On the variability of fuzzy debugging

Abstract

The development and maintenance of software reliability enjoys a high priority in the evolution of modern computer technology. Over the last three decades or so, a tremendous amount of research has gone into the theory and practice of producing reliable software. Given the many subjective factors that influence software reliability, it is surprising that very little of this research has been directed towards the application of fuzzy set ideas to problems in the area. In this paper, we extend the research began in Zeephongsekul and Xia (Fuzzy Sets and Systems 83 (1996) 233–247) where fuzzy debugging was introduced to looking at the variability associated with this mode of debugging. A measure of variability is introduced which incorporates both the imprecision of fuzzy debugging and the randomness due to the number of failures encountered. This measure generalizes the variance introduced by Körner (Fuzzy Sets and Systems 92 (1997) 83–93) to stochastic point processes consisting of random points and fuzzy marks. The results are useful for a proper assessment of the quality of the debugging process. An example illustrating the theory is provided.