NHPP Based Software Reliability Growth Models
Stochastic processes are used for the description of a system’s operation over time. There are two main types of stochastic processes: continuous and discrete. Among discrete processes, counting processes in reliability engineering are widely used to describe the appearance of events in time (e.g., failures, number of perfect repairs, etc). The simplest counting process is a Poisson process. The Poisson process plays a special role to many applications in reliability engineering.
As a general class of well-developed stochastic process model in reliability engineering, NHPP models have been successfully used in studying hardware reliability problems. They are especially useful to describe failure processes which possess certain trends such as reliability growth and deterioration. Therefore, an application of NHPP models to software reliability analysis is then easily implemented. The model provides the expected number of faults/failures at a given time.
Schneidewind [1975] proposed an error detection model. Goel & Okumoto [1979] proposed the time dependent failure rate model. Ohba [1984] proposed the inflection S-shaped model. Musa [1975] and Musa & Okumoto [1984] proposed the basic execution time model and Log Poisson model respectively. Yamada, Ohba & Osaki [1983] proposed a model based on the concept of failure observation and the corresponding fault removal phenomenon. Goel [1985] modified his original model by introducing the test quality parameters. Littlewood [1981] proposed a modification of Dunae model based on NHPP. Yamada, Osaki & Narihisa [1985] proposed a model with two types of faults. Yamada & Osaki [1985] also proposed two classes of discrete time models. One class describes an error detection process in which the expected number of errors detected per test case is geometrically