The steady rollout of Industrial IoT (IIoT) technology in the manufacturing domain embodies the potential to implement smarter and more resilient production processes. To this end, it is expected that there will be a strong reliance of manufacturing processes on cloud/edge services so as to act intelligently and flexibly. While automation is necessary to handle the environment's complexity, human-in-the-loop design approaches are paramount. In this context, Digital Twins play a crucial role by allowing human operators to inspect and monitor the environment to ensure stability and reliability. Integrating the IIoT with the Metaverse enhances the system's capabilities even further, offering new opportunities for efficiency and collaboration while enabling integrated management of assets and processes. This article presents a layered conceptual architecture as an enabler for smart manufacturing metaverse environments, targeting real-time data collection and representations from shopfloor assets and processes. At the bottom layer, our proposal relies on middleware technology, serving differentiated Quality of Service (QoS) needs of the Operation Technology (OT) monitoring processes. The latter contributes to feeding a virtual layer where data processes reside, creating representations of the monitored phenomena at different timescales. Metaverse applications can consume data by tapping into the metaverse engine, a microservice-oriented and accelerated Platform as a Service (PaaS) layer tasked with bringing data to life. Without loss of generality, we profile different facets of our proposal by relying on two different proof-of-concept inspection applications aimed at real-time monitoring of the network fabric activity and a visual asset monitoring one.

翻译：工业物联网（IIoT）技术在制造领域的稳步推广，体现了实现更智能、更具弹性生产流程的巨大潜力。为此，预计制造流程将高度依赖云/边缘服务，以实现智能化和灵活化运作。尽管自动化是应对环境复杂性的必要手段，但"人在回路"的设计方法至关重要。在此背景下，数字孪生技术通过允许人类操作员检查并监控环境以确保稳定性和可靠性，发挥着关键作用。将工业物联网与元宇宙相融合，可进一步增强系统能力，为效率和协作提供新机遇，同时实现对资产与流程的集成管理。本文提出一种分层概念架构，作为智能制造元宇宙环境的使能技术，旨在从车间资产和流程中实现实时数据采集与表示。在底层，我们的方案依托中间件技术，满足操作技术（OT）监控流程中差异化的服务质量（QoS）需求。该中间件为承载数据处理的虚拟层提供支撑，在不同时间尺度上创建被监控现象的表示。元宇宙应用可通过接入元宇宙引擎来消费数据，该引擎是一个面向微服务且经加速的平台即服务（PaaS）层，负责将数据转化为生动形态。在不失一般性的前提下，我们通过两个概念验证型检测应用来剖析方案的不同侧面：一个专注于网络架构活动的实时监控，另一个则聚焦于视觉资产监控。