Effect of Objective Function on the Optimization of Highway Vertical Alignment by Means of Metaheuristic Algorithms

Abstract

The main purpose of this work is the comparison of several objective functions for optimization of the vertical alignment. To this end, after formulation of optimum vertical alignment problem based on different constraints, the objective function was considered as four forms including: 1) the sum of the absolute value of variance between the vertical alignment and the existing ground; 2) the sum of the absolute value of variance between the vertical alignment and the existing ground based on the diverse weights for cuts and fills; 3) the sum of cut and fill volumes; and 4) the earthwork cost and then the value of objective function was compared for the first three cases with the last one, which was the most accurate ones. In order to optimize the raised problem, Genetic Algorithm (GA) and Group Search Optimization (GSO) were implemented and performance of these two optimization algorithms were also compared. This research proves that the minimization of sum of the absolute value of variance between the vertical alignment and the existing ground, which is commonly used for design of vertical alignment, can’t at all grantee the optimum vertical alignment in terms of earthwork cost.