Abstract: The current global climate anomaly is anticipated to lead to a heightened frequency of extreme natural disaster events in the future. With the rapid pace of urbanization, there is an increasing development of large urban complexes. This convergence of high-risk and fragile social systems will further underscore the interconnection, complexity, and importance of extreme disaster events. As a result, a single emergency entity can no longer adequately address the emergency needs arising from such complex disasters, necessitating collaboration among multiple entities within emergency organizations. However, cross-organizational emergency coordination does not always operate with optimal efficiency. The feasibility of emergency plans, fragmentation of inter-organizational collaboration, and inadequate crisis learning all impede the maximization of efficiency in inter-organizational collaborative governance. It is crucial to optimize and strengthen cooperation among emergency organizations and establish a coordinated and efficient risk management model.
This article focuses on a significant major emergency—the “7.20” super heavy rainstorm disaster in Zhengzhou, Henan province, as the subject of research. The primary source of research data consists of case-related news reports and emergency plans at all levels published on government websites. Initially, text analysis is employed to identify the emergency organizations involved in this major event and establish them as nodes within the emergency collaboration network. The interactive and collaborative relationships between organizations are considered as edges within the emergency collaboration network. From a systemic and practical perspective, inter-organizational planning and collaboration networks based on the emergency plan are constructed alongside actual response collaboration networks during two distinct time stages. Subsequently, social network analysis is utilized to compare structural differences and evolutionary characteristics across different types and stages of emergency collaborative networks by examining both overall network characteristics and individual organizational positions. Furthermore, this study delves into assessing the alignment between inter-organizational collaboration outlined in the emergency plan with actual responses during current disaster relief efforts, as well as evaluating local government learning effects during crises.
The main findings of this paper are as follows: (1)The alignment between the plan and the actual collaborative situation is suboptimal. Emergency planning poses challenges to facilitating efficient emergency response collaboration, exhibiting characteristics of “limited operability”. This is evidenced by a lack of forward extension, inadequate information communication mechanisms, insufficiently prominent unified command subject, and oversight of certain critical emergency response nodes. (2)Additionally, there are deficiencies in the initial actual response collaborative network with regards to collaborative efficiency and the role of the core organization. (3)Timely crisis learning can significantly enhance the efficacy of emergency response collaboration in major emergencies. The actual response collaboration network demonstrates characteristics of automatic evolutionary optimization under government crisis learning influence. It tends to form an emergency organizational network structure with core organizational leadership and coordinated division of labor.
Based on the findings of this study, future research will endeavor to investigate the influencing factors that promote the establishment of coordination relationships among diverse emergency responders in large-scale disaster scenarios, and enhance our comprehension of the dynamic evolution mechanism of emergency collaboration networks. Additionally, it would be advantageous to validate the scientific principles identified in this investigation through examination of additional similar cases of extreme disasters.

Key words: social networks, planning collaborative networks, response collaborative networks, Zhengzhou rainstorm, learning in crisis

摘要： 当前城市化进程中自然灾害事件的复杂性、耦合性日益加剧,亟待优化协同高效的风险治理模式。为此,本文以河南郑州“7.20”特大暴雨灾害为例,从制度和实践两个层面,构建其基于应急预案的组织间计划协作网络和两个不同时间阶段下的响应协作网络,并对三种网络特征进行对比分析。结果表明,根据应急预案和实际应急系统分别构建的协作网络结构存在较大差异,主要原因一方面源于应急预案本身存在不足,另一方面则是实际响应出现失灵;此外,实际响应协作网络在政府危机学习作用下呈现自动演进优化特征,趋于核心组织统一指挥、组织间紧密联系、协调分工的应急网络结构。建议应急管理体制改革加强应急预案动态调整并建立政府危机学习长效机制。

关键词: 社会网络, 计划协作网络, 响应协作网络, 郑州暴雨, 危机学习