Software testing optimization for large systems using agent-based and NSGA-II algorithms

The multiobjective optimization problem is addressed in this article using a novel evolutionary technique to find a global solution in the Pareto form. The proposed work is innovative because it applies an evolutionary multi-agent system (EMAS) and NSGA-II from various traditional evolutionary methods. The evolution process in NSGA-II and EMAS enables thorough exploration of search space, and the employed crowdsourcing mechanism facilitate the accurate approximation of the entire Pareto frontier. The technique is explained in this article, and report the initiatory experimental findings. The product line or large configurable system needs to set specifications, architecture, reusable components, and shared products to develop the features of new products. To maintain high quality, a thorough testing process is required. Testing is necessary for each product of the large system, each of which has a varied set of features. Consequently, a multi-objective optimization technique can be used to optimize the large system testing process. The performance of a multi-objective Non-Dominated Sorting Genetic Algorithm II (NSGA-II) and evolutionary multi-agent system (EMAS) on Feature Models (FMs) to enhance large System testing is reported in this study