[MA] Enabling Test Case Selection and Prioritization for Cardinality-based Software Product Lines

Test case selection and prioritization is an important challenge in search-based testing, whose goal is to find a minimal subsets of test cases to execute, whilst maximizing test quality. It is especially vivid in context of highly-configurable software systems, such as software product lines (SPLs) and their derivations, whose configuration spaces cannot be tested exhaustively. Moreover, if an SPL allows cardinalities, i.e., selecting the same feature multiple times, an additional level of complexity appears.

Traditional approaches for search-based testing of SPLs focus either on boolean feature models, or are limited to one-wise sampling, leaving testing of feature interactions out of scope. The goal of this thesis is to close this gap and to enable efficient test case selection and prioritization of cardinality-based SPLs. As a case study, it should use BRISE-MPL, a multiple product line for expensive blackbox optimization, developed at the chair of software technology.

To reach this goal the following tasks need to be accomplished.

• Perform a literature analysis on test case selection and prioritization.
• Analyze BRISE-MPL and its variability model. Formulate the problem of test case selection and prioritization for this system.
• Analyze state-of-the-art approaches for solving test case selection and prioritization and their capabilities in context of cardinality-based SPLs.
• Propose a new approach for performing test case selection and prioritization, which is able to handle BRISE-MPL.
• Implement the proposed approach and integrate it into BRISE-MPL.
• Conduct an empirical evaluation of the approach, investigating its efficiency and performance.

Betreuer: Dmytro Pukhkaiev