Real-time data-driven synchronous reconfiguration of human-centric smart assembly cell line under graduation intelligent manufacturing system

Abstract

Manufacturing systems are constantly required to leverage emerging technologies to increase flexibility and responsiveness to meet the diversified demands in an ever-changing market cost-efficiently. Assembly cell line (ACL), derived from the Toyota production system, is firstly sprouted in some leading electronics companies. In this unique layout, the conveyor line is divided into human-centric assembly cells (ACs) and each AC stores a certain number of parts that allow the multi-skilled worker to easily access and complete an entire product assembly on his or her own pace. Therefore, the proper production capacity of ACL can be achieved through frequent changeover and reconfiguration to meet the changing demands, which is an extension of the just-in-time strategy for organization systems (JIT-OS). However, due to the lack of real-time data on the up-to-date capacity of ACs and an efficient coordination mechanism of ACL based on real-time information feedback in dynamic and stochastic manufacturing environments, it is difficult to successfully implement JIT-OS to rapidly respond to the changing demands. This research proposes a real-time data-driven synchronous reconfiguration of smart ACL workshops (Sync-RAS) under the Graduation Intelligent Manufacturing System (GiMS) by leveraging Industry 4.0 (I4.0) enabling technologies. Computer vision (CV) is employed to monitor and analyze the real-time operation process within the human-centric AC. In addition, an appropriate information-sharing architecture with scalability and reconfigurability is developed to coordinate the smooth reconfiguration of the smart ACL workshop in rapid response to the changing demands. Based on these, GiMS is introduced to achieve simple and feasible operational tactics through real-time data-driven information visibility, traceability, and sharing, while a Sync-RAS mechanism is proposed to optimize the reconfiguration of ACL. Finally, a real-life case has been performed to verify and show the benefits of the proposed approach.