Reviewing and Proposing Approaches for Self-adaptive Testing in the Field

Software systems are an essential part of our everyday life, and over time, they are becoming larger and more complex. These characteristics make systems even more susceptible to faults. For this reason, there is an increasing need to prolong testing beyond software release, even while the system is operating in the field. This is called testing in the field. One strategy to handle the challenges associated with testing in the field, such as uncertainty, is to employ self-adaptive testing. The self-adaptation of the testing process guarantees that the system is functioning properly, even after the performance of its own adaptation. In this research proposal, we aim at proposing a novel approach for self-adaptive testing in the field. We first review the literature and propose a definition, a taxonomy, and a reference architecture for this type of approach. We also highlight the main challenges faced by these approaches. Finally, we also propose a strategy that combines Combinatorial Testing and Markov Chains to perform Self-Adaptive Testing in the Field.