Efficient approach for start-up transient process optimisation of cluster tools with equipment front-end module

In semiconductor manufacturing, cluster tools with equipment front-end modules (EFEM) play a critical role in wafer fabrication. The rapid growth of AI has sharply increased the demand for high-performance computing chips. As a result, the rapid increase in wafer processing lots has led to frequent switches between steady state and transient processes in cluster tools, which reduces tool efficiency and makes start-up process scheduling increasingly important. This work aims to solve the start-up process scheduling problem of cluster tools with EFEM. Such a cluster tool consists of three modules: a vacuum module (VM) for wafer processing, a loadlock module (LLM), and an EFEM under the atmospheric environment. Each of VM and EFEM has a robot for wafer transportation. In operating a cluster tool with EFEM, robots should closely coordinate with LLM to successfully transport wafers between VM and EFEM, and realise a smooth start-up process. This complicates the scheduling problem. To address these issues, operational strategies of the robots and LLM are proposed such that LLM is well coordinated with the robots. Based on these operational strategies, a generic linear programme is developed to optimise the start-up process. Finally, the effectiveness of the proposed method is verified through experiments.