Topological Resilience Reserve (TPRR)

GMI-MRO-025MROResearch-Forward

Definition

TRR measures the fraction of network connectivity (algebraic connectivity) the plant retains after the highest-risk assets are removed, reported at k = 1, 3, and 5.

The Measurement Gap

Criticality rankings are sorted lists; plants are networks. Lists rank nodes, networks have topology: connections, paths, redundancies, bottlenecks. MTBF, MTTR, OEE, and PM Compliance are all node-local and cannot describe what happens when three shared utilities go down at once. Albert, Jeong, and Barabási (2000) showed scale-free networks are robust to random failures but catastrophically fragile under targeted attack on hub nodes: exactly the profile of most industrial plants. The criticality ranking already says those nodes matter, but it cannot answer what happens when two or three of them fail in the same week, or which specific combination of removals would fragment the production network into isolated islands that can no longer support each other. It cannot say how much the network depends on them structurally.

Decision Supported:

Where to invest in redundancy, alternate paths, or operational segmentation to harden the network against simultaneous failures of the highest-risk assets.

Formula

Plain language: Build the plant as a graph (assets = nodes, dependencies = edges). Compute network connectivity. Remove the top-k highest-risk assets and recompute. The ratio is your reserve.

Notation:

TRR(k) = λ₂(L(G \ Sₖ)) / λ₂(L(G))

Variables:

Variable Definition Unit
λ₂ Fiedler value: second-smallest eigenvalue of the graph Laplacian; measures overall connectivity Dimensionless
L(G) Graph Laplacian of the plant network G Matrix
G \ Sₖ Plant network with the top-k highest-risk assets removed Subgraph
Sₖ Set of k assets with the highest risk scores from the Health/criticality platform Asset set
TRR(k) Fraction of baseline connectivity retained after removing k critical assets Ratio (0–1)

Report TRR(1), TRR(3), TRR(5) together with giant component fraction S(k), which tracks whether the network has fragmented into isolated subgraphs.

Proposed formula, pending validation pilots.

At a Glance

Unit
Ratio (0 to 1)
Trend Direction
Higher; approaching 1.0
Reporting Frequency
Quarterly; refresh after major topology changes
Category
Digital Twin & System Resilience
← Resilience Recovery Quality (RRQ)Twin Fidelity Drift (TFD) →

Assess Your Organization

The AI-GPS Assessment evaluates your AI governance maturity and recommends KPIs like this one for your implementation roadmap.

Take the Free AI-GPS Assessment