A Method Using Generative Adversarial Networks for Robustness Optimization

Abstract

The evaluation of robustness is an important goal within simulation-based analysis, especially in production and logistics systems. Robustness refers to setting controllable factors of a system in such a way that variance in the uncontrollable factors (noise) has minimal effect on a given output. In this paper, we present an approach for optimizing robustness based on deep generative models, a special method of deep learning. We propose a method consisting of two Generative Adversarial Networks (GANs) to generate optimized experiment plans for the decision factors and the noise factors in a competitive, turn-based game. In a case study, the proposed method is tested and compared to traditional methods for robustness analysis including Taguchi method and Response Surface Method.