Abstract
This article examines the assessment of behavior of a search engine optimization algorithm when it is directly integrated into a simulation model of a complex dynamic system and when the search direction is selected over relatively short sample trajectories of the simulation process. In this problem formulation, evaluating the optimization process effectiveness requires constructing probabilistic estimates of the nonstationary simulation process since it has no time to reach a steady state. Therefore, this paper derives analytical expressions for the trend of mean integral estimates of nonstationary processes defined by the simulation model and its variance as a function of the estimation interval. The resulting relationships provide the basis for estimating the convergence rate of search engine optimization.

This publication has 3 references indexed in Scilit: