An Imperfect SRGM based on NHPP

Abstract

SRGMs (Software Reliability Growth Models) are the probabilistic and statistical models widely used to evaluate software reliability. They explain how the reliability of the software varies over time by investigating failure data as the testing progresses. The paper presents a new SRGM in an imperfect debugging environment. The proposed model considers that with time, the fault detection rate will increase gradually. Also, the new model is tested with a real dataset that is available in public domain and compared the results with seven existing models in terms of four goodness-of-fit criteria.