Details:

Alternative Title:
Prediction Models and Feedback System
Creators:
George, K. P.
Corporate Creators:
University of Mississippi. Dept. of Civil Engineering
Corporate Contributors:
Mississippi. Dept. of Transportation. Research Division ; United States. Federal Highway Administration
Mississippi. Dept. of Transportation. Research Division ; United States. Federal Highway Administration Less -
Subject/TRT Terms:
[+]
Condition Data Data Files Distress Model Feedback Control Feedback System Mathematical Prediction Pavement Distress Pavement Management Pavement Management Systems Pavement Performance Performance Model Prediction Model Statistical Analysis
Publication/ Report Number:
FHWA/MS-DOT-RD-00-119
Resource Type:
Tech Report
Geographical Coverage:
Mississippi
Edition:
Final; Aug. 1995-Oct. 2000.
Corporate Publisher:
Mississippi. Dept. of Transportation
Abstract:
As a primary component of a Pavement Management System (PMS), prediction models are crucial for one or more of the following analyses:

maintenance planning, budgeting, life-cycle analysis, multi-year optimization of maintenance works program, and authentication of design

alternatives. The main focus of the study is to develop pavement deterioration models. Four cycles of pavement condition data and the required

inventory data are compiled from the Mississippi Department of Transportation (MDOT) PMS database. Though regression is the primary tool for

developing models, Bayesian regression is also employed whenever feasible. Expert opinion regarding the major distresses in pavements are

compiled, augmenting the field data. The study begins with a review of relevant literature with the aim of identifying the commonly employed

explanatory variables and various model forms.

Five pavement families are identified for the model development: original flexible, overlaid flexible, composite, jointed concrete, and continuously

reinforced concrete pavements. Models for each family are developed for predicting distresses, roughness, and a composite condition index

(Pavement Condition Rating). The database employed is divided into ‘in-sample’ data constituting a major portion (70 percent) with ‘out-of-sample’

data comprising the remaining. Totally 26 models are developed, with the in-sample data: six each for original flexible, overlaid flexible, and

composite, and four each for jointed concrete and continuously reinforced concrete pavements. The models are subsequently verified with the ‘outof-

sample’ data.

Among the scores of model forms attempted, power form or some variation of it fits all of the models while satisfying crucial boundary conditions.

The out-of-sample data provides an independent database to verify the validity of the models. A sensitivity analysis of the model equation is

presented in each case, substantiating the predictive capability of the model. In seven cases, incorporating expert opinion in the field data, employing

Bayesian regression, resulted in better prediction models. While these equations form a nucleus for condition prediction of MDOT pavement

network, for project level analyses, a shift adjustment of the prediction should be made to match the current observation.

The feedback program developed in this study computes load index of original pavements of all types and overlaid flexible pavements. Load index

is the ratio of the actual ESAL sustained by the pavement and the design ESAL. Also included is a routine to verify/substantiate the prediction

models by comparing the actual to the predicted distresses


Format:
PDF
Funding:
State Study 119
Collection(s):
US Transportation Collection
Main Document Checksum:
[+]
urn:sha256:81f85222ecd11eca3000b3c3b2396b80a7f3007e890dd75494b041ebf5c1f3f6
Download URL:
https://rosap.ntl.bts.gov/view/dot/24129/dot_24129_DS1.pdf
File Type:

Supporting Files

• No Additional Files

More +