Recently, manufacturing concepts, such as personalized production and distributed manufacturing, have attracted attention owing to the ongoing revolution in industrial technology. Connected micro smart factories in factory-as-a-service system with these new manufacturing paradigms and Industrial Internet of Things (IIoT) are inefficient in terms of cost and production. To solve these problems, a digital twin, which uses a digital representation of a process, with the same configuration of manufacturing elements, synchronized information, and functional units, was designed and implemented. The digital twin utilizes the latest information from the Internet to gather data from IIoT devices and interoperates in a variety of applications. In addition, it derives the components of a detailed design of the digital twin application, to which it performs procedure definition. This research differs from other digital twin studies that concentrate on the prognostic health management of only a single machine. This study could help managers organize the benefits of utilization through a digital twin based on a hierarchy as they could receive real-time monitoring of the present, tracking information from the past, and operational decision-making support for the future. In addition, the proposed application reduces the cost and production inefficiencies, ultimately resulting in the efficient operation of a manufacturing system.

Personalised production allows the supply chain (SC) to exist in various dynamic fluctuations within a make-to-order (MTO) environment. An SC for personalised production has redundant inventory and operation capacity; therefore, it requires a system that can achieve recoverability for operation resilience. Thus, a standalone cyber physical system (CPS) has limitation for SC control with MTO. To solve this problem, the CPS must be coordinated, and a systematic approach is required. This study proposes a cyber physical logistics system (CPLS) that is coordinated with the agent cyber physical production systems in a multi-level CPS structure. This multi-level architectural framework is designed to provide technical functionalities for resilient SC control. The service composition procedures of technical functionalities on the distributed digital twin (DT) simulation are divided into type and instance stages. The operation procedures of the DT application and technical functionality of the DT engine are suggested. The proposed CPLS has appropriate service composition and operation for the bullwhip and ripple effects, which are the two main SC-related problems. This study illustrates an early case of CPLS that can minimise differences among assets using distributed DT simulation; furthermore, the study establishes an SC and production plan based on the DT simulation results.

To achieve efficient personalized production at an affordable cost, a modular manufacturing system (MMS) can be utilized. MMS enables restructuring of its configuration to accommodate product changes and is thus an efficient solution to reduce the costs involved in personalized production. A micro smart factory (MSF) is an MMS with heterogeneous production processes to enable personalized production. Similar to MMS, MSF also enables the restructuring of production configuration; additionally, it comprises cyber-physical production systems (CPPSs) that help achieve resilience. However, MSFs need to overcome performance hurdles with respect to production control. Therefore, this paper proposes a digital twin (DT) and reinforcement learning (RL)-based production control method. This method replaces the existing dispatching rule in the type and instance phases of the MSF. In this method, the RL policy network is learned and evaluated by coordination between DT and RL. The DT provides virtual event logs that include states, actions, and rewards to support learning. These virtual event logs are returned based on vertical integration with the MSF. As a result, the proposed method provides a resilient solution to the CPPS architectural framework and achieves appropriate actions to the dynamic situation of MSF. Additionally, applying DT with RL helps decide what-next/where-next in the production cycle. Moreover, the proposed concept can be extended to various manufacturing domains because the priority rule concept is frequently applied.