Grey-based approach for estimating software reliability under nonhomogeneous Poisson process

Due to the randomness and time dependence of factors affecting software reliability, most reliability models are treated as stochastic processes, non-homogeneous Poisson process (NHPP) is used one. However, failure behavior does not follow NHPP in a statistically rigorous manner, pure random method might be enough describe behavior. To solve these problems, this paper proposes new integrated approach that combines grey system theory software. A model put forward discrete form, grey-based for estimating under proposed nonlinear multi-objective programming problem. Finally, four Reliability applied real datasets, dynamic R-square predictive relative error calculated. Comparing with original single model, it found modeling using has higher prediction accuracy reliability. Therefore, there characteristics uncertain information models, exploiting latent lead more accurate estimation.