Multi-Objective Test Suite Optimization for Event-B Models

Abstract

Event-B is a formalism that is used in modeling and proving the consistency of complex systems. The method has been successfully used in the development of several complex real-life applications. For complementing the theorem-proving and model-checking tools developed for Event-B models, test suite generation methods have been recently introduced as research theme. In order to optimize the large test suites produced by the existing approaches, in this paper the test suite optimization problem is introduced for Event-B models. However, there exist many optimization criteria in real-life testing problems. Given that, six specifically multi-objective test suite optimization problems are defined. Two modern Multi-Objective Evolutionary Algorithms are used for solving them: NSGA-II and SPEA-2. The experiments have been conducted using five test suites generated from two industrial inspired Event-B models (five different machines).