人工智能在企业安全生产管理工程中的应用综述——从视觉感知到大模型智能决策的演进

doi:10.12677/mse.2026.154082

期刊菜单

人工智能在企业安全生产管理工程中的应用综述——从视觉感知到大模型智能决策的演进
A Review of the Application of Artificial Intelligence in Enterprise Work Safety Production Management Projects—The Evolution from Visual Perception to LLM Intelligent Decision-Making

DOI: 10.12677/mse.2026.154082, PDF,
作者: 李龙：新疆鲁源应急科技有限公司，新疆喀什；杜晨晨：中国矿业大学(北京)力学与土木工程学院，北京
关键词: 人工智能；安全生产；管理工程；计算机视觉；预测性维护；风险预警；Artificial Intelligence； Safety Production； Management Engineering； Computer Vision； Predictive Maintenance； Risk Early Warning

摘要: 安全生产是企业发展的生命线，亦是现代企业管理工程领域的核心课题。随着工业化推进和企业规模扩大，传统人工巡查、被动响应的安全监管模式已凸显效率低、盲区大、实时性差等局限。本文从管理工程与信息技术交叉融合的视角，综述人工智能技术在企业安全生产中的应用现状与发展潜力，重点梳理计算机视觉、物联网与机器学习、数据挖掘与自然语言处理等核心应用场景，分析技术落地面临的数据孤岛、模型泛化不足及部署成本高的挑战，展望端云协同与大模型应用趋势，为企业构建闭环的智慧安全管理工程体系提供理论参考。

Abstract: Work safety is the lifeline of enterprise development and also a core research topic in the field of modern enterprise management engineering. With the advancement of industrialization and the expansion of enterprise scale, the traditional safety supervision mode relying on manual inspection and passive response has prominently exposed limitations such as low efficiency, widespread blind areas and poor real-time performance. From the interdisciplinary perspective of management engineering and information technology, this paper summarizes the application status and development potential of artificial intelligence technology in enterprise work safety. It mainly sorts out the core application scenarios of computer vision, Internet of Things combined with machine learning, data mining and natural language processing. The challenges faced by technology implementation are analyzed, including data silos, insufficient model generalization ability and high deployment costs. Furthermore, the development trends of edge-cloud collaboration and large model application are prospected, providing theoretical references for enterprises to construct a closed-loop intelligent safety management engineering system.

文章引用：李龙, 杜晨晨. 人工智能在企业安全生产管理工程中的应用综述——从视觉感知到大模型智能决策的演进[J]. 管理科学与工程, 2026, 15(4): 848-855. https://doi.org/10.12677/mse.2026.154082

参考文献

[1]	Cai, R., Li, J., Tan, Y., Tang, J. and Chen, X. (2025) Convolutional Neural Networks for Construction Safety: A Technical Review of Computer Vision Applications. Applied Soft Computing, 180, Article 113374. https://doi.org/10.1016/j.asoc.2025.113374
[2]	Wang, Z., Zhu, Y., Zhang, Y. and Liu, S. (2025) An Effective Deep Learning Approach Enabling Miners’ Protective Equipment Detection and Tracking Using Improved YOLOv7 Architecture. Computers and Electrical Engineering, 123, Article 110173. https://doi.org/10.1016/j.compeleceng.2025.110173
[3]	Wang, S., Park, S., Kim, J. and Kim, J. (2025) Safety Helmet Monitoring on Construction Sites Using Yolov10 and Advanced Transformer Architectures with Surveillance and Body-Worn Cameras. Journal of Construction Engineering and Management, 151, Article 04025186. https://doi.org/10.1061/jcemd4.coeng-16760
[4]	Chen, Q., Long, D., Wang, S., Chen, Q. and Yuan, B. (2025) Real-Time Detection of Personal Protective Equipment Violations for Construction Workers Using Semi-Supervised Learning and Video Clips. Journal of Construction Engineering and Management, 151, Article 04024213. https://doi.org/10.1061/jcemd4.coeng-15310
[5]	Jahan, M.K., Bhuiyan, F.I., Amin, A., Mridha, M.F., Safran, M., Alfarhood, S., et al. (2025) Enhancing the YOLOv8 Model for Realtime Object Detection to Ensure Online Platform Safety. Scientific Reports, 15, Article No. 21167. https://doi.org/10.1038/s41598-025-08413-4
[6]	Xiao, Y., Huo, Y., Cai, J., Gong, Y., Liang, W. and Kołodziej, J. (2024) ERF-XGB: An Edge-IoT-Based Explainable Model for Predictive Maintenance. IEEE Transactions on Consumer Electronics, 70, 4016-4025. https://doi.org/10.1109/tce.2024.3371440
[7]	Ngandu Kalala, G., Boujelben, A., Chiementin, X. and Rasolofondraibe, L. (2026) Towards Accurate Gear Fault Diagnosis: Hybrid Data Enrichment by Digital Twins for Machine Learning-Based Monitoring. Journal of Vibration Engineering & Technologies, 14, Article No. 13. https://doi.org/10.1007/s42417-025-02252-3
[8]	Ma, H., Fan, C., Zhang, Y., Wang, Q., Yu, K. and Ma, Z. (2025) Digital Twin-Inspired Methods for Rotating Machinery Intelligent Fault Diagnosis and Remaining Useful Life Prediction: A State-of-the-Art Review and Future Challenges. Mechanical Systems and Signal Processing, 232, Article 112770. https://doi.org/10.1016/j.ymssp.2025.112770
[9]	Pandey, S., Singh, A.K., Parhi, S. and Jha, S.K. (2025) Towards Safer Steel Operations with a Multi Model Framework for Accident Prediction and Risk Assessment Simulation. Scientific Reports, 15, Article No. 13293. https://doi.org/10.1038/s41598-025-96028-0
[10]	Khan, M., Billion-Polak, P. and Khoshgoftaar, T.M. (2026) Artificial Intelligence and Machine Learning for Job Hazard Analysis: A Survey. IEEE Access, 14, 25605-25622. https://doi.org/10.1109/access.2026.3664316
[11]	Wang, Z., Guan, Z., Liu, X., Qian, B., Sun, X., Li, J., et al. (2026) Research on the Construction Method of Industrial Internet Risk Assessment Indicator System Based on LLMs. International Journal of Machine Learning and Cybernetics, 17, Article No. 215. https://doi.org/10.1007/s13042-026-03013-8
[12]	Oral, M., Alboga, Ö., Aydınlı, S. and Erdis, E. (2025) Usability of Large Language Models for Building Construction Safety Risk Assessment. Engineering, Construction and Architectural Management, 33, 7021-7048. https://doi.org/10.1108/ecam-08-2024-1143

为你推荐

友情链接