Generative AI for Resilient Design of Manufacturing Processes

Abstract

Manufacturing systems often face unexpected disruptions such as machine failures or material shortages, which can severely impact the production performance. Traditional methods for addressing these disruptions tend to be time-consuming and resource-intensive, which cannot effectively maintain the resilience of manufacturing systems. Despite recent advances in digital twins (DT) and artificial intelligence (AI), very little has been done to mitigate high computational demands and generate resilient designs of process flows under uncertainty. Therefore, this paper presents a new generative AI (G-AI) approach for the on-the-fly designs of manufacturing systems in response to production disruptions, integrating digital twin models with neural networks to optimize process flows and increase system resilience. First, we propose a novel Generative Adversarial Network for system design (D-GAN) to generate diverse, adaptive system designs that align production performance with target key performance indicators (KPIs). Second, DT models are coupled with statistical metamodeling to optimize the sequential probability of design improvements. Experimental results show the high potential of new G-AI approaches to generate cost-effective system designs and enhance manufacturing resilience.