Present Serviceability Rating and Present Serviceability Index Concepts

PRESENT SERVICEABILITY RATING
AND PRESENT SERVICEABILITY INDEX CONCEPTS

Paul Irick
The purpose of this presentation is to review concepts that are associated with the
present serviceability rating (PSR) and the present serviceability index (PSI). In his
opening remarks, Carey introduced philosophical and practical considerations for
these concepts. I shall discuss the concepts from an analytical point of view. There
is, of course, much overinp among the philosophical, analytical, and practical aspects
of PSR-PSI concepts. I have separated my remarks into topics that start with general
considerations in experimental research and then progress through the conceptual and
analytical steps that were taken at the AASHO Road Test.
EXPERIMENTAL PAVEMENT RESEARCH

If research by experiment is to be successful, it is necessary to have clear-cut
objectives for the experiment and to have a sound experiment design for the experi-
mental units, their treatments, and their observation. It is also necessary to analyze
the observations in ways that are consistent with the experiment design and objectives.
It is not a simple matter to maintain consistency among objectives, design, and anal-
ysis. At the very least, the experiment design should provide an appropriate basis
for determining relations that are implied by the objectives and that can be expressed
as effects of the design factors on the observed variables.
Road tests are one form of experimental pavement research in which the general
objectives are to learn how pavement behavior depends on built-in pavement charac-
teristics and/or external influences such as environmental conditions and loading
factors. In these terms the experimental units are test sections. Treatments are
those factors that describe how test sections are constructed and subjected to environ-
mental and loading conditions. Pavement behavior is a response, or a combination of
responses, of test sections to treatments that have been applied.
In a road test it is possible to define a large number of variables that describe
elements of pavement behavior, and for each variable there may be several alter-
native measurement systems. At the AASHO Road Test many individual elements
of behavior were observed and analyzed with respect to the design factors. In ad-
dition, it was clearly desirable to define, observe, and analyze pavement performance
as a "supervariable" that could represent the overall response of a test section to
its treatment.
PAVEMENT PERFORMANCE
It was decided that pavement performance should indicate the amount of traffic car-
ried at an acceptable level of service. When more specifically formulated, pavement
performance becomes a "supervariable" for the external behavior of a pavement section
and can therefore be analyzed with respect to experiment design factors.
PRESENT SERVICEABILITY
The concept of pavement performance implies that, from the viewpoint of a per-
ceptive user, a pavement provides a particular level of service at any point in time.
Thus the concept of pavement performance leads to the concept of present service-
ability.
PRESENT SERVICEABILITY RATINGS

If the concept of present serviceability is to be useful for describing pavement per-
formance, then users can presumably discern various levels of serviceability among
pavements whose physical conditions cover the full spectrum of interest. The PSR of
a pavement is a user's judgment of the level of service that a particular pavement pro-
vides at a point in time.
In the AASHO Road Test studies, raters were asked to judge present serviceability
in 1 of 5 categories: 4.0 to 5.0, very good; 3.0 to 4.0, good; 2.0 to 3.0, fair; 1.0 to 2.0,
poor; or 0.0 to 1.0, very poor.
It was found that the average PSR given by a panel of raters was reproducible within
and among various panels. The next question was whether and how PSR was correlated
with objective measures of pavement condition.
PRESENT SERVICEABILITY CORRELATES

Candidates for correlation with PSR include only those variables whose variations
can be sensed by the present serviceability raters. Thus the list of potential correlates
includes surface irregularities and defects that can be measured in terms of longitudinal
and transverse profiles, cracking, spalling, faulting, and so forth.
There are also many alternative PSR correlates for longitudinal profiles, ranging
from surface elevations and their derivatives to measurement system values for the
integrated roughness of a given length of pavement.
PRESENT SERVICEABILITY INDEXES

A PSI was defined to be an algebraic function of PSR correlates. Moreover, the
PSI concept incorporated the view that coefficients in the function should be determined
through multiple regression analysis of the form Y = A0 + A1 X1 + A2 Xz + ... +E, where
Y is an average PSR for a pavement section; X1 , X2 are PSR correlates; A0 + A1 X1 +
A2 X2 + ... is a PSI determined by the regression analysis; and E is a discrepancy be-
tween the PSR and the PSI. In other words, PSR = PSI + E. Multiple regression pro-
cedures were used to find the set of correlates (X1, X2 ,...) that minimized the sum of
E2 and that excluded correlates that didnot contribute significantly to the goodness of
fit provided by those correlates that were included.
PRESENT SERVICEABILITY HISTORIES

A PSI value relates to only one point in time and is not in itself a measure of pave-
ment performance. If PSI values for a particular pavement are plotted against time,
however, it becomes possible to see (or to project) the period of time during which the
pavement provided (or will provide) acceptable service, regardless of the service-
ability level that is selected as being acceptable.
PAVEMENT PERFORMANCE INDEXES

The serviceability-time history of a pavement is necessary but not sufficient for
determining pavement performance because the performance concept includes the
amount of traffic that has been served.
The major performance index used to analyze AASHO Road Test data was log N,
where N was the number of axle loads of a given weight that the pavement carried at
a serviceability level greater than p.
10
ANALYSIS OF PERFORMANCE DATA

The biweekly measurement program at the AASHO Road Test produced serviceability
data for each test section. From these data a PSI was computed, and a PSI history was
thus maintained. At the end of the test, performance indexes were computed (or ex-
trapolated) for p = 2.5 and 1.5.
Finally, the performance index values were analyzed for their relation with pave-
ment design factors and loading factors.
CONCLUDING REMARKS
The main objective of the AASHO Road Test was to determine how pavement per-
formance depends on pavement design factors such as thickness and on loading factors
such as axle load and axle spacing. In this presentation we have tried to show how
performance was defined and evaluated through the concepts of PSR's and PSI's.
The measurement of surface profile was a dominant element for the implementation
of all these concepts. We are therefore pleased to see the continued interest and con-
cern that this workshop shows for the evaluation of surface profiles and pavement
serviceability.

Present Serviceability Rating and Present Serviceability Index Concepts

Uploaded by

Copyright:

Available Formats

Present Serviceability Rating and Present Serviceability Index Concepts

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Present Serviceability Rating and Present Serviceability Index Concepts

Uploaded by

Copyright:

Available Formats

PRESENT SERVICEABILITY RATING

AND PRESENT SERVICEABILITY INDEX CONCEPTS

EXPERIMENTAL PAVEMENT RESEARCH

PRESENT SERVICEABILITY RATINGS

PRESENT SERVICEABILITY CORRELATES

PRESENT SERVICEABILITY INDEXES

PRESENT SERVICEABILITY HISTORIES

PAVEMENT PERFORMANCE INDEXES

ANALYSIS OF PERFORMANCE DATA

You might also like