Optimizing Effort and Time Parameters of COCOMO II Estimation using Fuzzy Multi-objective PSO

Abstract

The estimation of software effort is an essential and crucial activity for the software development life cycle. Software effort estimation is a challenge that often appears on the project of making a software. A poor estimate will produce result in a worse project management. Various software cost estimation model has been introduced to resolve this problem. Constructive Cost Model II (COCOMO II Model) create large extent most considerable and broadly used as model for cost estimation. To estimate the effort and the development time of a software project, COCOMO II model uses cost drivers, scale factors and line of code. However, the model is still lacking in terms of accuracy both in effort and development time estimation. In this study, we do investigate the influence of components and attributes to achieve new better accuracy improvement on COCOMO II model. And we introduced the use of Gaussian Membership Function (GMF) Fuzzy Logic and Multi-Objective Particle Swarm Optimization method (MOPSO) algorithms in calibrating and optimizing the COCOMO II model parameters. The proposed method is applied on Nasa93 dataset. The experiment result of proposed method able to reduce error down to 11.891% and 8.082% from the perspective of COCOMO II model. The method has achieved better results than those of previous researches and deals proficient with inexplicit data input and further improve reliability of the estimation method.

Improved Effort and Cost Estimation Model Using Artificial Neural Networks and Taguchi Method with Different Activation Functions

“…The essential part of using artificial intelligence (AI) in software effort estimation is the construction of the Artificial Neural Network (ANN) architecture [ 11 ]. ANNs have learning adaptability and are good at modeling complex nonlinear relationships.…”