Incremental Nonlinear Model for Predicting Pavement Serviceability Jorge A. Prozzi 1 and Samer M. Madanat 2 Abstract: A recursive nonlinear model was developed for the prediction of pavement performance as a function of traffic characteristics, pavement structural properties, and environmental conditions. The model highlights some of the advantages of relaxing the linear restriction that is usually placed on the specification form of pavement performance models. First, a functional form that better represents the physical deterioration process can be used. Second, the estimated parameters are unbiased, owing to a proper specification and the use of sound statistical techniques. Finally, the standard error of the prediction is reduced by half that of the equivalent existing linear model. This improved accuracy has important economic implications in the context of pavement management. The model developed as part of this research enables the determination of an unbiased exponent of the so-called power law and of the equivalent loads for different axle configurations. The estimated exponent confirms the value of 4.2 traditionally used. However, it should be noted that this exponent is only to be used for determining damage in terms of serviceability. On the other hand, equivalent loads estimated for different axle configura- tions tend to differ from traditionally used values. DOI: 10.1061/ASCE0733-947X2003129:6635 CE Database subject headings: Pavements; Serviceability; Models; Cost allocations. Introduction The objective of this research was the development of sound flex- ible pavement performance models to be used primarily for the management of the road infrastructure because accurate predic- tion of pavement performance is important for effective manage- ment of the infrastructure. Owing to the great complexity of the road deterioration pro- cess, performance models are, at best, only approximate predic- tors of expected conditions. Models that only produce a determin- istic prediction of performance without any quantification of the accuracy of the prediction are unrealistic; hence, an estimation of the prediction error is essential. The absence of such an estimate imposes important limitations on the applicability of the perfor- mance prediction model. The Present Serviceability Index PSIwas developed in the early 1960s and constituted the first comprehensive effort to es- tablish performance standards based upon considerations of riding quality Carey and Irick 1960; Highway Research Board HRB 1962. A panel of highway users from different backgrounds evaluated seventy-four flexible pavement sections and rated them on a discrete five-point scale 0 for poor, 5 for excellent. This experiment gave origin to the Present Serviceability Rating PSR. The PSR was found to correlate highly with surface roughness and, to a lesser extent, with rutting, cracking, and patching. Models for the prediction of riding quality are very important to highway agencies for the purpose of managing their road net- work. The prediction of riding quality is also important for road pricing and regulation studies. Both the rate of riding quality de- terioration over time and the contribution of the various factors to such deterioration are important inputs to these studies. Useful models are those that establish the contribution of pavement structure, traffic, environment and any other factors that are rel- evant for cost allocation. Data Considerations There are a number of possible data sources that have been used for the development of pavement performance models. Some of these sources are: 1randomly selected in-service pavement sec- tions; 2in-service pavement sections selected according to an experimental design; 3purposely built pavement test sections subjected to the action of actual traffic and the environment; and 4purposely built pavement test sections subjected to the accel- erated action of traffic and environmental conditions. Data originating from in-service pavement sections subjected to the combined actions of highway traffic and environmental conditions are those that most closely represent the actual dete- rioration process of pavements in the field. All other data sources would produce models that would suffer from some kind of bias or restrictions unless special considerations are taken into account during the parameter estimation process. Some of these consider- ations are briefly described in the following paragraphs. The most common problems encountered in models developed from randomly selected in-service pavement sections are caused by unobserved events typical of such data sets, and by the prob- 1 Assistant Professor, Dept. of Civil Engineering, The Univ. of Texas, ECJ 6.8, Austin, TX 78712. corresponding author. E-mail: prozzi@mail.utexas.edu 2 Professor, Dept. of Civil and Environmental Engineering, Univ. of California, 114 McLaughlin, Berkeley, CA 94720. E-mail: madanat@ce.berkeley.edu Note. Discussion open until April 1, 2004. Separate discussions must be submitted for individual papers. To extend the closing date by one month, a written request must be filed with the ASCE Managing Editor. The manuscript for this paper was submitted for review and possible publication on November 26, 2001; approved on December 20, 2002. This paper is part of the Journal of Transportation Engineering, Vol. 129, No. 6, November 1, 2003. ©ASCE, ISSN 0733-947X/2003/6- 635– 641/$18.00. JOURNAL OF TRANSPORTATION ENGINEERING © ASCE / NOVEMBER/DECEMBER 2003 / 635