A successful pavement management system requires an accurate pavement performance prediction model. A novel pavement performance model using the piecewise approximation approach was developed to estimate the pavement serviceable life. It can be broadly applied to estimate pavement performance of any distress types or indexes. The basic theory of the piecewise approximation is to divide the whole pavement serviceable life into three zones: Zone 1 for early age pavement distress, Zone 2 in rehabilitation stage, and Zone 3 for overdistressed situations. Historical pavement performance data are regressed independently in each time zone. This approach can accurately predict pavement distress progression trends in each individual zone by eliminating possible impacts from biased data in other zones. This paper describes the theoretical piecewise approximation process of data classification and model regression and then demonstrates an implementation for a group of Washington State Department of Transportation asphalt concrete pavements. The results are compared with the Mechanistic–Empirical Pavement Design Guide incremental damage approach, the current Washington State Pavement Management System (WSPMS) exponential model, and ordinary regression on all data points. Results indicate that the proposed approach is able to estimate the most accurate rehabilitation due year and to predict the performance trends for each divided zone. The piecewise approximation approach is planned for implementation into the WSPMS and will play an important role in decision making for future pavement rehabilitations.
[1]
Samer Madanat,et al.
Adaptive Optimization of Infrastructure Maintenance and Inspection Decisions under Performance Model Uncertainty
,
2002
.
[2]
Linda M Pierce,et al.
Calibration of Flexible Pavement in Mechanistic–Empirical Pavement Design Guide for Washington State
,
2009
.
[3]
Jorge A Prozzi,et al.
Estimation of Pavement Performance Deterioration Using Bayesian Approach
,
2006
.
[4]
James C. Bezdek,et al.
Pattern Recognition with Fuzzy Objective Function Algorithms
,
1981,
Advanced Applications in Pattern Recognition.
[5]
Samuel H Carpenter,et al.
PAVEMENT PERFORMANCE PREDICTION MODEL USING THE MARKOV PROCESS
,
1987
.
[6]
R L Lytton.
Concepts of pavement performance prediction and modelling
,
1987
.
[7]
Linda M Pierce,et al.
Washington State's Web-Enabled Pavement-Management System
,
2004
.
[8]
Samer Madanat,et al.
Development of Pavement Performance Models by Combining Experimental and Field Data
,
2004
.
[9]
Gordon A. Sparks,et al.
Development of Bayesian Regression Model to Predict Hot-Mix Asphalt Concrete Overlay Roughness
,
1996
.
[10]
Linda M Pierce,et al.
Calibration of NCHRP 1-37A software for the Washington state department of transportation : Rigid pavement portion
,
2006
.