In this paper, a new software availability/ reliability model is developed where lifetimes and repair times have general system-state-dependent distributions. Multiple errors may be introduced or removed through repairs. The model is formulated as a multivariate Markov process and contains many other models which have appeared in the literature as special cases. The exponentiality assumption prevalent in the literature is totally eliminated. Expressions of various performance measures of practical interest combining availability and reliability of the software system at time t are derived. Using the matrix Laguerre transform of Sumita , corresponding computational procedures are also developed. A numerical example is given, demonstrating speed, accuracy and stability of these procedures.
ASJC Scopus subject areas