Software Development Effort Estimation Using Soft Computing

Abstract

Software development effort estimation is a daunting task that is being carried out by software developers as not much of the information about the software which is to be developed is available during the early stages of development. The information that is to be gathered for various attributes of software needs to be subjective which otherwise leads to imprecision and uncertainty. Inaccurate estimation of the software effort and schedule leads to financial loses and also delays in project deadline. In this paper, we present the use of soft computing technique to build a suitable model which improves the process of effort estimation. To do so, various parameters of Constructive Cost Model (COCOMO) II are fuzzified that leads to reliable and accurate estimates of effort. The results show that the value of Magnitude of Relative Error (MRE) obtained by applying fuzzy logic is quite lower than MRE obtained from algorithmic model. By analyzing the results further it is observed that Gaussian Membership Function (gaussmf) performs better than Triangular Membership Function (trimf) and Trapezoidal Membership Function (trapmf) as the transition from one interval to another is quite smoother. Here varying number of COCOMO II inputs are fuzzified with these membership functions. The validation of the experiment is carried on COCOMO public dataset.