An imperfect-debugging model with learning-factor based fault-detection rate

• Main Authors: Iqbal, Javaid, Quadri, S.M.K., Ahmad, N.
• Format: Conference Proceeding
• Language: English
• Subjects: Computational modeling; Debugging; Fault detection; Fault Detection Rate (FDR) function; Imperfect debugging; Learning effect; Non-Homogeneous Poisson Process (NHPP); Software; Software reliability; Software Reliability Growth Model (SRGM); Testing
• DOI: 10.1109/IndiaCom.2014.6828164

Learning and fault detection rate functions have been vigorously studied and analyzed in the modeling of software reliability growth functions. The behavior of learning and fault detection rate functions is being studied either under static assumptions or under dynamic assumptions where it can be affected by many factors, e.g., imperfect debugging, resource allocations etc. Thus, some software reliability growth functions/models model a fault detection rate function as a constant term and others take a variable (increasing) fault detection rate function. An S-shaped rate function meant to capture the learning patterns during the software testing/debugging process is being extensively employed to model a variable (increasing) fault detection rate function. The ultimate aim is to capture the realistic behavior of learning and fault detection rate functions. In this paper, we propose an NHPP based imperfect-debugging software reliability growth model with learning-factor based fault detection rate function by incorporating a learning-factor based fault detection rate function obtained from Chiu and Huang's learning model