This paper describes a computational approach, SOFTSYS, recently developed at the University of Illinois for determining layer properties from nondestructive pavement testing and evaluation. Specifically, it is capable of backcalculating properties of pavement layers using results of the Falling Weight Deflectometer (FWD) test. SOFTSYS is based on a set of algorithms derived from the use of artificial neural networks and genetic algorithms in the field of soft computing. The backcalculation approach considers geomaterial nonlinearity that may seriously affect the overall response. SOFTSYS includes structural models developed for various types of pavements including full-depth asphalt pavements and fulldepth asphalt pavements built on lime stabilized subgrade soils. In this paper, the algorithm of SOFTSYS is introduced and described in detail by solving an example backcalculation problem that uses synthetic FWD data generated from finite element analysis of a full-depth asphalt pavement structure.