An Analysis of Fuzzy Approaches for COCOMO II

Abstract

Software cost estimation is one of the mostchallenging task in project management. However, theprocess of estimation is uncertain in nature as itlargely depends upon some attributes that are quiteunclear during the early stages of development. In thispaper a soft computing technique is explored toovercome the uncertainty and imprecision in estimation.The main objective of this research is to investigatethe role of fuzzy logic technique in improving theeffort estimation accuracy using COCOMO II bycharacterizing inputs parameters using Gaussian,trapezoidal and triangular membership functions andcomparing their results. NASA (93) dataset is used inthe evaluation of the proposed Fuzzy Logic COCOMO II.After analyzing the results it had been found thateffort estimation using Gaussian member function yieldsbetter results for maximum criterions when comparedwith the other methods.