A Hybrid Optimization Model to Increase the Accuracy of Software Development Effort Estimation

Accurate software development effort estimation is a critical part of software projects. During recent years, software development effort estimation has become a challenging issue for developers, managers, and customers. Uncertainty of software projects, complexity of production process, intensive role of human, and ambiguity of needs are some of the reasons of challenge. Effective development of software is based on accurate effort estimation. Analogy based estimation (ABE) is the most popular method in this field. This model can easily estimate the development effort by comparing new projects with previous ones. Despite its benefits, the ABE is unable to produce accurate estimations when the importance level of project feature is not the same, or finding a relation among them is difficult. In this situation, efficient feature weighing can be a solution to improve the performance of ABE. This paper proposes a new hybrid estimation model based on combination of an invasive weed optimization algorithm (IWO) and ABE to increase the accuracy of software development effort estimation. Indeed, the process of attribute weighting is adjusted so that the performance of ABE is improved. Two real data sets are utilized to evaluate the accuracy of the proposed hybrid model. The promising results show that a combination of IWO and ABE could significantly improve the performance of existing estimation models.