Passenger Car Makes and
Models
51775
31989
33447
30096
31526
29420
29446
30558
12.496
12.497
12.519
12.521
12.537
12.539
12.545
12.547
2001
2003
2006
2001
2003
2000
2000
2002
Make
Model (Transmission Type)
Engine Size
CHEVROLET
VOLKSWAGEN
VOLKSWAGEN
LINCOLN
FORD
CHEVROLET
FORD
SAAB
TOYOTA
GMC
BUICK
VOLKSWAGEN
HYUNDAI
FORD
SUBARU
BUICK
GMC
ASTRO 2WD
GOLF
TOUAREG
TOWN CAR
E150 ECONOLINE
AVALANCHE 1500 4WD
EXPLORER 4DR
9-3
CELICA
SIERRA 1500 2WD
PARK AVENUE
GOLF
ELANTRA WAGON
F350 SUPER DUTY
LEGACY WAGON AWD
CENTURY
SAFARI 2WD PASSENGER
004.3
002.0
003.2
004.6
004.2
005.3
004.0
002.0
001.8
004.8
003.8
002.0
002.0
006.8
002.5
003.1
004.3
Multiple Full-Size Van Makes and Models
MERCURY
HUMMER
SUBARU
HYUNDAI
FORD
JEEP
MONTEREY
H3
IMPREZA WAGON AWD
ACCENT
EXPLORER SPORT TRAC
LIBERTY 4WD
Multiple Heavy Duty Light Truck Makes and
Models
FORD
TOYOTA
CHRYSLER
HONDA
FORD
GMC
HYUNDAI
CHEVROLET
RAM 1500 4WD
F350 DIESEL
COROLLA
PT CRUISER
ACCORD
MUSTANG COUPE (Auto)
SAFARI 2WD CARGO
ELANTRA SEDAN
C1500 SUBURBAN 2WD
8 Cylinder
Gasoline
004.2
003.5
002.5
001.5
004.0
003.7
8 Cylinder
Diesel
005.7
6 Cylinder
Gasoline
007.3
001.8
002.4
002.3
003.8
004.3
002.0
005.3
108
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
31268
31835
Odds
Ratio
12.558
12.558
Model
Year
2003
2003
1856
12.558
2000
29697
32826
29138
32314
32170
32019
33459
35058
30403
29439
31028
31386
32137
12.562
12.568
12.574
12.579
12.585
12.592
12.609
12.611
12.613
12.614
12.622
12.633
12.634
2000
2005
2000
2004
2004
2003
2006
2004
2001
2000
2002
2003
2004
2018
12.652
2003
Multiple Compact Light Truck Makes and Models
29754
30169
31293
30659
31831
31265
29759
29683
31021
30714
30533
31698
12.658
12.663
12.670
12.677
12.684
12.685
12.686
12.697
12.707
12.710
12.719
12.727
2000
2001
2003
2002
2003
2003
2000
2000
2002
2002
2002
2003
VOLVO
LEXUS
CADILLAC
DODGE
MITSUBISHI
BMW
VOLVO
TOYOTA
NISSAN
FORD
BUICK
JEEP
2009
12.727
2003
Multiple Sub-Compact Passenger Car Makes and
Models
29753
31026
29375
31432
29358
12.737
12.770
12.777
12.778
12.786
2000
2002
2000
2003
2000
Make
Model (Transmission Type)
Engine Size
BMW
MITSUBISHI
M3
GALANT
003.2
003.0
Multiple Compact Light Truck Makes and Models
TOYOTA
DODGE
BMW
GMC
CHEVROLET
VOLKSWAGEN
DODGE
BMW
TOYOTA
HONDA
NISSAN
CHRYSLER
CHEVROLET
VOLVO
NISSAN
FORD
DODGE
FORD
ECHO
CARAVAN 2WD
323I
SAFARI
MONTE CARLO
GTI
CHARGER
325CI CONVERTIBLE
TACOMA 2WD
ODYSSEY
XTERRA 2WD
PT CRUISER (Auto)
BLAZER 4WD
V40
GS 300
ESCALADE 2WD
RAM 1500 4WD
ECLIPSE
745LI
V70 AWD
4RUNNER 2WD
PATHFINDER 2WD
F150 REG CAB SHORT
REGAL
WRANGLER 4WD
S80
SENTRA (Auto)
RANGER SUPER CAB 2DR
RAM 1500 2WD
MUSTANG COUPE (Manual)
4 Cylinder
Gasoline
001.5
003.3
002.5
004.3
003.8
001.8
002.7
002.5
002.7
003.5
002.4
002.4
004.3
4 Cylinder
Gasoline
001.9
003.0
005.3
004.7
003.0
004.4
002.4
002.7
003.5
005.4
003.8
004.0
4 Cylinder
Gasoline
002.9
002.5
004.0
005.7
004.7
109
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
Odds
Ratio
Model
Year
1966
12.795
2002
Multiple Compact Light Truck Makes and Models
30846
32490
32831
30469
29236
33552
34808
35191
31008
12.813
12.821
12.825
12.829
12.837
12.849
12.853
12.857
12.864
2002
2004
2005
2002
2000
2006
2005
2006
2002
ISUZU
NISSAN
DODGE
AUDI
CHRYSLER
GMC
CHEVROLET
CHEVROLET
MITSUBISHI
50029
12.870
2005
32433
35142
12.884
12.888
2004
2005
2146
12.888
2005
29580
29615
29124
29758
29282
30807
30580
31394
31189
29411
29305
29819
29346
29290
31252
31764
31310
31246
29164
32777
29246
12.927
12.929
12.935
12.938
12.938
12.944
12.956
12.956
12.956
12.963
12.970
12.972
12.982
12.986
12.990
13.000
13.005
13.007
13.015
13.023
13.048
2000
2000
2000
2000
2000
2002
2002
2003
2002
2000
2000
2001
2000
2000
2003
2003
2003
2003
2000
2005
2000
Make
Model (Transmission Type)
RODEO 2WD
ALTIMA
DAKOTA 2WD
A4 QUATTRO (Auto)
300M
CANYON 2WD
EXPRESS 2500
COBALT
MONTERO SPORT 4WD
Multiple Light Truck Makes and Models
MAZDA
FORD
MPV
E350
Multiple Mini-Van Makes and Models
MERCURY
OLDSMOBILE
AUDI
VOLVO
DODGE
GMC
CHEVROLET
CHRYSLER
VOLVO
GMC
FORD
BMW
FORD
DODGE
BMW
MAZDA
CHEVROLET
BMW
BUICK
CHEVROLET
CHRYSLER
SABLE
INTRIGUE
A6 QUATTRO
V70
DURANGO 2WD
SONOMA 2WD
SILVERADO 1500 4WD
SEBRING CONVERTIBLE
S60
K1500 SIERRA 4WD
CONTOUR
740I
FOCUS 4-DR SEDAN
RAM 1500 2WD
330CI
PROTEGE/PROTEGE 5
SILVERADO 1500 2WD
325CI
REGAL
IMPALA
TOWN & COUNTRY 2WD
Engine Size
6 Cylinder
Gasoline
003.2
002.5
004.7
001.8
003.5
002.8
004.8
002.4
003.5
8 Cylinder
Diesel
003.0
005.4
6 Cylinder
Gasoline
003.0
003.5
002.8
002.4
005.9
004.3
004.8
002.7
002.4
005.3
002.0
004.4
002.0
3.9
003.0
002.0
004.8
002.5
003.8
003.4
003.8
110
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
34817
31006
30404
32432
32849
30994
33803
29700
Odds
Ratio
13.060
13.060
13.093
13.095
13.096
13.097
13.100
13.109
Model
Year
2006
2002
2001
2004
2005
2002
2006
2000
1948
13.121
2002
29863
32161
13.133
13.135
2001
2004
CHEVROLET
EXPRESS 2500
MITSUBISHI
MONTERO SPORT 2WD
TOYOTA
TACOMA 2WD
MAZDA
MIATA
DODGE
RAM 1500 2WD
MITSUBISHI
ECLIPSE
NISSAN
FRONTIER 2WD
TOYOTA
RAV4 2WD
Multiple Mid-Size Passenger Car Makes and
Models
CHEVROLET
CORVETTE
CHEVROLET
IMPALA
1876
13.138
2000
Multiple Mini-Van Makes and Models
2180
13.140
2006
Multiple Compact Light Truck Makes and Models
33729
31001
29724
29684
29817
33881
29442
31142
13.154
13.160
13.162
13.165
13.165
13.165
13.178
13.197
2006
2002
2000
2000
2001
2006
2000
2002
2102
13.206
2004
30578
30993
29662
34758
29622
29488
30348
52095
31627
30882
30689
29533
33448
13.224
13.233
13.236
13.245
13.269
13.274
13.275
13.276
13.284
13.309
13.319
13.327
13.336
2002
2002
2000
2002
2000
2000
2001
2003
2003
2002
2002
2000
2006
Make
MERCEDES
MITSUBISHI
VOLKSWAGEN
TOYOTA
BMW
SUZUKI
HONDA
TOYOTA
Model (Transmission Type)
C280
MIRAGE
JETTA
4RUNNER 2WD
540I
RENO
PRELUDE
MR2
Multiple Full-Size Van Makes and Models
CHEVROLET
MITSUBISHI
SUBARU
BMW
PLYMOUTH
JEEP
SUBARU
DODGE
HYUNDAI
JEEP
FORD
MAZDA
CHRYSLER
K1500 AVALANCHE 4WD
ECLIPSE
IMPREZA AWD
X5
VOYAGER 2WD
GRAND CHEROKEE 2WD
FORESTER AWD
RAM 3500 DIESEL
SANTA FE 2WD
WRANGLER 4WD
E250 ECONOLINE
B2500 REG CAB SHORT
SEBRING
Engine Size
006.0
003.5
003.4
001.8
004.7
003.0
002.5
002.0
6 Cylinder
Gasoline
005.7
003.8
6 Cylinder
Gasoline
4 Cylinder
Gasoline
003.0
001.5
001.8
003.4
004.4
002.0
002.2
001.8
8 Cylinder
Gasoline
005.3
002.4
002.5
003.0
002.4
004.0
002.5
005.9
002.7
004.0
005.4
002.5
002.4
111
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
29241
Odds
Ratio
13.349
Model
Year
2000
2119
13.352
2005
31172
30380
29355
33182
34747
33866
35065
29449
32185
32749
29453
30989
32297
30411
31211
33728
52059
13.353
13.356
13.362
13.367
13.387
13.390
13.391
13.395
13.406
13.407
13.416
13.420
13.459
13.466
13.471
13.472
13.476
2002
2001
2000
2005
2000
2006
2004
2000
2004
2005
2000
2002
2004
2001
2003
2006
2003
VOLKSWAGEN
TOYOTA
FORD
SATURN
GMC
SUBARU
BMW
HYUNDAI
CHRYSLER
CHEVROLET
INFINITI
MERCURY
GMC
TOYOTA
AUDI
MERCEDES
CHEVROLET
29730
13.479
2000
32850
34782
33591
30858
31800
30383
32233
31434
33484
33131
31022
34756
30566
30118
32638
32567
30540
13.491
13.494
13.495
13.499
13.505
13.515
13.532
13.538
13.543
13.545
13.547
13.550
13.558
13.574
13.585
13.586
13.604
2005
2003
2006
2002
2003
2001
2004
2003
2006
2005
2002
2002
2002
2001
2004
2004
2002
Make
Model (Transmission Type)
Engine Size
CHRYSLER
LHS
NEW BEETLE (Auto)
CAMRY
MUSTANG COUPE (Auto)
VUE FWD
SONOMA 2WD
IMPREZA
745LI
SONATA
300M
SILVERADO 1500 2WD
G20
SABLE
ENVOY 4WD
TUNDRA 4WD
A4
C230
SILVERADO 2500 DIESEL
003.5
6 Cylinder
Gasoline
002.0
003.0
003.8
002.2
002.2
002.5
004.4
002.4
003.5
004.3
002.0
003.0
004.2
004.7
001.8
002.5
006.6
VOLKSWAGEN
NEW BEETLE (Manual)
002.0
DODGE
MITSUBISHI
HONDA
JAGUAR
MERCEDES
TOYOTA
DODGE
DODGE
DODGE
NISSAN
NISSAN
BMW
CHEVROLET
INFINITI
VOLVO
SUZUKI
CADILLAC
RAM 1500 2WD
ECLIPSE
S2000
X-TYPE
ML500
CAMRY SOLARA
STRATUS
RAM 1500 2WD
STRATUS
FRONTIER 4WD
PATHFINDER 4WD
525I
CORVETTE
G20
S40
AERIO
ESCALADE EXT AWD
005.7
003.0
002.2
002.5
005.0
003.0
002.7
003.7
002.4
004.0
003.5
002.5
005.7
002.0
001.9
002.3
006.0
Multiple Station Wagon Makes and Models
112
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
34789
29340
29245
29274
Odds
Ratio
13.604
13.608
13.612
13.613
Model
Year
2004
2000
2000
2000
1868
13.623
2000
Multiple Full-Size SUV Makes and Models
32961
31205
29171
13.625
13.632
13.653
2005
2003
2000
1840
13.680
2000
31763
13.694
2003
HYUNDAI
ACCENT
ACURA
RSX
CHEVROLET
ASTRO 2WD PASSENGER
Multiple Mid-Size Passenger Car Makes and
Models
MAZDA
MX-5 MIATA
1886
13.696
2000
Multiple Full-Size Van Makes and Models
29614
30799
30482
30445
29163
30943
29726
29291
13.697
13.701
13.702
13.716
13.722
13.731
13.747
13.751
2000
2002
2002
2001
2000
2002
2000
2000
2030
13.753
2003
29169
30708
29544
29688
29323
32267
34803
30229
29518
29267
30017
30855
31134
13.758
13.768
13.775
13.778
13.784
13.793
13.796
13.812
13.837
13.841
13.843
13.846
13.846
2000
2002
2000
2000
2000
2004
2005
2001
2000
2000
2001
2002
2002
1955
13.854
2002
Make
Model (Transmission Type)
Engine Size
CHEVROLET
FORD
CHRYSLER
DODGE
EXPRESS 2500
F150 SUPER CAB SHORT
TOWN & COUNTRY 2WD
DAKOTA 2WD
004.8
004.2
003.3
004.7
OLDSMOBILE
GMC
AUDI
VOLVO
BUICK
MAZDA
VOLKSWAGEN
DODGE
BRAVADA AWD
K1500 YUKON DENALI XL
ALLROAD
S80/S80 EXECUTIVE
PARK AVENUE
TRIBUTE
JETTA
RAM 1500 2WD
Multiple Full-Size SUV Makes and Models
CADILLAC
FORD
MAZDA
TOYOTA
FORD
FORD
BMW
MERCEDES
LINCOLN
DODGE
FORD
JAGUAR
TOYOTA
SEVILLE
F150 REG CAB HD LONG
MX-5 MIATA
CAMRY
ESCORT 4DR
FOCUS
525I
E320
LS
CARAVAN 2WD
F150 SUPER CREWCAB
S-TYPE 4.0 LITRE
COROLLA
Multiple Station Wagon Makes and Models
8 Cylinder
Gasoline
001.6
002.0
004.3
6 Cylinder
Gasoline
001.8
8 Cylinder
Gasoline
004.3
006.0
002.7
002.9
003.8
003.0
002.8
005.2
8 Cylinder
Gasoline
004.6
005.4
001.8
002.2
002.0
002.0
002.5
003.2
003.0
002.4
005.4
004.0
001.8
4 and 6
Cylinder Gas
113
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
30686
34958
31320
32160
30460
29155
30694
30590
35293
32559
32622
30463
29283
52209
33553
30112
29695
31015
Odds
Ratio
13.857
13.869
13.873
13.877
13.887
13.887
13.890
13.901
13.927
13.946
13.957
13.964
13.982
13.988
13.993
13.993
13.995
13.999
Model
Year
2002
2002
2003
2004
2002
2000
2002
2002
2004
2004
2004
2002
2000
2003
2006
2001
2000
2002
1967
13.999
2002
Multiple Full-Size Light Truck Makes and Models
30152
29635
32504
29606
29363
30501
31894
30110
32136
29331
32696
30233
14.013
14.014
14.018
14.023
14.039
14.043
14.043
14.046
14.052
14.055
14.057
14.062
2001
2000
2004
2000
2000
2002
2003
2001
2004
2000
2005
2001
JEEP
PONTIAC
NISSAN
NISSAN
FORD
BMW
PONTIAC
HYUNDAI
CHEVROLET
FORD
BMW
MERCEDES
29372
14.062
2000
29467
29181
30620
32023
33485
14.064
14.071
14.073
14.088
14.089
2000
2000
2002
2003
2006
Make
Model (Transmission Type)
Engine Size
FORD
FORD
CHEVROLET
CHEVROLET
ACURA
BMW
FORD
CHEVROLET
BMW
SUBARU
VOLKSWAGEN
AUDI
DODGE
GMC
GMC
HYUNDAI
TOYOTA
NISSAN
E150 ECONOLINE
EXCURSION
CORVETTE (Manual)
IMPALA
RSX
X5
ESCAPE
C2500 2WD
545I
IMPREZA WRX
NEW BEETLE
A4
DURANGO 4WD
SIERRA 2500 DIESEL
CANYON 2WD
SANTA FE
CELICA
FRONTIER 2WD
GRAND CHEROKEE 4WD
SUNFIRE
SENTRA
QUEST
RANGER REG CAB LONG
330CI CONVERTIBLE
SUNFIRE
ELANTRA
BLAZER 2WD
F150 REG CAB LONG
3-SERIES
E430
004.6
006.8
005.7
003.4
002.0
004.4
003.0
006.0
004.4
002.0
002.0
001.8
004.7
006.6
003.5
002.7
001.8
002.4
8 Cylinder
Gasoline
004.7
002.2
001.8
003.3
002.5
003.0
002.2
002.0
004.3
004.2
002.5
004.3
FORD
RANGER SUPER CAB 2DR
002.5
ISUZU
CHEVROLET
CHRYSLER
VOLKSWAGEN
DODGE
TROOPER
SILVERADO 1500 2WD
SEBRING COUPE
JETTA
STRATUS
003.5
005.3
003.0
002.8
002.7
114
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
34976
31027
35011
29298
29182
33593
35006
29547
29470
32387
29392
Odds
Ratio
14.097
14.099
14.100
14.119
14.137
14.148
14.165
14.167
14.172
14.203
14.203
Model
Year
2002
2002
2003
2000
2000
2006
2003
2000
2000
2004
2000
Make
Model (Transmission Type)
Engine Size
MAZDA
NISSAN
FORD
DODGE
CHEVROLET
HYUNDAI
CHEVROLET
MERCEDES
JAGUAR
KIA
GMC
PROTEGE 5
SENTRA (Manual)
F250 SUPER DUTY
STRATUS
C1500 SUBURBAN 2WD
ACCENT
SILVERADO 2500 2WD
C230 KOMPRESSOR
S-TYPE
OPTIMA
SIERRA 1500 2WD
002.0
002.5
005.4
002.4
005.7
001.6
006.0
002.3
003.0
002.7
005.3
29725
14.212
2000
VOLKSWAGEN
JETTA
002.0
29585
33293
29256
29828
32188
32029
30860
29820
14.220
14.270
14.271
14.272
14.300
14.307
14.320
14.325
2000
2006
2000
2001
2004
2003
2002
2001
MITSUBISHI
AUDI
DODGE
BMW
CHRYSLER
VOLKSWAGEN
JAGUAR
BMW
GALANT
A3
B1500 VAN
X5 (Manual)
CROSSFIRE
NEW BEETLE (Manual)
X-TYPE
740IL
002.4
002.0
005.2
004.4
003.2
002.0
003.0
004.4
32375
14.330
2004
JEEP
GRAND CHEROKEE 2WD
004.0
32616
30997
32842
29586
34770
31264
29367
30517
29996
32648
31561
33559
31255
14.330
14.336
14.336
14.342
14.346
14.378
14.385
14.392
14.395
14.396
14.397
14.413
14.419
2004
2002
2005
2000
2003
2003
2000
2002
2001
2004
2003
2006
2003
VOLKSWAGEN
MITSUBISHI
DODGE
MITSUBISHI
BMW
BMW
FORD
BMW
FORD
VOLVO
GMC
GMC
BMW
JETTA
GALANT
MAGNUM
GALANT
525I
745I
RANGER REG CAB SHORT
M3 CONVERTIBLE
EXPLORER 4DR
V40
ENVOY 2WD
ENVOY 4WD
330I
2020
14.420
2003
29593
35299
14.432
14.437
2000
2004
001.8
002.4
003.5
003.0
002.5
004.4
002.5
003.2
004.0
001.9
004.2
004.2
003.0
6 Cylinder
Gasoline
003.0
004.4
Multiple Mid-Size Light Truck Makes and Models
MITSUBISHI
BMW
MONTERO SPORT 4WD
745I
115
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
30325
31563
33595
29338
30512
31372
29269
30251
29587
31874
34739
Odds
Ratio
14.447
14.453
14.460
14.461
14.477
14.478
14.480
14.483
14.497
14.506
14.506
Model
Year
2001
2003
2006
2000
2002
2003
2000
2001
2000
2003
2000
2198
14.511
2006
30910
32295
29326
33651
30500
30407
34791
31074
29444
32193
30727
30840
30496
31004
29727
30872
31284
35039
32437
29239
30355
31798
30495
31737
14.529
14.529
14.538
14.542
14.555
14.556
14.566
14.576
14.582
14.598
14.637
14.639
14.641
14.703
14.720
14.727
14.729
14.735
14.758
14.769
14.771
14.772
14.775
14.778
2002
2004
2000
2006
2002
2001
2004
2002
2000
2004
2002
2002
2002
2002
2000
2002
2003
2003
2004
2000
2001
2003
2002
2003
LINCOLN
GMC
FORD
JEEP
BMW
TOYOTA
CHEVROLET
SAAB
HYUNDAI
CHRYSLER
FORD
INFINITI
BMW
MITSUBISHI
VOLKSWAGEN
JEEP
BUICK
HONDA
MERCEDES
CHRYSLER
SUBARU
MERCEDES
BMW
LINCOLN
29352
14.786
2000
FORD
Make
Model (Transmission Type)
Engine Size
PORSCHE
GMC
HYUNDAI
FORD
BMW
CHEVROLET
DODGE
MERCURY
MITSUBISHI
OLDSMOBILE
CHEVROLET
BOXSTER
ENVOY 4WD
ELANTRA
F150 SUPER CAB LONG
745LI
TRAILBLAZER 4WD
CARAVAN 2WD
GRAND MARQUIS
MIRAGE
ALERO
S10 PICKUP 2WD
002.7
004.2
002.0
004.6
004.4
004.2
003.3
004.6
001.5
002.2
002.2
4 Cylinder
Gasoline
003.9
004.2
004.0
003.7
003.0
003.4
006.0
002.0
001.5
002.4
002.0
003.5
002.5
003.5
001.8
004.7
003.1
003.0
001.8
002.7
002.5
003.2
002.5
005.4
Multiple Mini-Van Makes and Models
LS
ENVOY 2WD
EXPLORER 2DR
LIBERTY 2WD
330CI
TACOMA 4WD
EXPRESS 3500
9-3 CONVERTIBLE
ACCENT (Auto)
PT CRUISER
FOCUS 4DR SEDAN
QX4 2WD
325I
MONTERO
NEW BEETLE
GRAND CHEROKEE 4WD
CENTURY
ACCORD COUPE
C230 KOMPRESSOR
CONCORDE
LEGACY WAGON AWD
ML320
325CI CONVERTIBLE
NAVIGATOR
MUSTANG CONVERTIBLE
(Auto)
003.8
116
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
Odds
Ratio
Model
Year
Make
Model (Transmission Type)
Engine Size
34767
14.799
2002
VOLKSWAGEN
NEW BEETLE
001.8
30592
14.802
2002
CHEVROLET
S10 PICKUP 2WD
004.3
31525
14.802
2003
FORD
MUSTANG CONVERTIBLE
004.6
31548
32802
32778
31626
29620
29324
32179
29223
30547
31345
29613
29583
29627
30376
31370
30652
31596
31024
14.810
14.815
14.816
14.841
14.841
14.846
14.853
14.883
14.889
14.909
14.910
14.919
14.926
14.931
14.932
14.943
14.957
14.965
2003
2005
2005
2003
2000
2000
2004
2000
2002
2003
2000
2000
2000
2001
2003
2002
2003
2002
FORD
CHRYSLER
CHEVROLET
HYUNDAI
PLYMOUTH
FORD
CHEVROLET
CHEVROLET
CHEVROLET
CHEVROLET
OLDSMOBILE
MITSUBISHI
PONTIAC
TOYOTA
CHEVROLET
DODGE
GMC
NISSAN
TAURUS
300
IMPALA
SANTA FE 2WD
NEON
EXPEDITION
TRAILBLAZER 2WD
S10 PICKUP 2WD FFV
AVALANCHE 2WD
SILVERADO 1500 4WD
ALERO
ECLIPSE
BONNEVILLE
4RUNNER 2WD
TRAILBLAZER 2WD
DURANGO 4WD
SAFARI 2WD PASSENGER
QUEST
003.0
003.5
003.8
002.4
002.0
004.6
004.2
002.2
005.3
005.3
003.4
002.4
003.8
003.4
004.2
005.9
004.3
003.3
1946
14.968
2002
31174
14.981
2002
50021
14.982
2003
30239
30982
32181
30661
30172
32169
29334
32481
52063
30569
30470
14.989
14.990
15.027
15.036
15.041
15.041
15.048
15.069
15.094
15.098
15.100
2001
2002
2004
2002
2001
2004
2000
2004
2003
2002
2002
Multiple Mid-Size Passenger Car Makes and
Models
VOLKSWAGEN
PASSAT
Multiple Light Truck Makes and Models
MERCEDES
MERCURY
CHEVROLET
DODGE
LEXUS
CHEVROLET
FORD
MITSUBISHI
CHEVROLET
CHEVROLET
AUDI
S430
COUGAR
TRAILBLAZER 4WD
RAM 2500
IS 300
MONTE CARLO
F150 REG CAB SHORT
GALANT
SILVERADO 3500 DIESEL
G1500/2500 EXPRESS
A4 QUATTRO (Manual)
4 Cylinder
Gasoline
001.8
8 Cylinder
Diesel
004.3
002.5
004.2
005.9
003.0
003.4
004.2
003.8
006.6
005.7
001.8
117
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
Odds
Ratio
Model
Year
2021
15.111
2003
31085
29211
29322
29190
31073
15.116
15.120
15.126
15.131
15.134
2002
2000
2000
2000
2002
SATURN
CHEVROLET
FORD
CHEVROLET
SAAB
LW300
K1500 TAHOE 4WD
ESCORT
CAMARO
9-3
003.0
005.3
002.0
005.7
002.0
30944
15.135
2002
MERCEDES
C230 KOMPRESSOR
002.3
31435
29408
30754
29289
31287
29418
32224
30842
29148
15.157
15.158
15.159
15.162
15.168
15.172
15.186
15.192
15.196
2003
2000
2002
2000
2003
2000
2004
2002
2000
DODGE
GMC
FORD
DODGE
BUICK
GMC
DODGE
ISUZU
BMW
RAM 1500 2WD
JIMMY 4WD
RANGER SUPER CAB 2DR
NEON
REGAL
K1500 YUKON 4WD
RAM 1500 2WD
AXIOM 2WD
740I/740I SPORT
004.7
004.3
002.3
002.0
003.8
005.7
003.7
003.5
004.4
33386
15.208
2006
CHEVROLET
COLORADO 2WD
003.5
31025
33193
31185
15.229
15.237
15.255
2002
2005
2002
NISSAN
SUBARU
VOLVO
SENTRA
IMPREZA STI
S40
001.8
002.5
001.9
34738
15.276
2000
CHEVROLET
C2500 SILVERADO 2WD
005.7
33130
31248
31586
30494
29630
30911
34839
15.289
15.294
15.305
15.332
15.335
15.339
15.348
2005
2003
2003
2002
2000
2002
2000
NISSAN
BMW
GMC
BMW
PONTIAC
LINCOLN
CHRYSLER
004.0
002.5
005.3
002.5
002.4
005.4
003.3
29353
15.352
2000
FORD
32759
32742
33547
34752
29217
29485
30611
30554
15.355
15.359
15.378
15.383
15.397
15.397
15.399
15.407
2005
2005
2006
2001
2000
2000
2002
2002
CHEVROLET
CHEVROLET
GMC
FORD
CHEVROLET
JEEP
CHRYSLER
CHEVROLET
FRONTIER 2WD
325I
SIERRA 1500 4WD
325CI
GRAND AM
NAVIGATOR
VOYAGER
MUSTANG CONVERTABLE
(manual)
COLORADO 2WD
ASTRO 2WD
SIERRA 1500 2WD
RANGER SUPER CAB 2DR
MALIBU
CHEROKEE 2WD
300M
C1500 AVALANCHE 2WD
Make
Model (Transmission Type)
Multiple Light Truck Makes and Models
Engine Size
8 Cylinder
003.8
002.8
004.3
004.3
003.0
003.1
004.0
003.5
005.3
118
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
30826
30155
32558
33446
30736
30240
32815
30292
31116
30015
30871
29207
29794
32480
32334
33139
29497
Odds
Ratio
15.422
15.422
15.425
15.427
15.429
15.432
15.433
15.443
15.449
15.458
15.459
15.459
15.463
15.473
15.489
15.518
15.518
Model
Year
2002
2001
2004
2006
2002
2001
2005
2001
2002
2001
2002
2000
2001
2004
2004
2005
2000
2000
15.534
2003
29215
31173
33385
30150
30658
31387
15.535
15.548
15.582
15.590
15.592
15.609
2000
2002
2006
2001
2002
2003
CHEVROLET
VOLKSWAGEN
CHEVROLET
JEEP
DODGE
CHRYSLER
LUMINA/MONTECARLO
NEW BEETLE (Manual)
COLORADO 2WD
GRAND CHEROKEE 2WD
RAM 1500 2WD
PT CRUISER (Manual)
003.8
002.0
002.8
004.7
005.9
002.4
32969
15.620
2005
HYUNDAI
SONATA
002.7
33016
30614
29222
29608
32926
30991
33138
29166
34994
15.637
15.647
15.649
15.651
15.676
15.676
15.685
15.696
15.698
2005
2002
2000
2000
2005
2002
2005
2000
2003
KIA
CHRYSLER
CHEVROLET
NISSAN
GMC
MERCURY
NISSAN
CADILLAC
BMW
SEDONA
CONCORDE
S10 PICKUP 2WD
SENTRA
ENVOY 2WD
VILLAGER
QUEST
DEVILLE
325I
003.5
003.5
004.3
002.0
004.2
003.3
003.5
004.6
002.5
1838
15.710
2000
Make
Model (Transmission Type)
Engine Size
HYUNDAI
JEEP
SUBARU
CHRYSLER
FORD
MERCEDES
CHRYSLER
NISSAN
SUZUKI
FORD
JEEP
CHEVROLET
AUDI
MITSUBISHI
HYUNDAI
NISSAN
KIA
ACCENT
WRANGLER 4WD
IMPREZA STI
PACIFICA AWD
MUSTANG CONVERTIBLE
S500
PACIFICA AWD
SENTRA
GRAND VITARA XL7 4WD
F150 SUPER CAB SHORT
GRAND CHEROKEE 4WD
SILVERADO 1500 4WD
TT COUPE QUATTRO
GALANT
SANTA FE
SENTRA
SPORTAGE
001.6
004.4
002.5
003.5
004.6
005.0
003.5
002.0
002.7
005.4
004.0
005.3
001.8
002.4
002.4
001.8
002.0
Multiple Compact Passenger Car Makes and
Models
Multiple Mid-Size Passenger Car Makes and
Models
4 Cylinder
Gasoline
4 Cylinder
Gasoline
119
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
Odds
Ratio
Model
Year
1903
15.731
2001
Multiple Station Wagon Makes and Models
29424
30278
15.737
15.743
2000
2001
GMC
NISSAN
2128
15.747
2005
30759
29441
29691
31631
30854
30885
31128
31163
29696
30544
30880
35023
15.752
15.758
15.761
15.762
15.782
15.786
15.790
15.798
15.800
15.808
15.809
15.810
2002
2000
2000
2003
2002
2002
2002
2002
2000
2002
2002
2003
FORD
HONDA
TOYOTA
HYUNDAI
JAGUAR
KIA
TOYOTA
VOLKSWAGEN
TOYOTA
CHEVROLET
JEEP
FORD
TAURUS
PASSPORT 4WD
CAMRY SOLARA
SONATA
S-TYPE 3.0 LITRE
RIO
CAMRY SOLARA
GTI
COROLLA
ASTRO 2WD PASSENGER
WRANGLER
E350 ECONOLINE
003.0
003.2
002.2
002.7
003.0
001.5
002.4
001.8
001.8
004.3
004.0
005.4
29214
15.820
2000
CHEVROLET
LUMINA/MONTECARLO
003.4
30045
31825
31166
30447
29629
31834
33377
29764
29734
31731
29624
32147
29206
15.826
15.841
15.844
15.850
15.875
15.875
15.881
15.883
15.889
15.911
15.913
15.918
15.921
2001
2003
2002
2001
2000
2003
2006
2001
2000
2003
2000
2004
2000
FORD
MERCURY
VOLKSWAGEN
VOLVO
PONTIAC
MITSUBISHI
CHEVROLET
ACURA
VOLKSWAGEN
LINCOLN
PLYMOUTH
CHEVROLET
CHEVROLET
RANGER SUPER CAB 2DR
SABLE
JETTA
V70
FIREBIRD/TRANS AM
GALANT
SILVERADO 1500 2WD
INTEGRA
PASSAT (Manual)
AVIATOR 2WD
VOYAGER 2WD
COLORADO 4WD
SILVERADO 1500 4WD
2048
15.931
2003
32522
15.938
2004
PONTIAC
GRAND PRIX
004.0
003.0
002.0
002.3
005.7
002.4
004.3
001.8
002.8
004.6
003.3
003.5
004.8
8 Cylinder
Gasoline
003.8
32482
15.959
2004
MITSUBISHI
LANCER
002.0
30534
32028
15.966
15.980
2002
2003
BUICK
VOLKSWAGEN
RENDEZVOUS AWD
NEW BEETLE (Auto)
003.4
002.0
Make
Model (Transmission Type)
SONOMA 2WD
ALTIMA
Multiple Compact Light Truck Makes and Models
Multiple Full-Size Van Makes and Models
Engine Size
6 Cylinder
Gasoline
004.3
002.4
6 Cylinder
Gasoline
120
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
30262
31167
31934
30377
31480
32548
30106
30651
34778
32613
Odds
Ratio
16.009
16.015
16.040
16.041
16.045
16.057
16.082
16.087
16.095
16.100
Model
Year
2001
2002
2003
2001
2003
2004
2001
2002
2003
2004
1901
16.127
2001
30624
32192
29659
32617
30503
30697
32135
30507
29737
32287
33877
33144
30998
31714
29175
33143
29174
16.133
16.138
16.154
16.160
16.181
16.188
16.198
16.215
16.244
16.252
16.258
16.260
16.268
16.289
16.298
16.299
16.340
2002
2004
2000
2004
2002
2002
2004
2002
2000
2004
2006
2005
2002
2003
2000
2005
2000
CHRYSLER
CHRYSLER
SATURN
VOLKSWAGEN
BMW
FORD
CHEVROLET
BMW
VOLKSWAGEN
GMC
SUZUKI
NISSAN
MITSUBISHI
LAND ROVER
CHEVROLET
NISSAN
CHEVROLET
SEBRING CONVERTIBLE
PACIFICA AWD
SW
JETTA
330I
EXPEDITION
AVEO
530I
PASSAT WAGON
SIERRA 1500 2WD
FORENZA
XTERRA 4WD
GALANT
RANGE ROVER
BLAZER 4WD
XTERRA 2WD
BLAZER 2WD
002.7
003.5
001.9
002.0
003.0
004.6
001.6
003.0
002.8
004.3
002.0
004.0
003.0
004.4
004.3
004.0
004.3
32272
16.348
2004
FORD
FREESTAR WAGON FWD
003.9
30424
30669
29391
34737
31289
31132
29880
31154
16.354
16.363
16.376
16.376
16.386
16.389
16.389
16.414
2001
2002
2000
2000
2003
2002
2001
2002
VOLKSWAGEN
DODGE
GMC
CHEVROLET
BUICK
TOYOTA
CHEVROLET
TOYOTA
JETTA (Manual)
STRATUS
SIERRA 1500 2WD
C1500 TAHOE 2WD
RENDEZVOUS FWD
CELICA
K1500 SUBURBAN 4WD
TUNDRA 2WD
002.8
003.0
004.8
005.3
003.4
001.8
005.3
003.4
Make
Model (Transmission Type)
Engine Size
MITSUBISHI
VOLKSWAGEN
SATURN
TOYOTA
FORD
SATURN
HONDA
DODGE
FORD
VOLKSWAGEN
ECLIPSE SPYDER
JETTA
VUE AWD
4RUNNER 4WD
EXPEDITION
VUE FWD
PRELUDE
DURANGO 4WD
E250 ECONOLINE
GOLF
003.0
002.8
003.0
003.4
004.6
002.2
002.2
004.7
005.4
002.0
Multiple Station Wagon Makes and Models
4 Cylinder
Gasoline
121
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
30586
32021
32030
Odds
Ratio
16.432
16.444
16.459
Model
Year
2002
2003
2003
Make
Model (Transmission Type)
Engine Size
CHEVROLET
VOLKSWAGEN
VOLKSWAGEN
MALIBU
JETTA
PASSAT
003.1
001.8
001.8
29584
16.460
2000
MITSUBISHI
ECLIPSE
003.0
31136
30212
33151
31350
30224
34765
31095
32801
29201
31290
34869
29440
31013
30660
31171
16.474
16.500
16.502
16.515
16.539
16.548
16.548
16.550
16.559
16.591
16.614
16.618
16.625
16.634
16.642
2002
2001
2005
2003
2001
2002
2002
2005
2000
2003
2001
2000
2002
2002
2002
TOYOTA
MAZDA
PONTIAC
CHEVROLET
MERCEDES
MITSUBISHI
SUBARU
CHRYSLER
CHEVROLET
CADILLAC
CHEVROLET
HONDA
NISSAN
DODGE
VOLKSWAGEN
HIGHLANDER 2WD
MX-5 MIATA
GRAND PRIX
MALIBU
CLK320
ECLIPSE
IMPREZA AWD
300
IMPALA
CTS
SILVERADO 2500 2WD
PASSPORT 2WD
ALTIMA
RAM 1500 4WD
NEW BEETLE
1994
16.643
2002
29212
30828
31311
30086
29297
29170
34863
16.649
16.652
16.654
16.699
16.703
16.725
16.727
2000
2002
2003
2001
2000
2000
2000
CHEVROLET
HYUNDAI
CHEVROLET
GMC
DODGE
CHEVROLET
MERCURY
K1500 TAHOE 4WD
SANTA FE 2WD
SILVERADO 1500 2WD
K1500 YUKON XL 4WD
STRATUS
ASTRO 2WD CARGO
VILLAGER
002.4
001.8
003.8
003.1
003.2
003.0
002.0
002.7
003.8
003.2
006.0
003.2
002.5
005.9
001.8
8 Cylinder
Gasoline
005.7
002.4
005.3
005.3
002.0
004.3
003.3
31177
16.732
2002
VOLKSWAGEN
PASSAT 4MOTION
002.8
29417
32018
31087
30516
31896
29224
29526
32191
34751
16.744
16.744
16.757
16.763
16.791
16.794
16.803
16.818
16.822
2000
2003
2002
2002
2003
2000
2000
2004
2001
GMC
VOLKSWAGEN
SATURN
BMW
PONTIAC
CHEVROLET
MAZDA
CHRYSLER
FORD
K1500 YUKON 4WD
GOLF
SC
M3
VIBE
S10 PICKUP 4WD
626
PACIFICA 2WD
RANGER REG CAB SHORT
005.3
002.0
001.9
003.2
001.8
004.3
002.5
003.5
003.0
Multiple Full-Size Van Makes and Models
122
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
29356
33250
32392
31083
29350
30026
29995
31104
29268
30087
31192
31539
30657
Odds
Ratio
16.862
16.863
16.877
16.884
16.893
16.915
16.920
16.932
16.951
16.968
16.975
16.986
16.989
Model
Year
2000
2005
2004
2002
2000
2001
2001
2002
2000
2001
2002
2003
2002
Make
Model (Transmission Type)
Engine Size
FORD
VOLKSWAGEN
KIA
SATURN
FORD
FORD
FORD
SUBARU
DODGE
GMC
VOLVO
FORD
DODGE
MUSTANG COUPE (Manual)
JETTA
SPECTRA
L300
MUSTANG
MUSTANG CONV. (Auto)
EXPLORER 2DR
LEGACY WAGON AWD
CARAVAN 2WD
K1500 YUKON XL 4WD
V40
RANGER REG CAB SHORT
RAM 1500 2WD
003.8
002.0
001.8
003.0
003.8
004.6
004.0
002.5
003.0
006.0
001.9
002.3
004.7
31300
16.994
2003
CHEVROLET
ASTRO 2WD PASSENGER
004.3
33019
29407
29752
29455
34784
32485
29383
30237
34928
30737
29378
30677
32003
31165
17.027
17.031
17.043
17.045
17.046
17.048
17.060
17.070
17.087
17.097
17.101
17.111
17.123
17.128
2005
2000
2000
2000
2003
2004
2000
2001
2002
2002
2000
2002
2003
2002
KIA
GMC
VOLVO
INFINITI
VOLKSWAGEN
MITSUBISHI
FORD
MERCEDES
CHEVROLET
FORD
FORD
FORD
TOYOTA
VOLKSWAGEN
SPECTRA
JIMMY 2WD
S80
Q45
NEW BEETLE
MONTERO
WINDSTAR 4DR WAGON
ML430
EXPRESS 3500
MUSTANG COUPE (Auto)
TAURUS
CROWN VICTORIA
SIENNA
JETTA
1921
17.151
2001
29654
29173
17.164
17.184
2000
2000
SATURN
CHEVROLET
LS
ASTRO AWD PASSENGER
002.0
004.3
002.8
004.1
001.8
003.8
003.8
004.3
005.7
003.8
003.0
004.6
003.0
001.8
6 Cylinder
Gasoline
003.0
004.3
33135
17.191
2005
NISSAN
PATHFINDER 2WD
004.0
30321
34783
33557
34837
31423
17.233
17.238
17.247
17.247
17.304
2001
2003
2006
2000
2003
PONTIAC
VOLKSWAGEN
GMC
CHEVROLET
DODGE
SUNFIRE
JETTA
ENVOY 2WD
EXPRESS 3500
DURANGO 2WD
002.2
001.8
004.2
005.7
004.7
Multiple Mid-Size SUV Makes and Models
123
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
29220
Odds
Ratio
17.312
Model
Year
2000
1912
17.326
2001
32813
30749
31213
34916
29286
31669
17.327
17.342
17.367
17.392
17.414
17.423
2005
2002
2003
2002
2000
2003
1964
17.433
2002
Multiple Compact Light Truck Makes and Models
33398
29200
30003
31610
32393
17.473
17.475
17.491
17.491
17.497
2006
2000
2001
2003
2004
CHEVROLET
CHEVROLET
FORD
HONDA
KIA
33069
17.524
2005
30245
30511
31691
29196
31402
30397
31309
33421
31868
29582
29209
17.549
17.553
17.560
17.565
17.565
17.566
17.576
17.581
17.582
17.603
17.607
32187
32145
32333
30442
31418
30021
30543
31050
30215
30760
17.614
17.620
17.641
17.655
17.669
17.670
17.677
17.690
17.710
17.717
Make
Model (Transmission Type)
Engine Size
CHEVROLET
PRIZM
EXPRESS 1500
IMPALA
F150 REG CAB LONG
CIVIC
SPECTRA
001.8
6 Cylinder
Gasoline
003.5
2.3
003.0
002.5
002.7
002.5
4 Cylinder
Gasoline
004.3
003.4
004.6
002.0
002.0
MERCEDES
C230 KOMPRESSOR
001.8
2001
2002
2003
2000
2003
2001
2003
2006
2003
2000
2000
MERCEDES
BMW
JEEP
CHEVROLET
CHRYSLER
TOYOTA
CHEVROLET
CHEVROLET
NISSAN
MITSUBISHI
CHEVROLET
SLK230 KOMPRESSOR
745I
GRAND CHEROKEE 4WD
G1500/2500 EXPRESS
VOYAGER/TOWN/CTRY
RAV4 2WD
SILVERADO 1500 2WD
MONTE CARLO
SENTRA
DIAMANTE SEDAN
K1500 SUBURBAN 4WD
002.3
004.4
004.7
005.7
003.8
002.0
004.3
003.9
001.8
003.5
005.3
2004
2004
2004
2001
2003
2001
2002
2002
2001
2002
CHRYSLER
CHEVROLET
HYUNDAI
VOLVO
DODGE
FORD
CHEVROLET
PONTIAC
MAZDA
FORD
CONCORDE
COLORADO 2WD
ELANTRA
S60
DAKOTA 2WD
FOCUS ZX3 3DR
ASTRO 2WD CARGO
FIREBIRD
PROTEGE/PROTEGE MPS
TAURUS WAGON
003.5
003.5
002.0
002.4
004.7
002.0
004.3
005.7
002.0
003.0
Multiple Mid-Size Light Truck Makes and Models
CHRYSLER
FORD
AUDI
BMW
DODGE
JAGUAR
PACIFICA 2WD
RANGER REG CAB SHORT
A4
325I
INTREPID
X-TYPE
124
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
34775
29598
30886
Odds
Ratio
17.724
17.731
17.736
Model
Year
2003
2000
2002
Make
Model (Transmission Type)
Engine Size
CHEVROLET
NISSAN
KIA
G3500 EXPRESS
ALTIMA
SEDONA
006.0
002.4
003.5
29180
17.753
2000
CHEVROLET
SILVERADO 1500 2WD
004.8
52045
33429
35053
30016
30398
33659
30236
30654
30019
17.762
17.771
17.774
17.784
17.789
17.794
17.798
17.805
17.809
2002
2006
2003
2001
2001
2006
2001
2002
2001
VOLKSWAGEN
CHEVROLET
VOLKSWAGEN
FORD
TOYOTA
KIA
MERCEDES
DODGE
FORD
JETTA TDI
TRAILBLAZER 4WD
PASSAT
F150 SUPER CREWCAB
RAV4 4WD
RIO
ML320
INTREPID
FOCUS WAGON
001.9
004.2
002.8
004.6
002.0
001.6
003.2
003.5
002.0
33426
17.815
2006
CHEVROLET
TRAILBLAZER 2WD
004.2
30448
32795
29827
29425
33136
29631
30738
32964
29495
31527
29213
34810
31048
29855
29238
17.815
17.831
17.831
17.834
17.842
17.869
17.892
17.906
17.918
17.940
17.964
17.983
17.985
18.000
18.002
2001
2005
2001
2000
2005
2000
2002
2005
2000
2003
2000
2005
2002
2001
2000
VOLVO
CHEVROLET
BMW
GMC
NISSAN
PONTIAC
FORD
HYUNDAI
KIA
FORD
CHEVROLET
CHEVROLET
PONTIAC
CHEVROLET
CHRYSLER
V70
TRAILBLAZER 2WD
X5 (Auto)
SONOMA 2WD FFV
PATHFINDER 4WD
GRAND AM
MUSTANG COUPE (Manual)
ELANTRA
SEPHIA/SPECTRA
MUSTANG COUPE (Manual)
LUMINA/MONTECARLO
EXPRESS 3500
BONNEVILLE
C1500 SUBURBAN 2WD
CIRRUS
002.4
004.2
004.4
002.2
004.0
003.4
003.8
002.0
001.8
003.8
003.1
006.0
003.8
005.3
002.5
50026
18.020
2004
Multiple Domestic Light Truck Makes and Models
29633
30342
29714
29746
29117
29337
29390
18.042
18.050
18.060
18.069
18.090
18.092
18.093
2000
2001
2000
2000
2000
2000
2000
PONTIAC
SATURN
VOLKSWAGEN
VOLVO
AUDI
FORD
GMC
GRAND PRIX
L300
CABRIO
S40
A4 QUATTRO
F150 SUPER CAB LONG
SIERRA 1500 2WD
8 Cylinder
Diesel
003.8
003.0
002.0
001.9
001.8
004.2
004.3
125
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
31259
32144
29379
30027
31425
29466
31147
34768
29588
30734
Odds
Ratio
18.098
18.112
18.114
18.139
18.167
18.169
18.171
18.181
18.188
18.206
Model
Year
2003
2004
2000
2001
2003
2000
2002
2002
2000
2002
Make
Model (Transmission Type)
Engine Size
BMW
CHEVROLET
FORD
FORD
DODGE
ISUZU
TOYOTA
VOLKSWAGEN
MITSUBISHI
FORD
530I
COLORADO 2WD
TAURUS WAGON
MUSTANG CONV. (Manual)
DURANGO 4WD
RODEO 4WD
SIENNA
PASSAT
MIRAGE
MUSTANG CONVERTIBLE
003.0
002.8
003.0
004.6
004.7
003.2
003.0
002.8
001.8
003.8
32797
18.223
2005
CHEVROLET
TRAILBLAZER 4WD
004.2
30247
30650
29168
30085
32221
29454
30644
30535
30010
30588
30649
29885
30628
18.232
18.235
18.237
18.252
18.281
18.283
18.303
18.322
18.351
18.366
18.369
18.391
18.394
2001
2002
2000
2001
2004
2000
2002
2002
2001
2002
2002
2001
2002
MERCEDES
DODGE
CADILLAC
GMC
DODGE
INFINITI
DODGE
BUICK
FORD
CHEVROLET
DODGE
CHEVROLET
CHRYSLER
SLK320
DURANGO 2WD
ESCALADE 4WD
K1500 YUKON 4WD
INTREPID
I30
DAKOTA 2WD
RENDEZVOUS FWD
F150 SUPER CAB LONG
MONTE CARLO
DURANGO 2WD
METRO
VOYAGER
1885
18.398
2000
31299
29419
30136
29710
30768
30137
31158
31461
31997
32022
30122
30594
18.398
18.407
18.431
18.437
18.446
18.461
18.488
18.526
18.535
18.542
18.595
18.598
2003
2000
2001
2000
2002
2001
2002
2003
2003
2003
2001
2002
003.2
005.9
005.7
006.0
003.5
003.0
004.7
003.4
004.6
003.8
004.7
001.3
003.8
6 Cylinder
Gasoline
004.3
005.3
003.0
003.4
004.3
004.0
002.0
002.4
001.8
002.0
003.5
004.3
Multiple Full-Size Van Makes and Models
CHEVROLET
GMC
JAGUAR
TOYOTA
GMC
JAGUAR
VOLKSWAGEN
DODGE
TOYOTA
VOLKSWAGEN
INFINITI
CHEVROLET
ASTRO 2WD CARGO
K1500 YUKON 4WD
S-TYPE
TUNDRA 2WD
SIERRA 1500 2WD
S-TYPE 4.0 LITRE
CABRIO
STRATUS
MATRIX
JETTA
QX4 4WD
S10 PICKUP 4WD
126
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
31412
31988
30549
30814
31377
29909
30555
30731
Odds
Ratio
18.598
18.598
18.607
18.609
18.612
18.635
18.637
18.673
Model
Year
2003
2003
2002
2002
2003
2001
2002
2002
2047
18.690
2003
30419
30218
29592
33658
30018
29731
30574
30030
31624
18.706
18.712
18.713
18.717
18.727
18.746
18.793
18.810
18.812
2001
2001
2000
2006
2001
2000
2002
2001
2003
VOLKSWAGEN
MERCEDES
MITSUBISHI
KIA
FORD
VOLKSWAGEN
CHEVROLET
FORD
HYUNDAI
GTI
C240
MONTERO SPORT 2WD
OPTIMA
FOCUS 4DR SEDAN
PASSAT
IMPALA
MUSTANG COUPE
ACCENT
001.8
002.6
003.5
002.7
002.0
001.8
003.8
004.6
001.6
30820
18.852
2002
HONDA
PASSPORT 2WD
003.2
31032
31088
32396
30396
31625
31169
29616
29381
31052
29985
32208
32271
29816
31115
18.856
18.868
18.874
18.896
18.925
18.953
18.998
19.027
19.041
19.078
19.096
19.100
19.105
19.107
2002
2002
2004
2001
2003
2002
2000
2000
2002
2001
2004
2004
2001
2002
OLDSMOBILE
SATURN
LAND ROVER
TOYOTA
HYUNDAI
VOLKSWAGEN
OLDSMOBILE
FORD
PONTIAC
FORD
DODGE
FORD
BMW
SUZUKI
ALERO
SL
DISCOVERY
MR2
ELANTRA
JETTA WAGON
SILHOUETTE FWD
WINDSTAR 3DR WAGON
GRAND AM
E250 ECONOLINE
CARAVAN 2WD
FREESTAR CARGO FWD
530I
GRAND VITARA XL7
003.4
001.9
004.6
001.8
002.0
002.0
003.4
003.8
003.4
005.4
003.3
003.9
003.0
002.7
1993
19.113
2002
30409
30014
19.118
19.163
2001
2001
Make
Model (Transmission Type)
Engine Size
DODGE
TOYOTA
CHEVROLET
HONDA
CHEVROLET
CHRYSLER
CHEVROLET
FORD
CARAVAN 2WD
CELICA
BLAZER 2WD
CIVIC
VENTURE FWD
PT CRUISER (Auto)
BLAZER 4WD
FOCUS ZX3
003.8
001.8
004.3
002.0
003.4
002.4
004.3
002.0
Multiple Mid-Size Van Makes and Models
Multiple Mid-Size Van Makes and Models
TOYOTA
FORD
TUNDRA 2WD
F150 SUPER CAB SHORT
6 Cylinder
Gasoline
6 Cylinder
Gasoline
004.7
004.6
127
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
29194
29774
Odds
Ratio
19.206
19.224
Model
Year
2000
2001
50022
19.234
2003
31553
34736
29591
29465
31450
19.253
19.273
19.305
19.310
19.326
2003
2000
2000
2000
2003
1941
19.327
2001
34776
19.343
2003
1922
19.349
2001
30562
29747
30536
31077
31943
30217
31054
30418
29860
31551
33376
30423
31090
30918
31351
29974
30220
29123
30128
30587
30550
30593
34932
31092
19.369
19.379
19.384
19.396
19.409
19.435
19.437
19.443
19.462
19.493
19.642
19.683
19.698
19.716
19.745
19.777
19.788
19.846
19.886
19.981
20.079
20.109
20.118
20.119
2002
2000
2002
2002
2003
2001
2002
2001
2001
2003
2006
2001
2002
2002
2003
2001
2001
2000
2001
2002
2002
2002
2002
2002
50034
20.129
2006
Make
Model (Transmission Type)
Engine Size
CHEVROLET
AUDI
G1500/2500 EXPRESS
A4 QUATTRO
004.3
001.8
Multiple Heavy Duty Light Truck Makes and
Models
GMC
CHEVROLET
MITSUBISHI
ISUZU
DODGE
SIERRA 1500 2WD
C1500 SUBURBAN 2WD
MONTERO SPORT 2WD
RODEO 2WD
STRATUS
Multiple Ford Vans and Light Truck Models
CHEVROLET
S10 PICKUP 2WD
Multiple Full-Size SUV Makes and Models
CHEVROLET
CAMARO
VOLVO
S40
CADILLAC
DEVILLE
SAAB
9-5
SUBARU
IMPREZA AWD
MAZDA
TRIBUTE
PONTIAC
GRAND PRIX
VOLKSWAGEN
GOLF (Manual)
CHEVROLET
CAVALIER
FORD
WINDSTAR CARGO VAN
CHEVROLET
AVEO
VOLKSWAGEN
JETTA (Auto)
SATURN
VUE AWD
MAZDA
626
CHEVROLET
MONTE CARLO
FORD
CROWN VICTORIA
MERCEDES
C320
AUDI
A6 QUATTRO
ISUZU
RODEO 4WD
CHEVROLET
MONTE CARLO
CHEVROLET
BLAZER 4WD
CHEVROLET
S10 PICKUP 2WD FFV
CHRYSLER
TOWN & COUNTRY 2WD
SATURN
VUE FWD
Multiple Heavy Duty Light Truck Makes and
Models
8 Cylinder
Diesel
004.3
005.3
003.0
003.2
002.4
006.8
002.2
8 Cylinder
Gasoline
005.7
002.0
004.6
002.3
002.0
003.0
003.8
002.0
002.2
003.8
001.6
002.8
003.0
002.0
003.4
004.6
003.2
002.7
003.2
003.4
004.3
002.2
003.8
002.2
8 Cylinder
Diesel
128
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
Odds
Ratio
Model
Year
Make
Model (Transmission Type)
Engine Size
30992
20.145
2002
MITSUBISHI
DIAMANTE SEDAN
003.5
32746
30182
20.246
20.273
2005
2001
CHEVROLET
LINCOLN
AVEO
TOWN CAR
001.6
004.6
29735
20.286
2000
VOLKSWAGEN
PASSAT 4MOTION
002.8
31889
31991
29184
20.289
20.329
20.330
2003
2003
2000
PONTIAC
TOYOTA
CHEVROLET
GRAND AM
HIGHLANDER 2WD
C1500 TAHOE 2WD
003.4
002.4
004.8
34834
20.338
2000
CHEVROLET
SILVERADO 3500 2WD
005.7
34746
29835
29836
30637
31565
30288
29179
34941
30417
34759
31529
20.355
20.406
20.406
20.458
20.458
20.513
20.572
20.606
20.630
20.651
20.659
2000
2001
2001
2002
2003
2001
2000
2002
2001
2002
2003
GMC
BUICK
BUICK
DODGE
GMC
NISSAN
CHEVROLET
FORD
VOLKSWAGEN
CHEVROLET
FORD
C1500 YUKON XL 2WD
LE SABRE
PARK AVENUE
CARAVAN 2WD
ENVOY XL 2WD
PATHFINDER 4WD
SILVERADO 1500 2WD
F250 SUPER DUTY
GOLF (Auto)
S10 PICKUP 2WD
RANGER 2WD
1940
20.676
2001
31767
31056
29814
31306
30008
32760
30089
20.687
20.752
20.767
20.771
20.791
20.806
20.830
2003
2002
2001
2003
2001
2005
2001
1914
20.859
2001
29244
29628
29780
31936
29603
29802
30121
31291
20.867
20.896
20.904
20.937
20.973
20.974
21.000
21.027
2000
2000
2001
2003
2000
2001
2001
2003
005.3
003.8
003.8
003.8
004.2
003.5
004.3
005.4
002.0
002.2
002.3
8 Cylinder
Gasoline
001.8
003.4
002.5
004.3
005.4
003.5
004.3
8 Cylinder
Gasoline
002.5
003.8
002.7
002.2
003.0
002.5
003.5
004.6
Multiple Full-Size Van Makes and Models
MERCEDES
PONTIAC
BMW
CHEVROLET
FORD
CHEVROLET
GMC
C230 KOMPRESSOR
MONTANA FWD
525I
BLAZER 2WD
F150 REG CAB SHORT
COLORADO 2WD
SAFARI 2WD PASSENGER
Multiple Light Truck Makes and Models
CHRYSLER
PONTIAC
AUDI
SATURN
NISSAN
BMW
INFINITI
CADILLAC
SEBRING CONVERTIBLE
FIREBIRD/TRANS AM
A6 QUATTRO
VUE FWD (Manual)
MAXIMA
325CI
QX4 2WD
DEVILLE
129
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
30589
30825
29804
29306
30386
Odds
Ratio
21.028
21.053
21.059
21.110
21.149
Model
Year
2002
2002
2001
2000
2001
Make
Model (Transmission Type)
Engine Size
CHEVROLET
HYUNDAI
BMW
FORD
TOYOTA
PRIZM
ACCENT
325I
CONTOUR
CELICA
1977
21.171
2002
FORD
Light Trucks and SUVs
32292
30256
30870
30296
32220
31935
30427
21.210
21.254
21.317
21.365
21.370
21.475
21.480
2004
2001
2002
2001
2004
2003
2001
1895
21.487
2001
31927
30117
29994
31356
32388
34766
30475
21.495
21.522
21.566
21.587
21.635
21.664
21.678
2003
2001
2001
2003
2004
2002
2002
GMC
CANYON 2WD
MERCURY
VILLAGER
JEEP
GRAND CHEROKEE 2WD
OLDSMOBILE
ALERO
DODGE
INTREPID
SATURN
VUE FWD (Auto)
VOLKSWAGEN
NEW BEETLE
Multiple Full-Size Passenger Car Makes and
Models
SATURN
L300
HYUNDAI
XG 300
FORD
EXPEDITION
CHEVROLET
S10 PICKUP 2WD
KIA
RIO
VOLKSWAGEN
JETTA
AUDI
A6 QUATTRO
1913
21.678
2001
Multiple Full-Size Light Truck Makes and Models
30622
21.736
2002
CHRYSLER
001.8
001.5
002.5
002.5
001.8
8 Cylinder
Gasoline
003.5
003.3
004.7
002.4
002.7
002.2
001.8
8 Cylinder
Gasoline
003.0
003.0
005.4
004.3
001.6
001.8
003.0
8 Cylinder
Gasoline
002.7
2049
21.738
2003
FORD
31875
29910
29844
31958
30388
30315
29524
30573
34753
29803
29733
21.767
21.807
21.839
21.854
21.861
21.891
21.909
21.910
21.916
21.947
21.954
2003
2001
2001
2003
2001
2001
2000
2002
2001
2001
2000
OLDSMOBILE
CHRYSLER
CHEVROLET
SUZUKI
TOYOTA
PONTIAC
MAZDA
CHEVROLET
MITSUBISHI
BMW
VOLKSWAGEN
SEBRING SEDAN
Multiple Vans and Light
Trucks
ALERO
PT CRUISER (Manual)
ASTRO 2WD PASSENGER
AERIO
COROLLA
GRAND AM
626
IMPALA
ECLIPSE
325CI CONVERTIBLE
PASSAT (Auto)
006.8
003.4
002.4
004.3
002.0
001.8
002.4
002.0
003.4
003.0
002.5
002.8
130
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
30909
30084
34908
Odds
Ratio
21.957
21.967
21.969
Model
Year
2002
2001
2001
Make
Model (Transmission Type)
Engine Size
LINCOLN
GMC
JEEP
LS
K1500 YUKON 4WD
CHEROKEE 2WD
003.0
005.3
004.0
29951
30004
30627
34889
30149
29990
21.972
22.011
22.014
22.044
22.070
22.095
2001
2001
2002
2001
2001
2001
DODGE
FORD
CHRYSLER
FORD
JEEP
FORD
RAM 1500 4WD
F150 REG CAB LONG
VOYAGER
F250 SUPER DUTY
GRAND CHEROKEE 2WD
ESCAPE
005.9
005.4
003.3
006.8
004.0
003.0
1892
22.100
2001
34865
34760
30471
29573
31051
34894
33604
30869
22.143
22.164
22.195
22.207
22.246
22.261
22.310
22.314
2001
2002
2002
2000
2002
2001
2006
2002
1984
22.323
2002
30290
51859
30430
31837
35087
33444
30390
22.329
22.336
22.337
22.350
22.366
22.373
22.393
2001
2002
2001
2003
2004
2006
2001
NISSAN
CHEVROLET
VOLKSWAGEN
MITSUBISHI
FORD
CHRYSLER
TOYOTA
QUEST
SILVERADO 2500 DIESEL
PASSAT
LANCER
E350
PACIFICA 2WD
HIGHLANDER 2WD
003.0
005.2
003.0
0025
002.2
006.8
002.7
004.0
6 Cylinder
Gasoline
003.3
006.6
001.8
002.0
005.4
003.5
002.4
31337
22.431
2003
CHEVROLET
IMPALA
003.8
29546
29634
30254
30304
31106
22.522
22.522
22.534
22.538
22.546
2000
2000
2001
2001
2002
MAZDA
PONTIAC
MERCURY
PLYMOUTH
SUZUKI
PROTEGE
MONTANA FWD
SABLE
NEON
AERIO
001.8
003.4
003.0
002.0
002.0
31397
22.586
2003
CHRYSLER
TOWN & COUNTRY
003.8
29810
31219
29932
22.697
22.706
22.725
2001
2003
2001
BMW
AUDI
DODGE
330CI
A4 QUATTRO
DAKOTA 2WD
003.0
003.0
004.7
Multiple Compact Passenger Car Makes and
Models
BMW
DODGE
AUDI
MERCURY
PONTIAC
FORD
HYUNDAI
JEEP
330CI CONVERTIBLE
RAM VAN 1500 2WD
A4 QUATTRO
COUGAR
GRAND AM
EXCURSION
TIBURON
GRAND CHEROKEE 2WD
Multiple Mini-Van Makes and Models
4 Cylinder
Gasoline
131
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
29657
29966
34888
30039
29952
30180
30259
29931
52255
29299
30894
30429
29235
30542
29940
32258
32010
31430
30250
29545
30422
30151
Odds
Ratio
22.737
22.814
22.817
22.825
22.825
22.873
22.881
22.976
22.982
23.003
23.042
23.077
23.135
23.184
23.203
23.213
23.252
23.263
23.321
23.336
23.358
23.385
Model
Year
2000
2001
2001
2001
2001
2001
2001
2001
2003
2000
2002
2001
2000
2002
2001
2004
2003
2003
2001
2000
2001
2001
1894
23.387
2001
31352
23.395
2003
2073
23.453
2004
29854
32041
30729
30181
30832
33015
32818
29189
29992
30655
29857
35080
23.507
23.529
23.550
23.557
23.579
23.581
23.585
23.607
23.646
23.647
23.651
23.659
2001
2003
2002
2001
2002
2005
2005
2000
2001
2002
2001
2004
Make
Model (Transmission Type)
Engine Size
SATURN
DODGE
FORD
FORD
DODGE
LINCOLN
MITSUBISHI
DODGE
VOLKSWAGEN
DODGE
LAND ROVER
VOLKSWAGEN
CHEVROLET
CADILLAC
DODGE
FORD
TOYOTA
DODGE
MERCURY
MAZDA
VOLKSWAGEN
JEEP
SC
STRATUS
F250 SUPER DUTY
RANGER REG CAB SHORT
RAM 2500 (CA Emission)
LS
ECLIPSE
DAKOTA 2WD
JETTA TDI
STRATUS
RANGE ROVER
NEW BEETLE (Manual)
VENTURE FWD
SEVILLE
DURANGO 4WD
F150 2WD
TUNDRA 2WD
NEON
COUGAR
PROTEGE
JETTA
GRAND CHEROKEE 4WD
001.9
003.0
005.4
002.5
005.9
003.9
002.4
003.9
001.9
002.5
004.6
002.0
003.4
004.6
005.9
004.2
003.4
002.0
002.5
001.6
002.0
004.0
Multiple Mid-Size Passenger Car Makes and
Models
CHEVROLET
MONTE CARLO
General Motors Light Trucks
CHEVROLET
VOLVO
FORD
LINCOLN
HYUNDAI
KIA
CHRYSLER
CHEVROLET
FORD
DODGE
CHEVROLET
FORD
SILVERADO 1500 2WD
S40
FOCUS WAGON
NAVIGATOR
SONATA
RIO
SEBRING
CAMARO
ESCORT 4DR
NEON
C1500 TAHOE 2WD
F350 SUPER DUTY
6 Cylinder
Gasoline
003.8
003.5
005.3
001.9
002.0
005.4
002.7
001.6
002.7
003.8
002.0
002.0
005.3
005.4
132
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
31600
29450
30721
31053
31937
Odds
Ratio
23.664
23.805
23.805
23.819
23.832
Model
Year
2003
2000
2002
2002
2003
Make
Model (Transmission Type)
Engine Size
GMC
HYUNDAI
FORD
PONTIAC
SATURN
004.3
002.5
004.2
003.1
003.0
30715
23.846
2002
FORD
30431
30298
30063
29843
29607
29839
30712
29165
29658
29250
23.851
23.877
23.934
23.940
23.965
23.980
24.061
24.086
24.094
24.108
2001
2001
2001
2001
2000
2001
2002
2000
2000
2000
VOLKSWAGEN
OLDSMOBILE
GMC
CHEVROLET
NISSAN
CADILLAC
FORD
CADILLAC
SATURN
DAEWOO
SONOMA 2WD
SONATA
F150 SUPER CAB (Auto)
GRAND PRIX
VUE FWD
F150 REG CAB SHORT
(Auto)
PASSAT
ALERO
C1500 YUKON 2WD
ASTRO 2WD CARGO
SENTRA
DEVILLE
F150 REG CAB (Manual)
CATERA
SL
LEGANZA
30428
24.157
2001
VOLKSWAGEN
NEW BEETLE (Auto)
002.0
30642
29935
29842
30049
31428
33202
32340
30158
30889
29834
30709
29503
34761
32222
30119
30281
30762
30671
29859
30893
24.197
24.221
24.228
24.229
24.264
24.359
24.430
24.475
24.507
24.508
24.542
24.549
24.642
24.654
24.710
24.718
24.783
24.801
24.842
24.895
2002
2001
2001
2001
2003
2005
2004
2001
2002
2001
2002
2000
2002
2004
2001
2001
2002
2002
2001
2002
DODGE
DODGE
CADILLAC
FORD
DODGE
SUZUKI
HYUNDAI
KIA
KIA
BUICK
FORD
LAND ROVER
DODGE
DODGE
INFINITI
NISSAN
FORD
DODGE
CHEVROLET
LAND ROVER
DAKOTA 2WD
DAKOTA 4WD
SEVILLE
TAURUS
INTREPID
FORENZA
TIBURON
RIO
SPORTAGE 4WD
CENTURY
F150 REG CAB LONG
RANGE ROVER
RAM VAN 2500 2WD
NEON
I30
FRONTIER 2WD (Manual)
WINDSTAR 4DR
STRATUS 4-DR
CAMARO
FREELANDER
003.9
004.7
004.6
003.0
002.7
002.0
002.7
001.5
002.0
003.1
004.2
004.6
005.2
002.0
003.0
002.4
003.8
002.7
005.7
002.5
004.2
002.8
003.4
005.3
004.3
001.8
004.6
004.2
003.0
001.9
002.2
133
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
30379
30264
30472
30148
30561
32232
31501
30653
30271
Odds
Ratio
24.913
24.917
24.941
24.948
25.047
25.056
25.060
25.126
25.137
Model
Year
2001
2001
2002
2001
2002
2004
2003
2002
2001
Make
Model (Transmission Type)
Engine Size
TOYOTA
MITSUBISHI
AUDI
JEEP
CHEVROLET
DODGE
FORD
DODGE
MITSUBISHI
CAMRY
GALANT
A6
CHEROKEE 4WD
CAMARO
STRATUS
F150 REG CAB LONG
INTREPID
MONTERO SPORT 2WD
002.2
003.0
003.0
004.0
003.8
002.4
004.2
002.7
003.0
30287
31416
32194
29882
31411
30024
30412
29946
25.188
25.194
25.198
25.228
25.233
25.292
25.332
25.396
2001
2003
2004
2001
2003
2001
2001
2001
NISSAN
DODGE
CHRYSLER
CHEVROLET
DODGE
FORD
VOLKSWAGEN
DODGE
PATHFINDER 2WD
DAKOTA 2WD
SEBRING
K1500 TAHOE 4WD
CARAVAN 2WD
MUSTANG CONVERTIBLE
CABRIO
NEON
003.5
003.9
002.4
005.3
003.3
003.8
002.0
002.0
30266
25.400
2001
MITSUBISHI
MIRAGE
001.8
29499
30610
31049
32923
30345
30626
25.423
25.455
25.616
25.664
25.668
25.798
2000
2002
2002
2005
2001
2002
LAND ROVER
CHEVROLET
PONTIAC
GMC
SATURN
CHRYSLER
DISCOVERY SER II
VENTURE FWD
FIREBIRD
CANYON 2WD
SC
VOYAGER
004.0
003.4
003.8
003.5
001.9
002.4
29878
25.824
2001
CHEVROLET
SILVERADO 1500 4WD
005.3
30113
31336
29243
30092
31690
30059
29939
29252
25.866
25.898
25.908
25.958
25.972
25.999
26.003
26.040
2001
2003
2000
2001
2003
2001
2001
2000
HYUNDAI
CHEVROLET
CHRYSLER
GMC
JEEP
GMC
DODGE
DAEWOO
SONATA
IMPALA
SEBRING
SONOMA 2WD
GRAND CHEROKEE 4WD
SIERRA 1500 2WD
DURANGO 4WD
NUBIRA
2038
26.056
2003
30159
31401
30446
26.072
26.095
26.140
2001
2003
2001
002.4
003.4
002.5
004.3
004.0
004.8
004.7
002.0
6 Cylinder
Gasoline
001.8
003.3
001.9
Multiple Mini-Van Makes and Models
KIA
CHRYSLER
VOLVO
SEPHIA/SPECTRA
VOYAGER/TOWN&CTRY
V40
134
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
29965
29948
30080
51985
29993
29904
Odds
Ratio
26.270
26.302
26.312
26.361
26.396
26.427
Model
Year
2001
2001
2001
2002
2001
2001
Make
Model (Transmission Type)
Engine Size
DODGE
DODGE
GMC
GMC
FORD
CHRYSLER
002.4
005.2
005.3
006.6
004.6
003.5
30722
26.433
2002
FORD
30060
29837
29848
31493
30763
30421
26.528
26.565
26.653
26.658
26.688
26.735
2001
2001
2001
2003
2002
2001
GMC
BUICK
CHEVROLET
FORD
FORD
VOLKSWAGEN
STRATUS
RAM 1500 2WD
SIERRA 1500 4WD
SIERRA 2500 DIESEL
EXPEDITION
300M
F150 SUPER CAB SHORT
(Manual)
SIERRA 1500 2WD
REGAL
BLAZER 4WD
F150 2WD
WINDSTAR CARGO VAN
JETTA
1910
26.758
2001
30213
31713
30635
30104
30668
30408
32568
34748
30346
26.952
26.991
27.003
27.017
27.018
27.021
27.025
27.063
27.096
2001
2003
2002
2001
2002
2001
2004
2001
2001
MAZDA
LAND ROVER
DODGE
HONDA
DODGE
TOYOTA
SUZUKI
AUDI
SATURN
PROTEGE/PROTEGE MPS
FREELANDER
CARAVAN 2WD
PASSPORT 2WD
STRATUS
TUNDRA 2WD
FORENZA
A4
SL
001.6
002.5
002.4
003.2
002.4
003.4
002.0
001.8
001.9
29890
27.114
2001
CHEVROLET
S10 PICKUP 2WD
004.3
32817
31702
29991
30440
29870
27.128
27.183
27.208
27.243
27.283
2005
2003
2001
2001
2001
CHRYSLER
KIA
FORD
VOLVO
CHEVROLET
SEBRING
RIO
ESCORT
S40
IMPALA
002.4
001.6
002.0
001.9
003.4
30280
27.330
2001
NISSAN
FRONTIER 2WD (Auto)
002.4
29888
30887
30272
29968
29912
31452
27.451
27.495
27.592
27.608
27.612
27.612
2001
2002
2001
2001
2001
2003
CHEVROLET
KIA
MITSUBISHI
DODGE
CHRYSLER
DODGE
PRIZM
SPECTRA
MONTERO SPORT 2WD
STRATUS 4-DR
SEBRING COUPE
STRATUS 4-DR
001.8
001.8
003.5
002.7
003.0
002.4
Multiple Light Truck Makes and Models
004.2
005.3
003.8
004.3
004.2
003.8
001.8
4 Cylinder
Gasoline
135
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
Odds
Ratio
Model
Year
1939
27.616
2001
30636
29853
31000
34931
31711
31890
30263
29812
30052
30051
27.652
27.755
27.762
27.836
27.850
27.879
27.883
28.060
28.109
28.132
2002
2001
2002
2002
2003
2003
2001
2001
2001
2001
DODGE
CHEVROLET
MITSUBISHI
CHRYSLER
LAND ROVER
PONTIAC
MITSUBISHI
BMW
FORD
FORD
CARAVAN 2WD
SILVERADO 1500 2WD
LANCER
TOWN & COUNTRY 2WD
DISCOVERY
GRAND PRIX
GALANT
330I
WINDSTAR 4DR WAGON
WINDSTAR 3DR WAGON
8 Cylinder
Gasoline
003.3
004.8
002.0
003.3
004.6
003.1
002.4
003.0
003.8
003.8
34878
28.216
2001
CHRYSLER
TOWN & COUNTRY 2WD
003.8
30764
30695
30127
32855
30939
29884
31511
29632
30028
29937
30260
29926
31031
29847
30022
28.220
28.274
28.310
28.506
28.554
28.670
28.674
28.711
28.735
28.804
28.832
28.847
28.909
28.950
29.059
2002
2002
2001
2005
2002
2001
2003
2000
2001
2001
2001
2001
2002
2001
2001
FORD
FORD
ISUZU
DODGE
MAZDA
CHEVROLET
FORD
PONTIAC
FORD
DODGE
MITSUBISHI
DODGE
OLDSMOBILE
CHEVROLET
FORD
WINDSTAR WAGON
ESCORT
RODEO 2WD
STRATUS
MPV
MALIBU
F150 SUPER CAB SHORT
GRAND PRIX
MUSTANG COUPE (Auto)
DURANGO 2WD
ECLIPSE
CARAVAN 2WD
ALERO
BLAZER 2WD
MUSTANG
003.8
002.0
003.2
002.4
003.0
003.1
004.2
003.1
003.8
004.7
003.0
002.4
002.2
004.3
003.8
29915
29.074
2001
CHRYSLER
SEBRING CONVERTIBLE
002.7
30274
30053
30161
31688
30382
30892
30178
29949
31706
29.233
29.428
29.431
29.554
29.624
29.696
29.716
29.790
29.811
2001
2001
2001
2003
2001
2002
2001
2001
2003
MITSUBISHI
FORD
KIA
JEEP
TOYOTA
LAND ROVER
LINCOLN
DODGE
KIA
MONTERO SPORT 4WD
WINDSTAR VAN
SPORTAGE 4WD
GRAND CHEROKEE 2WD
CAMRY SOLARA
DISCOVERY SER II
LS
RAM 1500 2WD
SPECTRA
003.5
003.8
002.0
004.0
002.2
004.0
003.0
005.9
001.8
Make
Model (Transmission Type)
Multiple Full-Size Van Makes and Models
Engine Size
136
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
30888
34749
29928
29947
30291
30265
29543
29871
31468
31552
30317
29249
29883
30038
Odds
Ratio
29.893
29.908
29.962
30.007
30.031
30.152
30.218
30.480
30.840
30.893
31.841
32.081
32.346
32.567
Model
Year
2002
2001
2001
2001
2001
2001
2000
2001
2003
2003
2001
2000
2001
2001
1930
32.584
2001
30401
30114
29852
30284
32.743
32.784
32.831
32.870
2001
2001
2001
2001
TOYOTA
HYUNDAI
CHEVROLET
NISSAN
29891
32.894
2001
29886
32571
29776
30311
30184
33.123
33.582
33.595
33.600
33.648
30268
30012
29887
31762
30160
30005
30884
30684
30156
29927
29903
30258
Make
Model (Transmission Type)
Engine Size
KIA
CHEVROLET
DODGE
DODGE
NISSAN
MITSUBISHI
MAZDA
CHEVROLET
FORD
FORD
PONTIAC
DAEWOO
CHEVROLET
FORD
SPORTAGE 2WD
S10 PICKUP 2WD
CARAVAN 2WD
RAM 1500 2WD
SENTRA
MIRAGE
MPV
IMPALA
E150 ECONOLINE
WINDSTAR WAGON
GRAND AM
LANOS
LUMINA
RANGER REG CAB SHORT
SIENNA
SONATA
SILVERADO 1500 2WD
MAXIMA
002.0
002.2
003.8
003.9
001.8
001.5
002.5
003.8
004.2
003.8
003.4
001.6
003.1
002.3
6 Cylinder
Gasoline
003.0
002.5
004.3
003.0
CHEVROLET
S10 PICKUP 2WD FFV
002.2
2001
2004
2001
2001
2001
CHEVROLET
SUZUKI
AUDI
PONTIAC
MAZDA
MONTE CARLO
VERONA
A6
AZTEK FWD
626
003.4
002.5
002.8
003.4
002.0
33.916
2001
MITSUBISHI
MONTERO
003.5
33.972
34.010
34.117
34.122
34.368
34.464
34.520
34.606
34.666
35.058
35.183
2001
2001
2003
2001
2001
2002
2002
2001
2001
2001
2001
FORD
CHEVROLET
MAZDA
KIA
FORD
KIA
FORD
KIA
DODGE
CHEVROLET
MITSUBISHI
F150 SUPER CAB (Auto)
MONTE CARLO
MPV
SPORTAGE 2WD
F150 REG CAB SHORT
OPTIMA
E150 ECONOLINE
OPTIMA
CARAVAN 2WD
VENTURE FWD
DIAMANTE SEDAN
004.2
003.8
003.0
002.0
004.2
002.7
004.2
002.4
003.3
003.4
003.5
Multiple Mini-Van Makes and Models
137
Appendix C: VLT ID Numbers Most Likely to Fail the OBD II Functional Test
VLT
ID
30302
29922
30111
31633
52135
34877
30002
29858
29914
Odds
Ratio
35.378
35.610
35.804
35.863
36.005
36.346
36.417
36.480
36.501
Model
Year
2001
2001
2001
2003
2003
2001
2001
2001
2001
Make
Model (Transmission Type)
Engine Size
OLDSMOBILE
DAEWOO
HYUNDAI
HYUNDAI
FORD
CHRYSLER
FORD
CHEVROLET
CHRYSLER
003.5
002.2
002.4
002.7
006.0
003.3
004.2
003.8
002.7
30013
36.526
2001
FORD
30938
30029
29918
29941
30319
30072
29921
30318
30163
30312
29979
30299
30320
30211
36.608
36.613
36.666
37.184
37.807
37.998
38.451
38.909
39.043
39.576
40.481
40.813
41.487
43.542
2002
2001
2001
2001
2001
2001
2001
2001
2001
2001
2001
2001
2001
2001
MAZDA
FORD
CHRYSLER
DODGE
PONTIAC
GMC
DAEWOO
PONTIAC
LAND ROVER
PONTIAC
FORD
OLDSMOBILE
PONTIAC
MAZDA
INTRIGUE
LEGANZA
SANTA FE
TIBURON
EXCURSION DIESEL
TOWN & COUNTRY 2WD
F150 REG CAB LONG
CAMARO
SEBRING SEDAN
F150 SUPER CAB SHORT
(Manual)
MILLENIA
MUSTANG COUPE (Manual)
VOYAGER
INTREPID
GRAND PRIX
JIMMY 2WD
LANOS
GRAND PRIX
DISCOVERY SER II
BONNEVILLE
E150 ECONOLINE
AURORA
MONTANA FWD
MPV
004.2
002.5
003.8
003.3
002.7
003.8
004.3
001.6
003.1
004.0
003.8
004.2
003.5
003.4
002.5
138
Appendix D: VLT ID Numbers Demonstrating Zero Failures of the Visual Inspection
VLT ID
2073
2236
2245
Model-Year
2004
2007
2007
Make
Model
General Motors
Light Trucks
Multiple Mid-Size Light Truck Makes and Models
Multiple Mid-Size SUV Makes and Models
Engine Size
003.5
6 Cylinder Gasoline
6 Cylinder Gasoline
2264
2007
Multiple Full-Size Van Makes and Models
8 Cylinder Gasoline
2269
2008
SMART
2274
2281
2318
2325
2335
2354
2372
2398
2426
2433
2450
2487
2489
2513
2516
2524
2008
2008
2008
2009
2009
2009
2009
2010
2010
2011
2011
2012
2012
2012
2012
2012
Multiple Luxury Car Makes and Models
Multiple Mid-Size Passenger Car Makes and Models
Multiple Full-Size Van Makes and Models
Multiple Mid-Size Passenger Car Makes and Models
Multiple Mid-Size SUV Makes and Models
Multiple Full-Size SUV Makes and Models
Multiple Light Truck Makes and Models
Multiple Compact Light Truck Makes and Models
Multiple Full-Size Van Makes and Models
Multiple Mid-Size Passenger Car Makes and Models
Multiple Compact Light Truck Makes and Models
Multiple Mid-Size Passenger Car Makes and Models
Multiple Full-Size Passenger Car Makes and Models
Multiple Mid-Size SUV Makes and Models
Multiple Full-Size SUV Makes and Models
Multiple Mini-Van Makes and Models
10 Cylinder Gas
6 Cylinder Gasoline
8 Cylinder Gasoline
5 Cylinder Gasoline
6 Cylinder Gasoline
8 Cylinder Gasoline
8 Cylinder Gasoline
6 Cylinder Gasoline
8 Cylinder Gasoline
5 Cylinder Gasoline
4 Cylinder Gasoline
5 Cylinder Gasoline
8 Cylinder Gasoline
4 Cylinder Gasoline
8 Cylinder Gasoline
6 Cylinder Gasoline
2534
2012
Multiple Full-Size Van Makes and Models
8 Cylinder Gasoline
29568
29575
29639
29641
29776
29817
30283
30452
30537
30794
30889
30903
30974
31048
31067
31485
2000
2000
2000
2000
2001
2001
2001
2002
2002
2002
2002
2002
2002
2002
2002
2003
MERCEDES
MERCURY
PORSCHE
PORSCHE
AUDI
BMW
NISSAN
ACURA
CADILLAC
GMC
KIA
LEXUS
MERCEDES
PONTIAC
PORSCHE
FORD
SMART CAR
SL500
MOUNTAINEER
BOXSTER
BOXSTER S
A6
540I
FRONTIER 4WD
3.2CL
ELDORADO
SIERRA 1500 DENALI AWD
SPORTAGE 4WD
LX 470
SL500
BONNEVILLE
BOXSTER
EXPLORER 2WD
001.0
005.0
004.0
002.7
003.2
002.8
004.4
003.3
003.2
004.6
006.0
002.0
004.7
005.0
003.8
002.7
004.6
139
Appendix D: VLT ID Numbers Demonstrating Zero Failures of the Visual Inspection
VLT ID
31565
31593
Model-Year
2003
2003
Make
GMC
GMC
Model
ENVOY XL 2WD
K1500 YUKON DENALI XL
Engine Size
004.2
006.0
31596
2003
GMC
SAFARI 2WD PASSENGER
004.3
31640
31722
31798
31800
32066
32073
32094
32109
32244
32350
32366
32403
32406
32413
32456
32458
2003
2003
2003
2003
2004
2004
2004
2004
2004
2004
2004
2004
2004
2004
2004
2004
INFINITI
LEXUS
MERCEDES
MERCEDES
AUDI
AUDI
BMW
BUICK
FORD
INFINITI
JAGUAR
LEXUS
LEXUS
LINCOLN
MERCEDES
MERCEDES
I35
GX 470
ML320
ML500
A4
A6 QUATTRO
5-SERIES
PARK AVENUE
E250
QX56 2WD
XJ8
GS 430
LX 470
LS
ML500
S500
003.5
004.7
003.2
005.0
003.0
002.7
003.0
003.8
004.6
005.6
004.2
004.3
004.7
003.9
005.0
005.0
32486
2004
MITSUBISHI
MONTERO SPORT
003.5
32499
32532
32542
32575
32582
32624
32640
32705
32707
32717
32739
32740
32752
32812
32835
32836
32884
32895
2004
2004
2004
2004
2004
2004
2004
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
NISSAN
PORSCHE
SATURN
TOYOTA
TOYOTA
VOLKSWAGEN
VOLVO
BMW
BMW
BUICK
CADILLAC
CADILLAC
CHEVROLET
CHRYSLER
DODGE
DODGE
FORD
FORD
PATHFINDER 4WD
BOXSTER
L300
4RUNNER 2WD
CAMRY SOLARA
PASSAT
S40
M3
X3
LACROSSE
STS
STS
C1500 SUBURBAN
CROSSFIRE
DURANGO 2WD
DURANGO 2WD
EXPLORER SPORT TRAC
FIVE HUNDRED
003.5
003.2
002.2
004.7
002.4
002.8
002.5
003.2
003.0
003.8
003.6
004.6
005.3
003.2
004.7
005.7
004.0
003.0
140
Appendix D: VLT ID Numbers Demonstrating Zero Failures of the Visual Inspection
VLT ID
32976
33003
33012
33026
33039
33067
33079
33080
33084
33089
33095
33133
33158
33166
33170
33188
33209
33224
33232
33233
33266
33321
33324
33339
33344
33360
33367
33369
Model-Year
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2005
2006
2006
2006
2006
2006
2006
2006
Make
INFINITI
JEEP
KIA
LAND ROVER
LEXUS
MAZDA
MERCEDES
MERCEDES
MERCEDES
MERCEDES
MERCEDES
NISSAN
PONTIAC
PORSCHE
SAAB
SUBARU
TOYOTA
TOYOTA
TOYOTA
TOYOTA
VOLVO
BMW
BMW
BMW
BUICK
CADILLAC
CADILLAC
CADILLAC
Model
FX35 RWD
GRAND CHEROKEE 4WD
AMANTI
LR3
SC 430
TRIBUTE
CLK320
CLK500
E500
ML350
SL500
MURANO AWD
VIBE
CAYENNE
9-3
FORESTER AWD
4RUNNER 2WD
HIGHLANDER 2WD
RAV4 2WD
RAV4 4WD
S40
3-SERIES
5-SERIES
Z4 COUPE
LUCERNE
DTS
STS
STS
Engine Size
003.5
003.7
003.5
004.4
004.3
003.0
003.2
005.0
005.0
003.7
005.0
003.5
001.8
004.5
002.0
002.5
004.0
002.4
002.4
002.4
002.4
002.5
003.0
003.0
004.6
004.6
003.6
004.6
33375
2006
CHEVROLET
AVALANCHE 1500 4WD
005.3
33455
33469
33494
33508
2006
2006
2006
2006
CHRYSLER
DODGE
FORD
FORD
TOWN & COUNTRY 2WD
DURANGO 2WD
E150
EXPLORER 4WD
003.8
005.7
005.4
004.0
33529
2006
FORD
FREESTAR WAGON FWD
004.2
33541
33601
33605
33608
2006
2006
2006
2006
FORD
HYUNDAI
HYUNDAI
HYUNDAI
RANGER 4WD
SONATA
TUCSON
TUCSON 4WD
004.0
002.4
002.0
002.7
141
Appendix D: VLT ID Numbers Demonstrating Zero Failures of the Visual Inspection
VLT ID
33609
33610
33624
33657
33665
33670
33680
33691
33692
33696
33723
33724
33737
33743
33749
33752
33763
33768
33772
33777
33788
33796
33837
33839
33840
33873
33876
Model-Year
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
2006
Make
INFINITI
INFINITI
JAGUAR
KIA
KIA
LAND ROVER
LAND ROVER
LEXUS
LEXUS
LEXUS
MAZDA
MAZDA
MERCEDES
MERCEDES
MERCEDES
MERCEDES
MERCEDES
MERCURY
MERCURY
MERCURY
MITSUBISHI
MITSUBISHI
PORSCHE
PORSCHE
PORSCHE
SUBARU
SUZUKI
Model
FX35
FX35 AWD
S-TYPE
OPTIMA
SPORTAGE
LR3
SPORT
LS 430
LX 470
SC 430
TRIBUTE
TRIBUTE
CLK500
E500
ML500
S350
SLK350
MARINER
MILAN
MOUNTAINEER 2WD
ENDEAVOR
OUTLANDER
BOXSTER
CAYENNE
CAYENNE
OUTBACK
AERIO
Engine Size
003.5
003.5
003.0
002.4
002.7
004.0
004.2
004.3
004.7
004.3
002.3
003.0
005.0
005.0
005.0
003.7
003.5
003.0
003.0
004.0
003.8
002.4
002.7
003.2
004.5
003.0
002.3
33878
2006
SUZUKI
GRAND VITARA
002.7
33904
33905
33955
33956
33965
34005
34025
34027
34044
34047
2006
2006
2007
2007
2007
2007
2007
2007
2007
2007
TOYOTA
TOYOTA
ACURA
ACURA
AUDI
BMW
CADILLAC
CADILLAC
CHEVROLET
CHEVROLET
RAV4 2WD
RAV4 4WD
MDX
RDX
A4
X3
DTS
ESCALADE AWD
SILVERADO 1500 2WD
C1500 SUBURBAN
003.5
002.4
003.7
002.3
002.0
003.0
004.6
006.2
004.3
005.3
142
Appendix D: VLT ID Numbers Demonstrating Zero Failures of the Visual Inspection
VLT ID
34053
34070
Model-Year
2007
2007
Make
CHEVROLET
CHEVROLET
Model
COLORADO 2WD
IMPALA
Engine Size
003.7
003.5
34081
2007
CHEVROLET
K1500 SUBURBAN
005.3
34099
34132
34154
34168
34173
34174
34183
34188
34199
34211
34264
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
CHEVROLET
DODGE
DODGE
FORD
FORD
FORD
FORD
FORD
FORD
FORD
HONDA
UPLANDER
CARAVAN 2WD
NITRO
EDGE
EXPEDITION 2WD
EXPLORER 2WD
F150 2WD
FIVE HUNDRED
FUSION
TAURUS
ELEMENT
003.9
003.3
003.7
003.5
005.4
004.0
004.2
003.0
002.3
003.0
002.4
34351
2007
LAND ROVER
RANGE ROVER
004.4
34359
34374
34375
34399
34403
34413
34414
34420
34452
34456
34457
34461
34462
34467
34469
34470
34509
34531
34547
34549
34560
34571
34573
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
LEXUS
LINCOLN
LINCOLN
MERCEDES
MERCEDES
MERCEDES
MERCEDES
MERCEDES
MITSUBISHI
NISSAN
NISSAN
NISSAN
NISSAN
NISSAN
NISSAN
NISSAN
SATURN
SUBARU
TOYOTA
TOYOTA
TOYOTA
TOYOTA
TOYOTA
GS 350
NAVIGATOR
TOWN CAR
C230
CLK350
GL450
ML350
S550
OUTLANDER
ALTIMA
ARMADA
MAXIMA
MURANO
TITAN 2WD
VERSA
XTERRA 2WD
AURA
OUTBACK
4RUNNER 2WD
4RUNNER 4WD
FJ CRUISER 4WD
RAV4 2WD
RAV4 4WD
003.5
005.4
004.6
002.5
003.5
004.6
003.5
005.5
003.0
003.5
005.6
003.5
003.5
005.6
001.8
004.0
003.5
002.5
004.0
004.0
004.0
003.5
003.5
143
Appendix D: VLT ID Numbers Demonstrating Zero Failures of the Visual Inspection
VLT ID
34574
34592
34817
34820
34825
35068
35168
35181
35229
35243
35257
35276
35277
50039
50041
50042
50045
52961
53016
53038
53069
Model-Year
2007
2007
2006
2007
2007
2004
2005
2006
2007
2007
2007
2007
2007
2009
2010
2010
2012
2008
2009
2010
2005
Make
Model
TOYOTA
SEQUOIA 2WD
VOLKSWAGEN
PASSAT
CHEVROLET
EXPRESS 2500
BMW
525I
CHEVROLET
EXPRESS 2500
BMW
325I SPORT WAGON
MINI
MINI COOPER
BMW
330CI CONVERTIBLE
BMW
328I COUPE
CHEVROLET
SILVERADO 2500 2WD
FORD
E350
MINI
MINI COOPER
MINI
MINI COOPER S
Multiple Mid-Size Passenger Car Makes and Models
Multiple Mid-Size Passenger Car Makes and Models
Multiple Light Truck Makes and Models
Multiple Mid-Size Passenger Car Makes and Models
DODGE
SPRINTER 2500 DIESEL
DODGE
RAM 2500 DIESEL
FORD
F350 DIESEL
FORD
F350 SRW SUPER DUTY
Engine Size
004.7
002.0
006.0
002.5
004.8
002.5
001.6
003.0
003.0
006.0
005.4
001.6
001.6
Diesel Engine
Diesel Engine
Diesel Engine
Diesel Engine
003.0
006.7
006.4
006.0
144
Appendix E: VLT ID Numbers Most Likely to Fail the Visual Inspection
VLT
ID
Odds
Ratio
Model
Year
50022
1.605
2003
31853
52293
29289
29349
29729
29102
29217
1.627
1.647
1.647
1.652
1.659
1.660
1.708
2003
2004
2000
2000
2000
2000
2000
NISSAN
DODGE
DODGE
FORD
VOLKSWAGEN
ACURA
CHEVROLET
350Z
RAM 2500 DIESEL
NEON
FOCUS ZX3 3DR
NEW BEETLE (Auto)
INTEGRA
MALIBU
Engine
Size
6 and 8
Cylinder
Diesel
003.5
005.9
002.0
002.0
002.0
001.8
003.1
29346
1.722
2000
FORD
FOCUS 4-DR SEDAN
002.0
29631
1.728
2000
PONTIAC
GRAND AM
003.4
30642
1.742
2002
DODGE
DAKOTA 2WD
003.9
34749
30109
30587
33281
29580
31095
29731
30617
30730
1.742
1.754
1.764
1.771
1.771
1.772
1.774
1.776
1.776
2001
2001
2002
2005
2000
2002
2000
2002
2002
CHEVROLET
HYUNDAI
CHEVROLET
VOLVO
MERCURY
SUBARU
VOLKSWAGEN
CHRYSLER
FORD
S10 PICKUP 2WD
ACCENT
MONTE CARLO
XC90
SABLE
IMPREZA AWD
PASSAT
PT CRUISER (Auto)
FOCUS ZX3
002.2
001.6
003.4
002.5
003.0
002.0
001.8
002.4
002.0
29730
1.784
2000
VOLKSWAGEN
NEW BEETLE (Manual)
002.0
29277
31428
32505
30654
30107
1.789
1.790
1.794
1.802
1.813
2000
2003
2004
2002
2001
DODGE
DODGE
NISSAN
DODGE
HONDA
DAKOTA 4WD
INTREPID
SENTRA
INTREPID
S2000
004.7
002.7
002.5
003.5
002.0
29200
1.815
2000
CHEVROLET
IMPALA
003.4
29764
30106
52265
30422
1.815
1.821
1.821
1.825
2001
2001
2004
2001
ACURA
HONDA
CHEVROLET
VOLKSWAGEN
INTEGRA
PRELUDE
SILVERADO 2500 DIESEL
JETTA
001.8
002.2
006.6
002.0
30735
1.826
2002
FORD
MUSTANG CONVERTIBLE (Auto)
004.6
29990
52341
29299
1.827
1.849
1.854
2001
2004
2000
FORD
FORD
DODGE
ESCAPE
EXCURSION DIESEL
STRATUS
003.0
006.0
002.5
Make
Model (Transmission Type)
Multiple Domestic Light Truck Makes and Models
145
Appendix E: VLT ID Numbers Most Likely to Fail the Visual Inspection
VLT
ID
30421
32521
Odds
Ratio
1.855
1.857
Model
Year
2001
2004
Make
Model (Transmission Type)
VOLKSWAGEN
PONTIAC
JETTA
GRAND AM
Engine
Size
001.8
003.4
32228
1.858
2004
DODGE
RAM 1500 4WD
004.7
29724
52171
31336
29322
31522
52493
29323
1.865
1.868
1.868
1.878
1.879
1.880
1.887
2000
2003
2003
2000
2003
2005
2000
VOLKSWAGEN
FORD
CHEVROLET
FORD
FORD
DODGE
FORD
JETTA
F250 DIESEL
IMPALA
ESCORT
MUSTANG
RAM 2500 DIESEL
ESCORT 4DR
001.8
006.0
003.4
002.0
004.6
005.9
002.0
31173
1.891
2002
VOLKSWAGEN
NEW BEETLE (Manual)
002.0
29728
33453
29727
30469
30268
29629
29493
1.896
1.896
1.899
1.900
1.919
1.925
1.932
2000
2006
2000
2002
2001
2000
2000
VOLKSWAGEN
CHRYSLER
VOLKSWAGEN
AUDI
MITSUBISHI
PONTIAC
JEEP
NEW BEETLE
SRT-8
NEW BEETLE
A4 QUATTRO (Auto)
MONTERO
FIREBIRD/TRANS AM
WRANGLER 4WD
001.8
006.1
001.8
001.8
003.5
005.7
002.5
29613
1.954
2000
OLDSMOBILE
ALERO
003.4
31434
30217
30019
34766
30994
31889
30418
29632
29584
52197
29915
29736
29223
31205
30250
29922
29870
29269
1.958
1.959
1.969
1.973
1.977
1.988
1.996
1.997
2.001
2.005
2.010
2.014
2.021
2.022
2.027
2.029
2.035
2.035
2003
2001
2001
2002
2002
2003
2001
2000
2000
2003
2001
2000
2000
2003
2001
2001
2001
2000
DODGE
MAZDA
FORD
VOLKSWAGEN
MITSUBISHI
PONTIAC
VOLKSWAGEN
PONTIAC
MITSUBISHI
FORD
CHRYSLER
VOLKSWAGEN
CHEVROLET
ACURA
MERCURY
DAEWOO
CHEVROLET
DODGE
RAM 1500 2WD
TRIBUTE
FOCUS WAGON
JETTA
ECLIPSE
GRAND AM
GOLF (Manual)
GRAND PRIX
ECLIPSE
F350 DIESEL
SEBRING CONVERTIBLE
PASSAT WAGON
S10 PICKUP 2WD FFV
RSX
COUGAR
LEGANZA
IMPALA
CARAVAN 2WD
003.7
003.0
002.0
001.8
003.0
003.4
002.0
003.1
003.0
006.0
002.7
001.8
002.2
002.0
002.5
002.2
003.4
003.3
146
Appendix E: VLT ID Numbers Most Likely to Fail the Visual Inspection
VLT
ID
30428
Odds
Ratio
2.036
Model
Year
2001
Make
Model (Transmission Type)
VOLKSWAGEN
NEW BEETLE (Auto)
Engine
Size
002.0
29909
2.038
2001
CHRYSLER
PT CRUISER
002.4
33812
30753
29249
30618
29886
34839
32103
29357
29442
29719
30650
2.044
2.049
2.055
2.060
2.064
2.066
2.066
2.068
2.079
2.080
2.084
2006
2002
2000
2002
2001
2000
2004
2000
2000
2000
2002
NISSAN
FORD
DAEWOO
CHRYSLER
CHEVROLET
CHRYSLER
BMW
FORD
HONDA
VOLKSWAGEN
DODGE
SENTRA
RANGER SUPER CAB 2DR SH
LANOS
PT CRUISER (Manual)
MONTE CARLO
VOYAGER
X5
MUSTANG COUPE (Auto)
PRELUDE
GOLF
DURANGO 2WD
002.5
003.0
001.6
002.4
003.4
003.3
004.4
004.6
002.2
002.0
005.9
51985
2.085
2002
GMC
SIERRA 2500 DIESEL
006.6
30659
29245
30417
29117
29624
29294
34739
30814
30825
33462
31610
32246
30030
32559
33866
29379
30260
29444
33249
29118
30148
52059
29725
2.105
2.124
2.126
2.132
2.137
2.141
2.162
2.177
2.186
2.211
2.236
2.262
2.263
2.263
2.267
2.284
2.284
2.289
2.309
2.313
2.335
2.341
2.353
2002
2000
2001
2000
2000
2000
2000
2002
2002
2006
2003
2004
2001
2004
2006
2000
2001
2000
2005
2000
2001
2003
2000
DODGE
CHRYSLER
VOLKSWAGEN
AUDI
PLYMOUTH
DODGE
CHEVROLET
HONDA
HYUNDAI
DODGE
HONDA
FORD
FORD
SUBARU
SUBARU
FORD
MITSUBISHI
HYUNDAI
VOLKSWAGEN
AUDI
JEEP
CHEVROLET
VOLKSWAGEN
RAM 1500 4WD
TOWN & COUNTRY 2WD
GOLF (Auto)
A4 QUATTRO
VOYAGER 2WD
RAM 1500 4WD
S10 PICKUP 2WD
CIVIC
ACCENT
CHARGER
CIVIC
ESCAPE
MUSTANG COUPE
IMPREZA WRX
IMPREZA
TAURUS WAGON
ECLIPSE
ACCENT (Auto)
JETTA
A4 QUATTRO
CHEROKEE 4WD
SILVERADO 2500 DIESEL
JETTA
004.7
003.3
002.0
001.8
003.3
005.9
002.2
002.0
001.5
006.1
002.0
003.0
004.6
002.0
002.5
003.0
003.0
001.5
001.8
001.8
004.0
006.6
002.0
147
Appendix E: VLT ID Numbers Most Likely to Fail the Visual Inspection
VLT
ID
30653
52063
29591
29966
32021
30731
29910
34764
51859
29219
30419
30668
32436
29921
31027
30729
29586
30108
31695
29445
30739
29593
52087
29358
31452
Odds
Ratio
2.362
2.367
2.370
2.385
2.406
2.433
2.459
2.466
2.475
2.489
2.494
2.494
2.500
2.515
2.535
2.545
2.566
2.584
2.590
2.639
2.662
2.697
2.700
2.707
2.712
Model
Year
2002
2003
2000
2001
2003
2002
2001
2002
2002
2000
2001
2002
2004
2001
2002
2002
2000
2001
2003
2000
2002
2000
2003
2000
2003
Make
Model (Transmission Type)
DODGE
CHEVROLET
MITSUBISHI
DODGE
VOLKSWAGEN
FORD
CHRYSLER
FORD
CHEVROLET
CHEVROLET
VOLKSWAGEN
DODGE
MAZDA
DAEWOO
NISSAN
FORD
MITSUBISHI
HYUNDAI
JEEP
HYUNDAI
FORD
MITSUBISHI
DODGE
FORD
DODGE
30736
2.714
2002
FORD
29286
33794
29965
33193
52209
29941
29968
31943
31528
29714
29298
2.734
2.746
2.768
2.783
2.831
2.867
2.908
2.943
2.950
2.958
3.044
2000
2006
2001
2005
2003
2001
2001
2003
2003
2000
2000
DODGE
MITSUBISHI
DODGE
SUBARU
GMC
DODGE
DODGE
SUBARU
FORD
VOLKSWAGEN
DODGE
INTREPID
SILVERADO 3500 DIESEL
MONTERO SPORT 2WD
STRATUS
JETTA
FOCUS ZX3
PT CRUISER (Manual)
RANGER SUPER CAB 4DR
SILVERADO 2500 DIESEL
METRO
GTI
STRATUS
TRIBUTE
LANOS
SENTRA (Manual)
FOCUS WAGON
GALANT
ACCENT
LIBERTY 4WD
ACCENT (Manual)
MUSTANG COUPE
MONTERO SPORT 4WD
RAM 2500 DIESEL
MUSTANG COUPE (Manual)
STRATUS 4-DR
MUSTANG CONVERTIBLE
(Manual)
INTREPID
LANCER EVOLUTION
STRATUS
IMPREZA STI
SIERRA 2500 DIESEL
INTREPID
STRATUS 4-DR
IMPREZA AWD
MUSTANG COUPE
CABRIO
STRATUS
Engine
Size
002.7
006.6
003.0
003.0
001.8
002.0
002.4
003.0
006.6
001.3
001.8
002.4
003.0
001.6
002.5
002.0
003.0
001.5
003.7
001.5
004.6
003.0
005.9
004.7
002.4
004.6
002.7
002.0
002.4
002.5
006.6
002.7
002.7
002.0
004.6
002.0
002.4
148
Appendix E: VLT ID Numbers Most Likely to Fail the Visual Inspection
VLT
ID
29885
31163
32614
52095
29239
32019
30412
31158
Odds
Ratio
3.082
3.100
3.116
3.175
3.394
3.439
3.706
4.331
Model
Year
2001
2002
2004
2003
2000
2003
2001
2002
Make
Model (Transmission Type)
CHEVROLET
VOLKSWAGEN
VOLKSWAGEN
DODGE
CHRYSLER
VOLKSWAGEN
VOLKSWAGEN
VOLKSWAGEN
METRO
GTI
GTI
RAM 3500 DIESEL
CONCORDE
GTI
CABRIO
CABRIO
Engine
Size
001.3
001.8
001.8
005.9
002.7
001.8
002.0
002.0
149
REFERENCES
Acock, A. C. (2010). A gentle introduction to Stata. College Station, TX: Stata Press
Air Resources Board. (2009). On-board diagnostic II (OBD II) systems – fact sheet / FAQs.
California Environmental Protection Agency, Air Resources Board. Retrieved from
http://www.arb.ca.gov/msprog/obdprog/obdfaq.htm
Austin, T. C., McClement, D., & Roeschen, J. D. (2009). Evaluation of the California smog
check program using random roadside data. Sacramento, CA: Sierra Research, Inc.
Barone, S. (2006). Statistics-driven development of OBD systems: an overview. Quality and
Reliability Engineering International 22(2006), 615-628.
Bin, O. (2003). A logit analysis of vehicle emissions using inspection and maintenance testing
data. Transportation Research Part D: Transport and Environment, 8(3), 215-227.
Retrieved from:
http://www.sciencedirect.com/science/article/pii/S136192090300004X
Bureau of Automotive Repair. (2009). Smog check inspection procedures manual. Retrieved
from http://www.bar.ca.gov/80_BARResources/05_Legislative/Regulatory
Actions/Smog%20Check%20Manual%20w-diesel%208-09%20V3.pdf
Bureau of Automotive Repair. (2010a). Directed vehicles. Retrieved from
http://www.autorepair.ca.gov/80_BARResources/02_SmogCheck/TestOnly_Directed_
Vehicles.html
Bureau of Automotive Repair. (2012a). 2012 Smog check performance report. Retrieved from
http://www.autorepair.ca.gov/80_BARResources/ftp/pdfdocs/2012%20Smog%20Chec
k%20Performance%20Report.pdf
Bureau of Automotive Repair. (2012b). Executive summary report. Smog check statewide.
Year: 2011. Retrieved from http://www.bar.ca.gov/80_BARResources/02_
SmogCheck/Engineering/ExecutiveSummaryReports/ArchivesContents.asp
Bureau of Automotive Repair. (2012c). STAR program. Retrieved from http://www.bar.ca.gov
/80_BARResources/03_Standards&Training/Star/StarProgram.html
Bureau of Automotive Repair. (2012d). New equipment and inspection procedures workshops
presentation. http://www.smogcheck.ca.gov/80_BARResources/05_
Legislative/RegulatoryActions/NewEquipandInspProceduresWorkshops.html
Bureau of Automotive Repair. (2013). Executive summary report. Smog check statewide. Year:
2012. Retrieved from http://www.bar.ca.gov/80_BARResources/02_
SmogCheck/Engineering/ExecutiveSummaryReports/ArchivesContents.asp
150
REFERENCES
California Inspection and Maintenance Review Committee. (2000). Smog check II evaluation.
Retrieved from http://www.feat.biochem.du.edu/assets/reports/CA%20IMRC%20
Report/IMRC20000619Part1.PDF
California State Assembly. (1994). Bill analysis for Assembly Bill 2018: Proposed Conference
Report No. 1. Retrieved from: http://www.leginfo.ca.gov/pub/93-94/bill/asm/ab_20012050/ab_2018_cfa_940316_174230_asm_floor
Cascio, F., Console, L., Guagliumi, M., Osella, M., Panati, A., Sottano, S. & Dupre, D. T.
(1999). Generating on-board diagnostics of dynamic automotive systems based on
qualitative models. AI Communications 12 (1999) 33-43
Choo, S., Shafizadeh, K., & Niemeier, D. (2007). The development of a prescreening model to
identify failed and gross polluting vehicles. Transportation Research Part D:
Transport and Environment. 12(3), 208-218. Retrieved from:
http://www.sciencedirect.com/science/article/pii/S1361920907000132
Eisinger, D. S. (2010). Smog check: science, federalism, and the politics of clean air. New
York, NY: Taylor and Francis Group
Environmental Protection Agency. (2013). Status of state and local OBD I/M programs. United
States Environmental Protection Agency. Retrieved from
http://www.epa.gov/obd/status.htm
Lyons, A., & McCarthy, M. (2009) Transitioning away from smog check tailpipe emission
testing in California for OBD II equipped vehicles. Sacramento, CA: California Air
Resources Board, Mobile Source Control Division
Moghadam, A. K. & Livernois, J. (2010). The abatement cost function for a representative
vehicle inspection and maintenance program. Transportation Research Part D:
Transport and Environment. 15(5), 285-297. Retrieved from:
http://www.sciencedirect.com/science/article/pii/S1361920910000180
Singer, B. C. & Wenzel, T. P. (2003) Estimated emission reductions from California's enhanced
smog check program. Environmental Science & Technology. 37(11), 2588-2595
Sosnowski, D. & Gardetto, E. (2001). Performing onboard diagnostic system checks as part of
a vehicle inspection and maintenance program. United States Environmental Protection
Agency, Office of Transportation and Air Quality, Transportation and Regional
Programs Division. Publication number EPA420-R-01-015, June 2001
Singleton, JR., R. A. & Straits, B. C. (2010). Approaches to social research. New York, NY:
Oxford University Press
Studenmund, A. H. (2011). Using econometrics: A practical guide. Boston, MA: AddisonWestley
151
REFERENCES
Supnithadnaporn, A., Noonan, D. S., Samoylov, A., & Rodgers, M. O. (2011). Estimated
validity and reliability of on-board diagnostics for older vehicles: comparison with
remote sensing observations. Journal of the Air & Waste Management Association,
61(10), 996-1004.
United States Environmental Protection Agency. (1992). 1992 Enhanced I/M support
document. Retrieved from: http://www.epa.gov/otaq/epg/regs.htm
Washburn, S., Seet, J., & Mannering, F. (2001). Statistical modeling of vehicle emissions from
inspection/maintenance testing data: an exploratory analysis. Transportation Research
Part D: Transport and Environment. 6(1), 21 – 36. Retrieved from:
http://www.sciencedirect.com/science/article/pii/S1361920900000110
Wenzel, T. P. & Ross, M. (2003) Comment on “Statistical modeling of vehicle emissions from
inspection/maintenance testing data: an explanatory analysis”. Transportation Research
Part D: Transport and Environment. 8(1), 69-71. Retrieved from:
http://www.sciencedirect.com/science/article/pii/S1361920902000093