G2WAY: A PAIRWISE TEST DATA GENERATION STRATEGY WITH AUTOMATED EXECUTION SUPPORT

Abstract

This paper discusses a new strategy, called G2Way, for pairwise test data generation. Unlike existing strategies, G2Way also supports (concurrent) automated execution integrated within the strategy itself. Furthermore, empirical evidence demonstrates that G2Way, in some cases, outperformed existing strategies in terms of the number of generated test data (as compared to that of AETG and its variations, IPO, SA, GA, ACA, and All Pairs). Notwithstanding the differences in the computing environment employed as well as the overhead incurred to permit automated execution, the test generation time is also within reasonable value.