Software Reliability Modeling and Assessment Integrating Time Dependent Fault Reduction Factor in Random Environment

Abstract

Growing demand of software in all application domains have led to the rising expectations and requirement for more reliable software systems from user end. Paradoxically, while achieving the reliability goals, the complexity of software turn to be very high and consequently it becomes critical to have influential approaches which evaluate reliability measures accurately. Based on distinct set of assumptions, a very large number of software reliability growth models (SRGMs) have already been developed over past few decades and still ongoing to evaluate various reliability metrics. In this chapter, we derive a software reliability model with the key consideration that the operating environment of software is unalike from the controlled testing environment and is accountable to affect software execution and its reliability significantly. To deal with randomness of operating environment and variations of fault detection rate subject to time we consider time dependent fault reduction factor in random environment. In addition, to suggest release time of the software, cost and reliability criteria are discussed and illustrated with numerical example. To conduct the comprehensive evaluation of goodness of fit, we worked out several selection criteria and comparative analysis with the existing models and it is worth noting that results offered by the proposed model are dependable and highly consistent with the observations procured from the real life data sets.