ABC Tester - Artificial Bee Colony Based Software Test Suite Optimization Approach

In this paper we present a new, non-pheromone-based test suite optimizationapproach inspired by the behavior of biological bees. Our proposed approachis based on ABC (Artificial Bee Colony Optimization) which is motivated bythe intelligent behavior of honey bees. In our proposed system, the sites arethe nodes in the Software under Test (SUT), the artificial bees modify the testcases with time and the bee’s aim is to discover the places of nodes with highercoverage and finally the one with the highest usage by the given test case. SinceABC system combines local search methods carried out by employed beeswith global search methods managed by onlookers and scouts, we attain nearglobal optima. We investigate whether this new approach outperforms existingtest optimization approach based on Genetic Algorithms (GA) in the task ofsoftware test optimization. Taking into account the results of our experiments,we conclude that (i) the proposed approach uses fewer iterations to completethe task; (ii) is more scalable, i.e., it requires less computation time to completethe task, and finally (iii) our approach is best in achieving near global optimalsolution.