Network resilience measures how well a network maintains its connectivity and functionality when nodes or edges are removed. This is crucial for understanding organizational vulnerability to disruptions, from employee turnover to system failures.
Key Resilience Metrics:
Largest Connected Component: Size of the largest group of nodes that can reach each other
Network Connectivity: Proportion of original network still connected after removals
Average Path Length: Mean distance between connected nodes (increases as network fragments)
Network Efficiency: Inverse of average path length, normalized to [0,1]
Fragmentation Point: Threshold where network breaks into isolated components
Attack Strategies:
Random Failure: Nodes removed randomly, simulating equipment failures or random turnover
Targeted Attack (Degree): Removes highest-degree nodes first, targeting central hubs
Targeted Attack (Betweenness): Removes critical bridges that connect different parts of network
Cascading Failure: Removal triggers subsequent failures in connected nodes
Network Topology and Resilience:
Scale-Free Networks: Robust to random failures but vulnerable to targeted hub removal
Random Networks: Similar resilience to both random and targeted attacks
Small-World Networks: Maintain connectivity well due to local clustering and shortcuts
Centralized Networks: Very vulnerable to central node failure
Distributed Networks: More resilient but may be less efficient
Organizational Applications:
Succession Planning: Identify critical employees whose departure would fragment teams
Knowledge Management: Ensure expertise is distributed to prevent knowledge loss
Communication Networks: Design redundant pathways for information flow
Supply Chains: Assess vulnerability to supplier failures or disruptions
Project Teams: Build resilience through overlapping memberships and backup roles
Improving Network Resilience:
Redundancy: Create multiple paths between critical nodes or units
Decentralization: Distribute authority and expertise across multiple nodes
Cross-Training: Develop backup capabilities to compensate for node loss
Early Warning: Monitor centrality metrics to identify emerging vulnerabilities
Strategic Connections: Add edges between disconnected clusters to prevent fragmentation
The Robustness-Efficiency Tradeoff:
Highly efficient networks (short paths, central hubs) are often less resilient
Highly resilient networks (many redundant paths) may be less efficient
Organizations must balance resilience needs with operational efficiency
