Title: Cost estimation of software development: improving the COCOMO model using a genetic algorithm approach

Authors: Jafar Razmi, Reza Ghodsi, Marzieh Jokar

Addresses: Department of Industrial Engineering, College of Engineering, University of Tehran, Tehran, Iran. ' Department of Industrial Engineering, College of Engineering, University of Tehran, Tehran, Iran. ' Department of Industrial Engineering, College of Engineering, University of Tehran, Tehran, Iran

Abstract: The use of computers and Information Technology (IT) solutions is a vital necessity for enterprises, which requires continuously increasing investments in hardware and software applications. Prior to any software development project estimation of cost has to be carried out. Because of the complex nature of software applications, it is often difficult to predict the cost of software development accurately. Recently, various methods have been proposed by researchers to predict the effort of software projects and estimate the cost accordingly. In this study, first a discussion on the major available models for software cost estimation along with their strengths and weaknesses is presented. Next, using Genetic Algorithms (GAs), three new models are introduced in order to estimate the cost of software development projects. The performances of these three models are tested using real data. The results show that the proposed models are able to provide better estimates in comparison to previously discussed models.

Keywords: SCE; software cost estimation; COCOMO; constructive cost models; genetic algorithms; GAs; software development projects; cost prediction; information technology.

DOI: 10.1504/IJMP.2009.026961

International Journal of Management Practice, 2009 Vol.3 No.4, pp.346 - 368

Published online: 11 Jul 2009 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article