Application of Artificial Neural Network for Software Reliability Growth Modeling with Testing Effort

Background/Objectives: To design a relatively simple Software Reliability Growth Model (SRGM) with testing effort function using Artificial Neural Network approach. Methods/Statistical Analysis: The results evaluation of the proposed SRGM using Artificial Neural Network (ANN) is measured by calculating the three vital criterians namely; AIC (Akaike Information Criterian), R2 (Coefficient of determination) and RMSE (Root Mean Squared Error). Findings: Traditional time-based models may not be appropriate in some situations where the effort is varying with time. Estimating the total effort required for testing the software in the Software Development Life Cycle (SDLC) is important. Hence, a multi-layer feed-forward Artificial Neural Network (ANN) based SRGM using back propagation training is proposed in this paper by incorporating test effort. The proposed ANN based model provides consistent performance for both exponential and S-shaped growth of mean value functions witnessed in software projects. Application/Improvements: The proposed SRGM using ANN will be performed to be eminently useful for software reliability applications, since it is able to maintain its performance in all situation.