A Detailed Study of NHPP Software Reliability Models (Invited Paper)

Abstract—Software reliability deals with the probability that software will not cause the failure of a system for a specified time under a specified condition. The probability is a function of the inputs to and use of the system as well as a function of the existing faults in the software. The inputs to the system determine whether existing faults, if any, are encountered. Software Reliability Models (SRMs) provide a yardstick to predict future failure behavior from known or assumed characteristics of the software, such as past failure data. Different types of SRMs are used for different phases of the software development life-cycle. With the increasing demand to deliver quality software, software development organizations need to manage quality achievement and assessment. While testing a piece of software, it is often assumed that the correction of errors does not introduce any new errors and the reliability of the software increases as bugs are uncovered and then fixed. The models used during the testing phase are called Software Reliability Growth Models (SRGM). Unfortunately, in industrial practice, it is difficult to decide the time for software release. An important step towards remediation of this problem lies in the ability to manage the testing resources efficiently and affordably. This paper presents a detailed study of existing SRMs based on Non-Homogeneous Poisson Process (NHPP), which claim to improve software quality through effective detection of software faults