10.6 Metrics Reliability Estimation

After the testing is done and the software is delivered, the development is considered over. It will clearly be very desirable to know, in quantifiable terms, the reliability of the software being delivered. As testing directly impacts the reliability and most reliability models use data obtained during testing to predict reliability, reliability estimation is the main product metrics of interest at the end of the testing phase. We will focus our attention on this metric in this section.

Before we discuss the reliability modeling and estimation let us briefly discuss a few main metrics that can be used for process evaluation at the end of the project.

Once the project is finished, one can look at the overall productivity achieved by the programmers during the project. As discussed earlier, productivity can be measured as lines of code (or function points) per person-month.

Another process metric of interest is defect removal efficiency. The defect removal efficiency of a defect removing process is defined as the percentage reduction of the defects that are present before the start of the process [104]. The cumulative defect removal efficiency of a series of defect removal processes is the percentage of defects that have been removed by this series. The defect removal efficiency cannot be determined exactly as the defects remaining in the system are not known. However, at the end of testing, as most defects have been uncovered, removal efficiencies can be estimated.

Let us now return...