Lindley Type Distributions and Software Reliability Assessment
Qi Xiao, Tadashi Dohi, Hiroyuki Okamura · 2020 Asia-Pacific International Symposium on Advanced Reliability and Maintenance Modeling (APARM) · 2020
Dennis Victor Lindley proposed an interesting one- parameter continuous probability distribution, which is called Lindley distribution, in the context of Fiducial and Bayesian statistics. Recently, the Lindley distribution and its variations have received considerable attentions and have been applied to the life data analysis for several products in stead of the common exponential distribution. In this paper, we attempt to use the Lindley type distributions to describe the software fault- detection time distribution, and develop somewhat different nonhomogeneous Poisson process (NHPP)-based software reliability models (SRMs). The resulting SRMs are compared with the existing NHPP-based SRMs having the well-known fault-detection time distributions which can be categorized into the exponential family and the extreme type distributions. Throughout numerical examples with the fault count data observed in actual software development projects, we show the usefulness of the Lindley type SRMs.