|
Items in cart [0] |
Questions? Call 888-624-8373 |
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 10
4.0 M.O.E. USER GUIDE
This M.O.E. User Guide provides practitioners with techniques to evaluate truck
weight enforcement activity and applies validated M.O.E.s that were developed and
tested in the current research project. The user guide consists of two parts: sampling
guidelines and a software data analysis tool.
Sampling (Data Colleciion) Guidelines are applied to estimate the number of
WIM data collection sites and required sample sizes required to measure an en-
forcement effect. This guideline provides users with estimates for specified
roadway classification and truck percentage conditions.
Software (Data Analysts) Too! calculates and statistically compares M.O.E. val-
ues between two observed enforcement conditions. This procedure also allows
users to conduct an automated pavement design life analysis, estimating the theo-
retical pavement-life effect resulting from differences produced by the two ob-
served enforcement activities.
It is Important to distinguish between procedural guidelines and a methodological
tool. A guideline (i.e., a method by which to undertake a course of action, which may be
modified at the discretion of the user) provides the user in this case with the starting point
for determining site number and data-collection sample sizes. However, final sampling
requirements in the applied evaluation watt depend upon observed data characteristics,
due to statistical requirements for data stability (i.e., degree of measured variance).
On the other hand, a too! (i.e., an instrument to perform an operation in a speci-
fied manner) is to be strictly applied throughout the evaluation. In fact, the software too}
in this case is designed to refine site-number requirements, on the basis of measured data
characteristics, and to advise the user of final sampling requirements.
in
OCR for page 11
4.1 M.O.E Sampling (Data CotIection) Guidelines
Sampling guidelines described in this section provide practitioners with straight-
forwnrd data-collection requirements to measure enforcement effects using the validated
M.O.E.s. This guideline provides users with estimates of observation site numbers and
associated truck sample sizes. These estimates are provided for specified roadway classi-
fication and truck percentage conditions.
Statistical M.O.E. sampling requirements were based on an analysis of nationwide
WIM data. This developmental analysis examined M.O.E. data generated for representative
locations (i.e., exhibiting prerequisite highway functional classification and truck mix crite-
ria) and determined the minimum number of observation sites required to produce repre-
sentative M.O.E. distributions. Based on these results, M.O.E. sampling guideline proce-
dures were developed to enable users to estunate equivalent sampling requirements.
Sampling guidelines are directed toward WIM database gathering. It must also be
emphasized that the soundness of the WIM input data and its subsequent analysis to
measure the effectiveness of truck weight enforcement is highly dependent on calibration
and maintenance of that equipment.
Users of this Sampling Guide are not expected to apply expertise in the area of
statistics. However, due to the fact Mat this guide was developed via the application of
various statistical concepts that affect its output, two statistical concepts and their appli-
cation in the guides development are briefly explained as follows.
Sampling requirements contained in the guide utilized two statistical concepts,
Level of Significance, and Power of Test. Each of these terms is defined as follows, only
as a matter of information for users of this guide.
Level of Significance refers, in this case, to the probability that the user is willing to
risk the error of rejecting a valid change in M.O.E. occurrence. In statistical terminology,
OCR for page 12
the Level of Significance is the maximum probability with which we would be willing to
risk a Type I error. A Type 1 error occurs when a true hypothesis is rejected, i.e., that base-
line (no enforcement) versus enforcement M.O.E. vanable sets are statistically different.
The .05 Level of Significance was applied in Me development of this guide.
Power of Test refers to the likelihood of making a correct statistical assessment, i.e.,
that the proper hypothesis is accepted, statistically speaking. The issue is to what extent is
the user milling to risk accepting an invalid change in M.O.E. occurrence. In statistical ter-
minology, the Power of a Test is the maximum probability with which we would be willing
to risk a Type 2 error. A Type 2 error occurs when a false hypothesis is accepted, i.e., Mat
baseline versus enforcement M.O.E. variable sets are not statistically different. The .80
Power of Test was applied in the development of this guide.
4.~.~. Sampling Observation Levels
Separate observation levels for sampling truck-weight violations were devised in
order to meet the diverse evaluation requirements of varied truck weight enforcement op-
erations. Three designated sampling observation levels are as follows: (1) statewide or
regional, (2) highway corridor or local level, and (3) spot or location-specific. Figure 1
on the next page is a conceptual representation of the three designated observational lev-
els.
At the broadest level, the implementation of revised regional or statewide policy
may require sampling over a vast geographic area, covering hundreds of square miles. At
the opposite end of the spectrum, spot truck-weight enforcement procedures are fre-
quently required due to location-specific factors, e.g., pertaining to local hauling condi-
tions. Finally, a major concern for enforcement and highway agencies is weight-law
compliance along specific highway corridors. The critical nature of weight monitoring
alone corridors stems from a number of factors, including trucker avoidance of weight
enforcement and costly pavement damage to local highways.
12
OCR for page 13
t: :.
A: ~ :-:::::
A::::::.:: :~::
A:: ::::::::.: :::
A. ~: ~: :-: ::: :.
:
::: :
it: ~ . ~:: . ~
I::
~ :::::: .
A::: : :: ::
:: ::::. ::.: :::: :::::
- .~;(ide.J'
: .: .... ^.R.eg':ona.'.l.-'.::. ':
Corridor't-~ ~~ at' ~ 'at
I.,: :., :.: :. .-. -. ,., : .-. : :: . : -: . . -. - .:. .:: :, . ~ Location
_ _ . ................ . .. ... - ...... . . . . _ . ..... _ . ~
Figure ~ . Illustration of Varied Data-sampling Observation Level Concept
4.~.2 Statewide or Regional M.O.E. Sampling
Statewide or regional M.O.E. sampling is applied to evaluate any truck weight en-
forcement program that affects large geographic areas which exceed the bounds of a defin-
able highway corridor. The derivation of sampling requirements was based on actual ob-
served statewide M.O.E. distributions; however, this guide is also applicable for smaller
geographic regions. Site number requirements contained in this guide indicate minimum
numbers to produce representative results for a designated region. Data collection site re-
quirements are designated on the basis of regional characteristics, i.e., highway functional
class and associated truck percentage combinations, which comprise the area under study.
An example application of this procedure is shown in Section 4.~.3 ofthis report.
WIM Data Site Number Requirements Guidelines for determining the required num-
ber of observation sites for a statew~de/regional study of truck weight enforcement e~ec-
tiveness were determined on the basis of observed M.O.E. distributions2 from representative
See Appendix F.
13
OCR for page 14
nation-we locations. Site number requirements for a designated region were based on the
region's composition in terms of specified highway functional classification and associated
truck percentage criteria.
The user's deteITnination of study site numbers shall commence via the application
of the guidelines shown in Table 2, which specify site number requirements for each func
tional-cIass/truck-percentage category. Site numbers indicated in the table are intended as a
starUng point for establishing final regional observation site number requirements. The
~ta-analysis software generated in this project is designed to refine site number requirement
based on individual user's specific data characteristics.
Table 2. Minimum Site-number Guideline for Selected M.O.E.s
in State/Regional Truck Weight Enforcement Evaluations
Rural Interstate
< 15 % Trucks
15to30% Trucks , ,
> 30% Trucks
Rural Primary Arterial
< 9 TO Trucks
9 to 30 % Trucks
l l
> 30% Trucks
Rural Minor Arterial
Urban Interstate
< 9% Trucks
. .
TANDEM AXLE :- i: i: ~ -:
VIOLATIONS ~::~:: --: i: ~
--SINGLE- Amp- E--
:-~0~4T~ -~-
. ~ ! ~
EXPRESS :
~E~ --- : ~:~
Urban Primary Arterial
< 9% Trucks
> 9% Trucks , ,
NOTE: flue accompanying NCHRP Project 20-34 software generates site number requirements based on
user's data.
The statewide/regional M.O.E. sampling procedure involves two preparatory steps.
First, the geographic area, e.g., jurisdictional territory, to be affected by the enforcement
program under study must be clearly defined. Second, the highway network within the de
fined study region must be reviewed to determine its composition, in terms of route fi~nc
tional classification and associated truck percentage as a function of overall traffic volume,
on each affected route.
14
OCR for page 15
The initial number of required study sites is then determined on the basis of corre-
sponding site-number designations shown in Table 2, subject to revision by the associated
software. The total number of study sites In a given region wait be the sum of those applied
in each functional class and truck-ratio which are represented in the region, as demonstrated
In the next paragraph. Each functional class represented In the region under study must be
included in the array of designated observation sites.
For example, when designating We primary M.O.E. of interest to be the "Proportion
of Gross Weight Violations", then the number of required sites for each highway category
watt be denved from numbers shown in the left-most column of Table 2. That is, at least
three data collection sites are required to represent Rural Interstates with less Man ~ 5 percent
trucks, six to represent Rural Interstate s with ~ 5 to 30 percent trucks, etc. The total number
of sites for the study region wall be equal to the sum of site numbers for all functional-
cIass/truck-percentage categories represented In the region, i.e., 36 sites. This procedure is
illustrated In Me example application of a regional sampling plan development shown in
section 4.~.3.
It is important to note a number of user precautions and associated considerations
underlying the development of site numbers contained in Table 2. These caveats relate to
the analysis and application of nationwide representative data used to estimate requ~re-
ments for conducting a regional truck weight enforcement evaluation.
First, the nationwide analysis determined that a single observation site, within
selected functional-cIass/truck-percentage categories, was occasionally sufficient to sta-
tistically detect certain enforcement effects. However, application of sound sampling
strategy to a regional enforcement study requires a significant degree of generality to en-
sure its validity, therefore, Table 2 mandates a minimum of two sites for each functional
highway classification condition.
OCR for page 16
Second, site number requirements outlined in Table 2 were based on observed
M.O.E percentage reductions found to be associated with enforcement activity3. However,
for situations in which an observed enforcement activity is expected to produce greater or
lesser percentage M.O.E differences, an appropriate adjusunent to the number of observa-
tion sites would be required to statistically measure the effect. For example, in a given re-
gion where 7 data collection sites may be required to detect an LO-percent reduction in gross
weight violations, only 5 sites would likely be required to detect a 20-percent reduction.
Importantly, with the current application, the user wail be appropriately informed of the level
of affected M.O.E. change (and the associated number of required sites to validly observe
this effect) via application of the software package accompanying this guide. Ike software
application is explained in Section 4.2 of this report.
Third, site numbers designated in Table 2 were based on measured statistical M.O.E.
distributions. By taking into account normal sample sizes and associated variability of these
M.O.E.s, they indicate the number of observation sites required to capture representative
M.O.E. distributions. However, a number of application-specific considerations are neces-
sary in He user's interpretation of the table. Specifically, truck weight surveillance over a
large geographical area may logically require larger site numbers Han indicated In He table.
For example, many cells in He table indicate the necessity of only 2 or 3 study sites, given
certain highway classification and truck ratio conditions. Yet, in He case of a statewide en-
forcement program over a very large area, the limitation of 2 or 3 study sites may be consid-
ered inadequate.
Thus, the final designation of observation sites must consider prevalent conditions,
e.g., specific hauling and commodity demands that affect truck-loading operations and the
subregional areas to which they apply. Specifically, He user is cautioned against combin-
ing sites characterized by known non-homogenous loading conditions when applying the
sampling procedure.
3 See Appendix F. Tables F-40 through F-42.
16
OCR for page 17
Finally, as previously noted, Table 2 is a guideline (i.e., a procedure by which to
undertake a course of action, which may be modified at the discretion of the user) to pro-
vide the user with the starting point for determining site number and data-collection sam-
ple sizes. Its final application relies on engineering judgement in the context of specific
study situations.
Designation of Data Collection Periods In view of known commodity shipping pat-
terns, both weekend and weekday data collection periods are recommended in applied re-
gional M.O.E. sampling efforts to evaluate truck weight enforcement programs. Impor-
tantly, designated data collection periods need to be sensitive to seasonal conditions, e.g.,
agncultural commodity hauling patterns. A minimum two-day data collection duration is
required at each site for each observed enforcement condition.
Based on NCE~P Project 20-34 findings, the user is advised to expect maximum
violation to occur during the early morning hours, e.g., 3 a.m. to 7:30 a.m. on weekdays, and
during Me late evening hours on Sundays.
Minimum site-specific truck sample sizes are shown In Table 3 for designated com-
binabons of highway functional class and associated truck percentages for designated
M.O.E.s. Sample size estimations shown in the table are based on the requirement to de-
tect differences in truck proportions exhibiting the array of generally applied M.O.E.s at
the .05 level of statistical confidence.
4.~.3 Example of a Regional M.O.E. Sampling Application
Consider the hypothetical example of a geographic region with a distribution of
100 WIM data collection sites as shown in Table 4. This distribution was estimated on
Me basis of traveled vehicle-miles by functional cIassification4 with adjusunents for traf-
fic monitoring prioritization.
4 U.S. Department of Transportation, Bureau of Transportation Statistics, National Transportation Statis-
tics, Washington, D.C. 1996
17
OCR for page 18
Table 3. Minimum Site-specific Number of Required Truck Observations
.
. :~C~.~SS' -I ~ -'' ~'].m9D-Mid, - .,~, :-:,-. ,
.~ .-A- -- :- :~ ~
.. .. .
Rural Interstate
< 15 % Trucks 175
l5to 30 % Trucks 300
> 30% Trucks 200
Rural Primary Arterial
< 9 % Trucks 225
9to 30 % Trucks ~325
> 30% Trucks 100
Rural Minor Arterial 200
Urban Interstate~
< 9% Trucks 100
> 9% Trucks 200
Urban Primary Arterial
< 9% Trucks 125
> 9% Trucks 100
* Over a minimum 2-day data collection period.
The assignment of available WIM sites to monitor an ongoing regional truck
weight enforcement program, according to the scheme previously shown in Table 2, pro-
duces the sampling scheme shown in Table 5 on the next payee
Table 4. Available WIM Monitoring Locations
in Example Sampling Application
A...::..
:',,2~' ?.,, .. i.,x;,: j,. ,.":; ~ i?.. be. ,,... :.:.A,.,.... ~
'2'".~. .Id~.. Bilge''"' i"'''' - ~A~...~.S~
Rural Interstate
< 15 % Trucks
15to 30 % Trucks
_
> 30% Trucks
.
Rural Primary Arterial
< 9 % Trucks
9to 30 % Trucks
> 30% Trucks
Rural Minor Arterial
Urban Interstate
< 9% Trucks
> 9% Trucks
Urban Primary Arterial
< 9% Trucks
> 9% Trucks
TOTAL
4
8
4
7
7
14
15
8
8
11
100
18
OCR for page 19
An examination of Table 5 indicates that of the 100 available WIM-monitor~ng
sites, only 36 sites are required for region-wide monitoring of two M.O.E.s, i.e., Gross
Weight Violations, and Tandem Axle Violations, on all highway fi~nctional-class/truck-
percentage categories. In order to obtain a non-biased estimation of truck weight en-
forcement effects, the user agency is advised to assign data collection locations in a ran-
dom fashion (within appropriate fi'nctional-class/truck-percentage categories) when all
available WIM installations are not statistically required for the evaluation.
A larger number of data collection sites is required within a region to statistically
represent less-frequently-occulting M.O.E.s (See Appendix F Sampling Plan Develop-
ment). Consequently, in the cuIrent example, the latter two M.O.E.s required more data
collection sites within certain fi~nctional-class/truck-percentage categories than were
available. In these instances, the percentage of available WIM sites is indicated In the
appropriate cells of Table 5. For example, while 8 sites were previously suggested in the
Table 2 Guide as the minimum number of sites to detect enforcement effects in terms of
Single-Axle Violations, the 4 available sites comprise 50% of this requirement. In such
instances, Me regional evaluation is necessarily limited by WIM-site availability, and the
issue site selection bias defers to applied site-location decision rationale.
Table 5. Reco~runended WIM Data Collection Site Distribution
for ~> ~_~-~ ·< ~
Rural Interstate
< 15 % Trucks
15 to 30 °/0 Trucks
> 30°/0 Trucks
Rural Primary Arteriai
< 9 °/0 Trucks
9 to 30 °/0 Trucks
> 30°/0 Trucks
Rural Minor Arterial
Urban Interstate
< 9°/n Trucks
2
2
Urban Primary Arterial
< 9°/0 Trucks
> 9°/0 Trucks
_
TOTAL
19
4 (50%)
8 (38%)
4(31%)
7 (64%)
7 (29%)
14 (58%)
~ 31~)
T
74
4 (44%)
8 (25%)
4 (12%)
7 (47%)
14 (93%)
9
8 (80%)
15
5 (10%)
1 1 (79°/0)
87
OCR for page 20
The non-availability of 100 percent of the Sample Guide's recommended site
numbers does not necessarily mean than the region can not be evaluated in teens of the
latter two M.O.E.s in this instance. Conversely, more sites may be needed in some in-
stances than indicated by the Table 2 Guide. The reason is that the exact number of re-
quired sites is determined by the data variance that is actually measured. Again, we em-
phasize that site numbers indicated in Table 2 are guidelines, based on nationwide obser-
vations of expected M.O.E.s vanances, and these estimates are prescribed as the starting
point for development of the final sampling plan.
Precise site number requirements are determined via application of the data analy-
sis software developed in this project, the Truck Weight Enforcement Evaluation Too!
(TWEET), described in Section 4.2 of this report. This software computes customized
site number requirements based on the user's collected data. Specifically, it performs
site-number requirement calculations based on actual measured variances, as is statisti-
cally appropriate. Thus, this process provides the necessary site-number refinement cal-
culation to define final sampling requirements. However, it can not replace the Table 2
Guide, as the user needs initial estimates for evaluation study planning purposes.
The data analysis software contains a "Sampling Guide" dialog box that computes
site number requirements for various levels of statistical precision (see Figure 2~. The ex-
ample dialog box in the figure hypothetically considers data collected at 36 sites. This is
He minimum number of prescribed sites in Table 2 assuming that the study region con-
tains sites in all eleven functional-cIass/truck-percentage categories. This software sam-
pling aid prescribes site-number requirements as a function of He specific enforcement-
program effectiveness threshold, i.e., designated percent change in specific M.O.E.s, that
the user wishes to consider. For example, looking at site-number requirement shown in
the figure for Gross Weight Violators M.O.E., we see that if users want to detect an en-
forcement effect based on an expected 40-percent violation reduction, only two data col-
lection sites are required. However, to apply a more rigorous statistical requirement, for
example a statistical test that is sensitive to a ten-percent reduction, seven sites would
Hen be required.
20
OCR for page 29
File Name:
enforce.p~n
Directories.
] .c:\tweetl 2
List Files of Type:
Data files (' tWI, -.pi
Olives:
......
3c: 1uS-DOS_61 ~
Figure 9. The "Data File for Enforcement Condition" Dialog Box
The Pavement Analysis dialog box (Figure 10) provides the user with an option to
conduct a pavement design-life enforcement-effects analysis. The program asks for specific
(and detailed) pavement design data. Because of the complexity of the pavement design-life
analysis, He user has the option of skipping the pavement analysis simply by clicking the
'skip pavement analysis' option.
Depending upon whether the user selects Flexible or Rig~d pavement, different
variables appear in the Pavement Characteristics portion of the dialog box. This box will
prompt the user for appropriate pavement design parameters. A comprehensive "Help"
screen associated with the Pavement Analysis Dialog boxes explains the design theory,
including He AASHTO design equations, underlying He computations utilized in the
software.
As further assistance to He user, Appendix E to this report contains a comprehensive
explanation of relevant pavement design considerations and background references
regarding overweight axle effects on pavement life.
29
OCR for page 30
Please enter information about the characteristics of your pavement.
If you are unsure about any of these, you may use the default values.
~ Skip pavement analysis .
... _- .. .._ ~_.. ......... ~__.,
rS erect Pavement lulaterial- -
fib Flexible [i e. Asphalt]
C, Rigid [i e., Portland Cement Concreted
Flexible Pavement Characteristics
Please enter values for each of the following pavement
variables. If you do not know one or more of the values for
your particular pavement use the defaults. as they were
chosen as the most probable values for a flexible pavement.
If you do not know the value of SN. but you do know the
materials which comprise: your pavement, press the "Calculate
SN .~.' button, and TWEET will compute SN based on the
material composition of the pavement.
SN: ~ I
Po 14-2 1
1 1
hi R 15000
ZR: |-1 64
So: .35
Figure 10. The "Pavement Analysis" (Flexible Example) Dialog Box
1
Flexible pavements are discussed first. Default values are shown on the dialog
box for the following parameters.
.
SN Pavement Structural Number. TWEET offers Me option of computing this variable
based on input values provided by the user.
pa Initial Serviceability Index
pi Tenninal Serviceability Index
MR Default Resilient Modulus
ZR Standard Normal Deviate corresponding to design reliability
SO Standard Deviation associated wad pavement performance prediction
The above parameters are defined arid their design Implications are explained In detailed
'Help' screens in the software.
Because of the importance of the pavement's Structural Number (SN), TWEET
provides the user with alternative approaches to its calculation. First, the user may accept
3Q.
OCR for page 31
the commonly applied value shown as the default. Second, the user may apply a known
value for SN, based on his knowledge of the site. Third, if the user knows the material
composition of Me pavement, TWEET can automatically calculate the SN value. In this
case, the user clicks on 'Calculate SN', and the Automatic Calculation of SN dialog box
appears as shown in Figure ~ I.
The dialog box shown in We figure allows the user to select We appropriate surface,
base, and sub-base charactenstics, i.e., pavement layer thickness (in inches), and strength
coefficient. According to We specified matenal type, the program wait suggest We most
appropriate default Strength Coefficient. Pavement materials and design personnel who curl
this software have the option of overriding default values, depending upon their own
knowledge of pavement matenals and design procedures along win specific pavement
characteristics associated wad We truck weight enforcement location.
._ _ . ~. ......... ....................... ...... . - ~
;~ A: lo,:- ;~ :` pi ~ · !~
. . . ... ...... . . .... . . .. ..... ..... . , , . .... it. ,. ~ ~ . . ... .... . ............................................................................
Please enter the correct information about your parement by checking the radio
buttons below. and press OK when you are finished TWEET will then calculate your
pavement's structural number {SN] and set that number in the SN entry field in the
Pavement Analysis dialog box Each radio button has a default Strength coefficient
associated with it; however if you know these values exactly you can enter them
direct4 in their ~espet:tive fields. Press Help for more information.
-Surface Characteristics
~.~.High stability asphalt concrete.
_~ ~.~_~_.~.~
C) Low stability asphalt concrete
Thickness: ~|
Strength coefficient: [-~4 |
1
-Subbase Characteristics
~ Sandy gravel
C' Sand or sandy clay
Thickness: l
.
Strength coefficient: .11 1
- Base Characteristics
At. Gushed Stone
C; Sandy gravel
~ Cement treated stone
G.4sphalt treated stone
~ Lime treatment
Thickness: l l
Strength coefficient: .14
-Second Subbase Characteristics
~ Sandy gravel
C' Sand of sandy clay
Thickness: [=
Strength coefficient ~ 11
Figure ~ i. The "Automatic Calculation of SN" Dialog Box
~3 ~
OCR for page 32
In the event Me user had selected Rigid Pavement, the Pavement Analysis (Rigic!
Pavements dialog box would appear, as shown in Figure 12.
; vocables. If you do not l:now one or ~ e at the wa yes for
i ~ yo ~ particular pavement use the defaults, as they were
chosen as the most probable values for a rigid pavement.
k: ~ 00 1
Please enter information about the characteristics of your pavement.
If you are unsure about any of these, you may use the default values.
Skip pavement analysis.
-Select Pavement 1~1ate~i~1
C~ Flexible [i.e., Asphalt]
, _ ~. ~_ ~_
Nonrigid [i e, Portland Cement Concrete];
........ _ . ~_ _ ~ ._
-Rigid Pavement Cha'.cteristics -
Please enter values for each of the following pavement
E- 5000000 1
id,
11
1650 1 ;
Po- 14-5 ~
D: 11 To | Pt i§-5 l
Figure 12. "Pavement Analysis" (Rigid Example) Dialog Box
This dialog box provides the following design values for user application:
k
E
D
s c
Po
Modulus of Subgrade Reaction
PCC Elastic Modulus
Slab Thiclmess (inches)
Modulus of Rupture
Initial Serviceability Index
Terminal Serviceability Index
As was the case with the Flexible Pavement Charactenstics box, the most likely de-
fault design values have been provided in the case of Rigid Pavements. The user has the
option of manually entering values specific to the highway study site.
32
OCR for page 33
The program wall then read the WIM data files and perform all calculations. Unless
data files are extraordinanly large, these calculations should take no more than a few
seconds. An animated graphic Status dialog box (Figure 13) will appear to advise of the
program's progress on the computational process. The truck on Me screen moves from left
to right on the roadway section as the calculation is completed.
Calc^~;n~..
71 ~ din
o% 108~
_1_
~ _ . . .
Figure ~ 3. "Calculation Status" Dialog Box
4.2.2 Viewing Results of Calculations
Once We calculations are completed, the user wall be presented with a series of "out-
put" dialog boxes that display calculated values based on input data. The first M.O.E. out-
put dialog box, Severity of Violations (Figure 14 on the next page), also reports sublunary
information, i.e., enforcement condition, highway type, total vehicle, and Muck sample. The
first part of the dialog displays Me observed number of violations, i.e., gross vehicle
weight, single axle weight, tandem axle weight, tandem axle weight, and Bridge Formula
violations. The second dialog displays the average number of overweight pounds (or
Metnc equivalent) for each grouping noted above.
The Calculatecl Percentages of Overweight Trucks dialog (see Figure 15 on the
next page) displays the calculated percentages of overweight trucks in the sample. It lists
four calculations based on the data files, i.e., (1 ~ percentage of trucks over the legal gross
weight limit, (2) percentage of trucks over the single axle weight limit, (3) percentage of
trucks over the tandem axle weight limit, and (4) percentage of Sucks violating the
Bridge Formula.
-$ ~
~-
OCR for page 34
Enforcement Condition: Baseline Condition
-Number of overweight trucks
Gross Vehicle Weight:
Single Axle Weight:
Tandem Axle Weight:
T'idem Axle Weight:
Bridge Formula Violations:
33
82
47
o
13
out of
out of
out of
out of
out of
266
266
266
266
266
trucks total
trucks total
trucks total
trucks total
trucks total
-average Ibs overweight ~
1
Gloss Vehicle Weight: 6872 Ibs overweight
Single Axle Weight: 1304 Ibs overweight
Tandem Axle Weight: 3393 Ibs overweight
T'idem Axle Weight:
Bridge Formula Violation:
o
2307
Ibs o~relweight
Ibs overweight
Figure 14. "Seventy of Violations" Dialog Box
Enforcement Condition:
Baseline Condition
Percentage of trucks over the legal gross weight limit:
1~41
Percentage of trucks over the legal single axle weight limit:
Percentage of trucks violating the legal tandem axle-weight limit: 17.67Z
Percentage of trucks Violating the legal t~idem axle weight limit:
Percentage of trucks violating the Bridge Formula:
30.83:t
O OOZ
~ 89Z
Figure ~ 5. "Calculated Percentage of Overweight Trucks" Dialog Box
The Violation Data by Truck Classification dialog box (Figure 16) indicates viola-
tors, by Duck number and percentage, for each class of Duck. This dialog box displays
OCR for page 35
violation information, broken down by truck classification. This information is usefiA in
determining violation distributions according to truck type.
1. i,` ~ 1,..e L i;. :,, ~ ~ ;
Enforcement Condition Baseline Condition
Select truck classification:
,
i, ,.
. .
i:
.
iE2
I:
. -
. .
::
.
, .
E:
1.
I.,
~2
At
. '
-Violation Data for Selected Classification: ------
l~lotorcycles
Passenger Cars
O the' Two-Axle, Fou'-Tire Single Unit Lehigh
Buses
Two-Axle, Six-Tire. Single Unit Trucks
Thee-Axle Single Unit Trucks '`
Four or Hare Axle Single Unit Trucks
Fou' or Less Axle Sinule Trailer Trucks
~ _ ~
~ ~_~
Total Number of Tracks:
Number of Violators:
Percentage Violating:
Proportion of total violations
committed by this type:
211
78
36.97:t
87.64z
Figure 16. "Violation Data by Truck Classification" Dialog Box
The dialog consists of two parts:
Truck Classification List Box This box lists all of the truck classifications which
were input by the user during the beginning of the analysis, or if Me default was
selected, the FHWA 13-type classifications.
Violation Data This part of the dialog lists violation data for the currently selected
truck classification. First, the Total Number of Trucks field displays the number
of trucks of the selected type which were in We sample (regardless of whether
they were violators). Second, the Number of Violators field lists the number of
trucks of the selected type that violated the weight limits. Third, the Percentage
Violating field lists the percentage of trucks of the selected type which were
violators (this percentage is simply the Number of Violators divided by the Total
Number of Trucks). Finally, the proportion of the tote sampled violations
represented by the selected truck class is indicated.
35
OCR for page 36
The Breakdown of Violations by Day-of-Week (not shown) dialog then displays
the percentage of violations occuning on each day of the week. The dialog simply lists
each day of the week, and next to it lists the percentage of all violations which occuITed
on that day.
The Breakdown of Violations by Time-of-Day (not shown) dialog then displays
the percentage of violations occulting at different hours of the day. Because it would be
overly complex to display the percentage of violations occurring at each of Me 24 hours
of the day, the five hours with the most violations are listed. If it is necessary to know
what percentage of violations occulted at every hour of the day, the Print option will be
of use. The printed copy of the data, urdike We on-screen display, does display We per-
centage of violations for each hour of the day.
The f?SAL Data dialog box (Figure 17) indicates average ESAL calculations using
the FHWA Traffic Monitoring Guicle procedure according to We number of axles. This
dialog also indicates computed Excess ESAL violations by Suck axle-count.
Enforcement Condition:
it.
I!
..,
i:
if?
;}
.
ok?
? i
Baseline Condition
-Average Number of ESALs:
Truck Type: Average ESALs: Average Excess ESALs:
2-axle trucks .2468189 0
3-axle trucks .4756302 0
4-axle trucks 2083355 0
5-axle trucks 1.622841 1.573099
6-axle trucks 0 0
7-axle tucks 0 0
All Trucks: 1.?419015 1.573099
Figure ~ 7. The "ESAL Data" Dialog Box
36
OCR for page 37
Now, TWEET begins its presentation of the data analysis results.
The Comparison
of Enforcement Conditions dialog (Figure I8) indicates to the user whether or not the ob
served M.O.E. differences are statistically significant. This dialog box contains results of
applied statistical significance tests to the computed M.O.E.s and indicates to the user
whether or not Me observed differences are significant. Separate tests of statistical signifi-
cance are applied to M.O.E.s depending upon whether the measure was calculated as a mean
(i.e., average gross weight violation) or a proportion (i.e., proportion of gross weight viola-
tors). Significance tests are applied at Me .05 level of statistical confidence.
This dialog allows you to determine the effectiveness of your enforcement
activity by comparing the differences in the calculated violation data for each
enforcement condition. For each I.IOE, a check has been placed in either the
"Significant.' or "Non-significant " column. This shows whether the difference
in the calculated value of that MOE between the first and second enforcement
condition is statistically significant. Press Help for more information
insignificance of Proportions and l~leans
.
HOE
Percentage of Gross Weight Yiolators
Percentage of Single Axle Weight Yiolators
Percentage of Tandem Axle Weight Yiolators
Percentage of Tridem Axle Weight Violators
Percentage of Bridge Formula Violators
Average Pounds over the Gross Weight Limit
Average Pounds over the Single Axle Limit
Average Pounds over the Tandem Axle Limit
Average Pounds over the Trident Axle Limit
Average Pounds over the Bridge Formula Limit
Average ESALs
Average Excess ESALs
Sionificant Won-sionificant
f
Figure I8. "Comparison of Enforcement Conditions" Dialog Box
The Sampling Guide dialog box (Figure 19 on page 39) page is an aid to determine
how many sites wall be needed to be surveyed in order to detect regional changes for desig-
nated M.O.E.s given specified levels of statistical confidence. The user is first presented
with a "Sampling Guide Options" table, allowing the option of specifying two parameters
37
OCR for page 38
related to the precision of Me statistical estimate. These are the desired [eve! of Significance
and Me Power of Test, explained as follows.
[eve! of Significance refers, in this case, to Me probability that the user is willing to
risk the error of rejecting a valid change in M.O.E. occurrence. In statistical terminology,
the Level of Significance is the maximum probability with which we would be willing to
risk a Type ~ error. A Type 1 error occurs when a true hypothesis is rejected, i.e., that base-
line versus enforcement M.O.E. variable sets are statistically different. In practice, a sign~fi-
cance level of.O5 or .01 is customary.
Power of Test refers to Me likelihood of making a correct statistical assessment.
This is achieved when the proper hypothesis is accepted. At issue is the extent to which the
user is willing to risk accepting an invalid change in M.O.E. occurrence. In statistical te~mi-
nology, the Power of a Test is the maximum probability with which we would be willing to
risk a Type 2 error. A Type 2 error occurs when a false hypothesis is accepted, i.e., baseline
versus enforcement M.O.E. variable sets are not statistically different.
The main feature of the Sampling Guide dialog box is a table indicating the number
of sites which are required for data collection if specified levels of M.O.E.s changes (i.e., 5,
10, 15, or 20 percent) are to be detected. These numbers are based on TWEET's analysis of
the measured statistical characteristics (e.g., variance) of Me observed M.O.E.s. The user
will note Mat fewer sites are necessary for larger differences. This effect is due to the fact
that smaller differences in real-world truck-weight enforcement compliance are subtler and
therefore require more statistical rigor to detect.
The final dialog box (Figure 20 on the next page) presents results of Me Pavement
Effects Analysis. Results contained in this dialog box are based on a theoretical pavement
design-life effect, associated with differential enforcement-related ESAL loading conditions.
Had the user opted to include the pavement design-life effect computation, this screen
would be displayed. Displayed information consists of the calculated pavement ESAL ca-
pacity, the estimated pavement life under both observed enforcement conditions, and esti
38
OCR for page 39
mated percentage pavement-life charge due to the observed ESAL-Ioading difference asso
ciated with Me enforcement activity.
-Sampling Guide Options -
Level of Significance:
Power of Test: | 0.80
s ~. . ,
This guide is intended to assist you in determining the number of data collection
t sites required to detect specified level* of change for various hlOEs You can
i~ change the following options to control the creation of the sampling guide
it
Hi
.
. ~
it
.. s
.,
. ~
i}
Cal current number of sites:
Number of Required Data Collection Sites
~ MOE
Percentage of Gross Weight `Violators
Percentage of Single Axle Wt. Violators
Percentage of Tandem Axle Wt. Violators
Percentage of Bridge Formula Violators
Average Pounds Over Gross Weight Limit
Average Pounds Over Single Weight Limit
Average Pounds Over Tandem Weight
Average Pounds Over Bridge Weight Limit
Average ESALs
Average Excess ESALs
Percent change to be detected
_ 10 15 20
26
12
26
25
82
77
78
106
369
246
3
6
6
41
38
39
53
185
123
3
3
3
27
26
26
35
123
82
2
1
2
20
19
20
27
92
61
Figure 19. "Sampling Gliide" Dialog Box
Calculated pavement E SAL capacity:
Estimated pavement life BEFORE enforcement activity ~rears]:
Estimated pavement life AFTER enforcement activity [years]:
Percentage increase in pavement life due to enforcement activity:
4162490
21.s
23.4
8.81 2:
Figure 20. The "Pavement Effects Analysis" Dialog Box
39
Representative terms from entire chapter:
truck weight
