Abstract Modern warehousing systems for fresh and cold-keeping storage, have presented characteristics of complex operation procedures, accelerated operating pace and high labour intensity. Thus, the working environment has become complicated and hazardous. Two recent fatal accidents occurred in cold warehouses have shifted the wide focus to safety management. The invisibility of operators’ status and location causes late responsiveness for rescuing. This paper first proposes an IoT and digital twin-enabled tracking solution framework for safety management. Then an indoor safety tracking mechanism for detecting motionless behaviour and self-learning genetic positioning is developed for recognizing the abnormal condition and obtaining precise location information in a real-time manner. A real-life case study with physical and cyber world implementation is conducted to demonstrate the feasibility and effectiveness of our proposed techniques. The results show that the detection of abnormal motionless behaviour is fulfilled, and the indoor positioning algorithm with self-learning ability not only achieves high accuracy up to 96.5% but also ensure the long-term use through adaptation.

中文翻译：

---

物联网和数字孪生为安全管理启用智能跟踪

摘要 现代保鲜冷藏仓储系统具有操作流程复杂、操作速度快、劳动强度高等特点。因此，工作环境变得复杂和危险。近期发生的两起冷库致命事故，将注意力转移到安全管理上。操作员状态和位置的不可见性导致救援响应迟缓。本文首先提出了一个用于安全管理的物联网和数字孪生跟踪解决方案框架。然后开发了一种用于检测静止行为和自学习基因定位的室内安全跟踪机制，以实时识别异常情况并获取精确的位置信息。进行了物理和网络世界实施的现实案例研究，以证明我们提出的技术的可行性和有效性。结果表明，实现了异常静止行为的检测，具有自学习能力的室内定位算法不仅达到了高达96.5%的高精度，而且通过适应保证了长期使用。