Spare Parts AI-Demand Alignment (SPADA)

GMI-DC-027Data CentersPilot-Ready

Definition

SPADA measures the alignment between actual spare parts demand (as shifted by AI-induced failure mode changes) and current stocking levels, weighted by the lead time and downtime cost consequences of stock-out events for each part category.

The Measurement Gap

Spare parts stocking levels are typically set using historical consumption data with simple min/max or EOQ calculations. When AI changes operating conditions and failure modes shift (as AIFMSI measures) the historical consumption pattern becomes a lagging indicator rather than a forward-looking predictor. A part that failed frequently under pre-AI conditions may become rare; a part that was rarely needed may become critical. Standard inventory management KPIs (stock turnover, fill rate, carrying cost) measure inventory efficiency but do not detect whether stocking levels are misaligned with the new demand pattern. The mismatch between stocking and demand is invisible until a critical stockout occurs at the worst possible time: during a corrective maintenance event on AI-modified equipment.

Decision Supported:

SPADA determines whether the spare parts inventory is aligned with the current AI-shifted failure demand pattern, or whether stocking levels need to be recalibrated to reflect the failure modes that AI operational changes have made more or less common.

Formula

Plain language: For each part category, compare the trailing 6-month average monthly demand to the current stocking target. Weight the discrepancy by the criticality of the gap: parts with long lead times and high downtime costs are weighted most heavily. Subtract the weighted misalignment from 1.0. Also compute the Critical Stockout Risk percentage to identify categories at most immediate risk.

Notation:

SPADA = 1 − Σ_c (w_c × |Demand_trend_c − Stock_level_c| / max(Demand_trend_c, Stock_level_c)) / Σ_c w_c
CSOR = Σ_c (w_c × Stockout_risk_flag_c) / Σ_c w_c × 100

Variables:

Variable Definition Unit
c Part category index (e.g., CRAC compressors, VFD modules, UPS capacitors, cooling tower fan motors) n/a
Demand_trend_c Average monthly demand for category c over trailing 6 months Units/month
Stock_level_c Current stocking target (not on-hand, but the target that reorder points are set to) Units
w_c Criticality weight: lead_time_days × downtime_cost_per_day (normalized) Dimensionless weight
Stockout_risk_flag_c 1 if current stock-on-hand < 1 month of demand_trend_c, 0 otherwise Binary

Use Poisson distribution for parts with lumpy (infrequent, variable-size) demand rather than simple averaging. Parts with approved substitutes may have effective availability greater than stocked quantity: account for this in the alignment calculation.

Proposed formula, pending validation pilots.

At a Glance

Unit
Index (0 to 1)
Trend Direction
Higher; closer to 1.0
Reporting Frequency
Monthly; with quarterly full review
Category
Failure Mode Shift
← AI-Induced Failure Mode Shift Index (AIFMSI)Inspection Domain Coverage Rate (IDCR) →

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