Inspection Skill Retention Index (ISRI)

GMI-TEL-040TelecommunicationsInfrastructure-DependentPattern O-9: Skill Atrophy

Definition

Change in defect detection rate and severity grading accuracy for physical (climber/field) inspectors over time, measured against a baseline established before drone inspection became the primary method. ISRI uses a multiplicative composite (detection rate multiplied by grading accuracy) because both capabilities must be maintained for the human backup assumption to hold.

The Measurement Gap

As drone inspection replaces climber inspection, the frequency of hands-on structural inspection practice drops dramatically. Physical inspectors who once climbed 100 towers per year may now climb 10. No telecom operator currently measures whether this reduced practice is degrading the inspection workforce's ability to deliver reliable results when physical inspection is required: for regulatory mandates, emergency damage assessment, or validation of questionable CV findings. The skill degradation is silent because it only manifests when someone actually needs a physical inspection and discovers the workforce cannot deliver baseline-quality results.

The Decision

When a drone/CV inspection identifies something requiring human follow-up, or when a regulatory mandate requires physical inspection: is the inspection workforce still proficient enough to deliver reliable results? If ISRI shows degradation, the organization faces a compounding risk: the human backup it relies on when the CV system is inadequate is itself becoming inadequate. The most dangerous scenario is discovering this degradation only during a regulatory audit that requires physical inspection.

Research Foundation

Skill atrophy in safety-critical inspection roles follows predictable degradation curves when practice frequency drops below maintenance thresholds, with detection rate and diagnostic accuracy declining at different rates: detection degrades faster than grading because it requires active pattern recognition that atrophies without practice (Human Factors in Infrastructure Inspection, National Academy of Engineering, 2023). OSHA requires that tower climbers performing structural inspection maintain demonstrated competency, though the definition of competency assessment frequency is left to the employer (OSHA 29 CFR 1926.1427: Operator Qualification and Certification).

Formula

ISRI = (Detection_rate_current × Grading_accuracy_current) / (Detection_rate_baseline × Grading_accuracy_baseline)

Variable Definitions (every variable resolved)

  • Detection_rate_current = percentage of known defects detected by the inspector in the current proficiency assessment (percentage)
  • Grading_accuracy_current = percentage of detected defects correctly severity-graded in the current assessment (percentage)
  • Detection_rate_baseline = same metric from the baseline period (pre-drone deployment or first 12 months) (percentage)
  • Grading_accuracy_baseline = same metric from the baseline period (percentage)
  • Multiplied to create a composite because both detection and grading must be maintained: degradation in either dimension pulls the index down

Units: Dimensionless ratio. 1.0 = no change from baseline. Below 1.0 = skill degradation. The product formulation means that degradation in either detection or grading pulls the index down.

Edge Cases

Inspectors hired after drone deployment (no personal baseline): use the team baseline established at drone deployment as their reference. Validation structures may not represent the full range of defect types: rotate validation structures and defect types across assessment cycles. Inspector awareness of being tested (Hawthorne effect): supplement structured assessments with blind reviews of drone imagery where inspector severity grades are compared to expert panel grades. Minimum sample: assess each inspector annually with a minimum of 20 known defects per assessment.

These are proposed starting points: calibrate to your operational context during pilot: ISRI above 0.85 means skill retention is adequate. Below 0.7 means the inspection workforce has meaningfully degraded and the human backup assumption is unreliable. If ISRI drops below 0.8, implement mandatory annual proficiency inspections (real climbs on validation structures) for all inspection personnel.

Confidence Grading (required alongside every ISRI value):

Grade Basis Interpretation
A: Empirical 5+ observed events in measurement window provide direct tail observation Measurement grounded in observed outcomes; highest confidence
B: Hybrid 1-4 observed events plus Monte Carlo simulation with Bayesian updating Blends observation and simulation; moderate confidence
C: Simulated Zero observed events; derived entirely from Monte Carlo simulation Modeled, not observed; treat as indicative, not validated

A ISRI of any value at Grade C is a simulation that agrees with itself. It tells you the model is internally consistent but not that it matches reality. Grade C scores should trigger proactive stress-testing against historical near-miss events and cross-asset analogues rather than acceptance as calibrated.

Proposed formula, pending validation pilots.

At a Glance

Unit
Ratio
Trend Direction
Toward 1.0; no decay
Reporting Frequency
Semi-annually
Category
AI Inspection Accuracy and Regulatory Sufficiency
← Inspection Detection Coverage Index (IDCI)Regulatory Inspection Sufficiency Score (RISS) →

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