[论文解读] Dynamical Diagnosis and Solutions for Resilient Natural and Social Systems

Tatyana Kovalenko, Sornette, Didier|arXiv (Cornell University)|Nov 8, 2012

Complex Systems and Decision Making参考文献 133被引用 11

一句话总结

本文提出了一种动态诊断框架，通过区分压力源与内部压力、引入'风险时间'以进行预测性情景分析，并倡导使用'危机飞行模拟器'来应对认知偏见，从而增强自然与社会系统的韧性。该研究将韧性作为风险的互补指标，强调持续监控、多样化策略以及个体训练，以提升复杂系统的适应能力。

ABSTRACT

The concept of resilience embodies the quest towards the ability to sustain shocks, to suffer from these shocks as little as possible, for the shortest time possible, and to recover with the full functionalities that existed before the perturbation. We propose an operation definition of resilience, seeing it as a measure of stress that is complementary to the risk measures. Emphasis is put on the distinction between stressors (the forces acting on the system) and stress (the internal reaction of the system to the stressors). This allows us to elaborate a classification of stress measures and of the possible responses to stressors. We emphasize the need for characterizing the goals of a given system, from which the process of resilience build-up can be defined. Distinguishing between exogenous versus endogenous sources of stress allows one to define the corresponding appropriate responses. The main ingredients towards resilience include (1) the need for continuous multi-variable measurement and diagnosis of endogenous instabilities, (2) diversification and heterogeneity, (3) decoupling, (4) incentives and motivations, and (5) last but not least the (obvious) role of individual strengths. Propositions for individual training towards resilience are articulated. The concept of "crisis flight simulators" is introduced to address the intrinsic human cognitive biases underlying the logic of failures and the illusion of control. We also introduce the "time-at-risk" framework, whose goal is to provide continuous predictive updates on possible scenarios and their probabilistic weights, so that a culture of preparedness and adaptation be promoted. These concepts are presented towards building up personal resilience, resilient societies and resilient financial systems.

研究动机与目标

• 将韧性操作化定义为风险的补充，重点关注系统压力与恢复动力学。
• 区分外源性压力源与内源性应激反应，以实现有针对性的干预。
• 建立一种系统化的方法，用于诊断不稳定性并增强复杂系统的韧性。
• 通过'风险时间'框架持续更新预测情景，促进准备文化。
• 通过'危机飞行模拟器'对个人与组织进行训练，以应对危机响应中的人类认知偏见。

提出的方法

• 基于压力源（外部力量）与压力（内部系统反应）的区分，提出压力测量的分类体系。
• 提出'风险时间'框架，以提供潜在危机情景及其发生概率的持续性、概率性更新。
• 开发'危机飞行模拟器'，用于模拟高压事件，并训练个体克服控制错觉等认知偏见。
• 强调对内生不稳定性进行持续的多变量监测，以检测早期预警信号。
• 倡导结构性韧性策略：多样化、异质性、解耦合以及激励机制。
• 将个体优势与动机作为韧性建设的关键组成部分，尤其通过针对性训练实现。

实验结果

研究问题

• RQ1 如何以一种与传统风险指标互补的方式，对韧性进行操作化定义与测量？
• RQ2 外源性压力源与内源性系统压力有何区别？各自如何被诊断与管理？
• RQ3 持续的危机情景预测建模如何提升准备度与适应性响应？
• RQ4 认知偏见以何种方式削弱危机准备度？如何通过模拟加以缓解？
• RQ5 构建个人与系统韧性所必需的结构性与行为性要素有哪些？

主要发现

• 韧性最好理解为对压力源的动态内部响应，而不仅仅是失败的缺失。
• '风险时间'框架能够实现实时、概率性的场景更新，支持适应性决策。
• 危机飞行模拟器能有效抵消控制错觉等认知偏见，通过向个体暴露于高压、逼真的模拟情境中。
• 对内生不稳定的持续多变量监测，对于早期发现系统性脆弱性至关重要。
• 多样化、解耦合以及激励机制是系统级韧性的关键结构性促成因素。
• 通过模拟高压环境的针对性训练，可显著提升个体韧性，从而改善现实世界中的危机应对能力。