浅析空中交通管制工作风险因素及技防措施的应用

Analysis of Risk Factors in Air Traffic Control Operations and the Application of Technical Defense Measures

摘要: 空中交通管制系统是一个典型的“人–机–环–管”多维嵌套的社会–技术系统，其运行安全性产生于管制员、飞行员、基础设施、运行程序和动态交通环境的持续交互之中。基于此，本文系统梳理了空管工作中的主要风险因素，重点阐述了西北空管局部署的ATC安全防护系统、管制员画像系统和气象智能服务系统3套技防措施的技术架构与应用成效。通过对系统投用前后关键安全指标的对比分析，验证了技防措施在降低最小安全间隔破坏预警事件(降幅61.5%)、减少通信错误(降幅55.6%)、缩短复杂天气响应时间(降幅51.1%)等方面的显著效能。同时，本文客观分析了技防措施在数据依赖、语音识别准确率、自动化偏见及系统实施等方面存在的局限性，从风险识别系统性、预警前置智能化、能力评估数据化、人机协同适用性及人工智能嵌入性五个维度，提炼了空管安全管理工作的基本原则。最后，对“人机增强智能”在空管领域的未来应用模式进行了展望，以期为空管安全管理从“被动应对”向“主动防御”转型提供理论依据和实践参考。

Abstract: The air traffic control (ATC) system is a typical social-technical system with multi-dimensional and nested “human-machine-environment-management” characteristics, whose operational safety emerges from the continuous interaction among controllers, pilots, infrastructure, operational procedures, and the dynamic traffic environment. Based on this, this paper systematically identifies the major risk factors in ATC operations, and focuses on elaborating the technical architectures and application effectiveness of three technical defense systems deployed by the Northwest Air Traffic Management Bureau—namely the ATC safety protection system, the controller capability profiling system, and the meteorological intelligent service system. Through a comparative analysis of key safety indicators before and after the implementation of these systems, the significant effectiveness of the technical defense measures is verified in terms of reducing minimum separation infringement warning events (a decrease of 61.5%), reducing communication errors (a decrease of 55.6%), and shortening response time under complex weather conditions (a decrease of 51.1%). Meanwhile, this paper objectively analyzes the limitations of the technical defense measures in aspects such as data dependency, speech recognition accuracy, automation bias, and system implementation, and distills the fundamental principles for ATC safety management from the five dimensions of systematic risk identification, intelligent early warning, data-driven capability assessment, human-machine collaboration applicability, and AI embeddability. Finally, this paper provides an outlook on the future application models of “human-machine augmented intelligence” in the ATC domain, aiming to provide a theoretical basis and practical reference for the transformation of ATC safety management from “passive response” to “active defense”.