AdapTSoS: adaptive testing of systems of systems

Testing of System of Systems (SoS) presents many challenges due to the dynamic and continuous evolution of its Constituent Systems (CSs). Existing SoS testing strategies support the definition of test cases but fail to address the dynamic changes of CSs. New test cases need to be added or existing ones must be updated to cope with the variable functionalities of CSs. Adaptive testing is a strategy to dynamically adapt a set of test configurations or test cases to changes in the system under test over time. The main idea of this paper is to provide a new adaptive testing strategy, AdapTSoS, for the adaptive testing of SoSs in order to deal with the continuous changes of variable functionalities of the CSs. The proposed strategy leverages a variability model representing the variable functionalities of the CSs and a Discrete Time Markov Chain simulating the SoS changes over time, to then provide a set of adaptive test configurations to be executed on the SoS. We illustrate the application of AdapTSoS to an existing SoS example and discuss the benefits provided by the proposed approach.