Software developed for mission-critical systems employ rigorous verification and validation techniques to achieve good quality. Traditional software reliability models use execution time during software testing for reliability estimation. Although testing duration is an important factor in reliability, it is possible to improve the accuracy of estimation by factoring other techniques that influence the final software quality. In this paper, we propose a novel method to utilize the effectiveness of the verification and validation process to predict reliability. The main idea is that failure detection is not only related to the time that the software experiences under testing, but also to different levels of review rigors the software has undergone. Our experimental results with multi-version software used in mission-critical systems show our model achieves substantial reliability estimate.

Software Reliability Growth as an Offshoot of Verification and Validation Process

Static analysis as a part of software reliability assessment in general aims in the detection of faults and anomalies, in the preparation of tests and in the collection of metrics for the qualitative and quantitative description of software attributes. It is shown that in addition to these objectives the demonstration of compliance between functions which are implemented in a program and those which are specified tin the software requirements specification) can be achieved by means of static analysis.

Manfred Kersken

Papers

The Role of Static Analysis in Software Reliability Assessment