Performance of Cost Assessment on Reusable Components for Software Development using Genetic Programming

Reusability is the quality of a piece of\udsoftware, which enables it to be used again, be it partial,\udmodified or complete. A wide range of modeling\udtechniques have been proposed and applied for software\udquality predictions. Complexity and size metrics have\udbeen used to predict the number of defects in software\udcomponents. Estimation of cost is important, during the\udprocess of software development. There are two main\udtypes of cost estimation approaches: algorithmic methods\udand non-algorithmic methods. In this work, using genetic\udprogramming which is a branch of evolutionary\udalgorithms, a new algorithmic method is presented for\udsoftware development cost estimation, using the\udimplementation of this method; new formulas were\udobtained for software development cost estimation in\udwhich reusability of components is given priority. After\udevaluation of these formulas, the mean and standard\uddeviation of the magnitude of relative error is better than\udrelated algorithmic methods such as COCOMO formulas.\u