Home· Govern· Inventory Management
Inventory Management · M.03

Right part. Right truck. Right time.

Spare parts forecast against the actual failure curve, not the calendar. Multi-warehouse balancing, vendor lead-time aware, tied to the truck rolls Asset Intelligence is about to dispatch.

inv.live · 14 warehouses 98.4% FIRST-PICK · −24% CARRY
Critical SKUs · forward-looking 30d
$8.4M · inventory · vs $11M before
Days on hand
14
PO triggers
42
Stockouts
2
AC contactor 200A · Charlotte
forecast: 14 dispatches/30d
PO 18
fiber pigtail · Atlanta
vendor lead +5d slip
EXPEDITE
Live · the stock curves 4 SKU stock levels · reorder triggers fire · restock signal emitted
REORDER · 20% ⚑ RESTOCK 100% 60% 0% 4 SKUs · 1 RESTOCK FIRED · 0 STOCKOUTS
▸ Vault deposit
What this guardrail produces
▸ 6 mo in vault → first tier unlocked
▸ Evidence
Asset visibility · working-capital cycle · stockout exposure
▸ Tier unlocked
Tier 1 · banksTier 3 · insurance-linked
Guardrail rule: every SKU level snapshotted on change, never overwritten — see /the-vault for the full underwriting fabric.
What it looks like

Track. Allocate. Predict.

Real-time stock with floor markers. Depot allocation heatmap. Pipeline-aware reorder triggers.

§ 01 · Stock health

Top SKUs · live · Montréal

Floor enforced at 30%
PP-24
4/28
SW-48POE
14/33
AP-6E
72/106
NET-6A
180/215
§ 02 · Depot allocation

Pipeline-balanced · 4 regions

Top 4 SKUs · East/Central/West/EU
SKU
E
C
W
EU
AP-6E
72
110
38
14
SW-48
14
42
68
8
PP-24
4
22
38
§ 03 · Reorder queue

From the deal, not the average

Next 14 days · 3 reorders
PP-24 · 24 units1d
AP-6E · 72 units4d
SW-48POE · 12 units9d
The problem

Inventory carries 2× what it needs to.

Safety stock based on usage averages, not forecasted failure curves. Half your warehouse is junk you'll never need; the other half stocks out at the wrong time.

Without Inventory Management

Today's status quo

  • Safety stock = average × multiplier — overcarries by 80%, stockouts anyway
  • PO triggers fire on min/max — no awareness of upcoming truck rolls
  • Vendor lead-time variance not modelled — ghosts in the schedule
  • Multi-warehouse balancing manual; trucks roll for parts that exist 50mi away
With Inventory Management

What changes

  • Forward-forecast from Asset Intelligence — stock for tomorrow's failures, not yesterday's average
  • PO triggers fire against expected demand and vendor lead time, not min/max
  • Vendor lead-time variance modelled per SKU — expedite policies trigger automatically
  • Multi-warehouse balancing continuous — parts redistribute before trucks roll
Capabilities

What's inside.

The six capabilities that make this module work end-to-end. Pick any one as your starting point — they compound.

01

Demand forecasting

Forecasts driven by Asset Intelligence's failure curves — not historical averages. The right SKU, in the right warehouse, before the truck dispatches.

Forecast · Failure-curve
02

Vendor lead-time modeling

Per-SKU, per-vendor lead-time distribution. Expedite, alternate-source, and dual-source policies trigger automatically when slips appear.

Lead time · Variance
03

Multi-warehouse balancing

Continuous re-balancing across warehouse network. Stock follows demand, not yesterday's geography.

Balance · Network
04

Truck-roll binding

Inventory reserves at dispatch — Asset Intelligence sees the part is on a truck, not in a warehouse. No double-allocation.

Bind · Reserve
05

Carry-cost optimization

Carry vs stockout cost modelled per SKU. Slow-movers retire automatically; critical SKUs over-stocked deliberately. Total cost minimized.

Carry vs Stockout
06

Vendor scorecards

Lead time, fill rate, RMA rate, recall events — scorecarded per vendor. Procurement leverage at the next negotiation.

Scorecards · Procure
The autonomous loop

From failure curve to right truck.

Asset Intelligence forecasts failures. Inventory pre-positions parts. Truck rolls with the diagnosis and the right SKU. POs fire from real demand, not min/max.

§ 01 · Forecast

Failure curves

Asset Intelligence's failure model becomes the demand forecast — per SKU, per region.

Daily
§ 02 · Position

Multi-warehouse

Network-aware balancing redistributes parts before trucks roll.

Continuous
§ 03 · Reserve

Truck binding

Dispatch reserves the part; inventory state visible to all modules in real time.

< 1s
§ 04 · Replenish

PO with intent

POs fire with expected demand, lead-time variance, alternate-source policy.

Per cycle
Policy you can read

Stock policy per SKU class.

Critical safety SKUs (breakers, contactors) overstock deliberately. High-velocity commodities (fasteners) run lean. Long-lead vendor-locked items dual-sourced where possible.

Each SKU class has its own policy. Total carry minimized. Stockout risk capped per criticality.

policy · sku-class.alm
# Stock Policy · SKU classes
class "safety-critical":
  service_level   = 99.9%
  cover_days      = 21
  expedite_when_below = 7d
  dual_source     = required

class "high-velocity-commod":
  service_level   = 99%
  cover_days      = 5
  blanket_po      = annual

class "long-lead-vendor-locked":
  service_level   = 98%
  cover_days      = vendor.lead_time × 1.5
  forecast_horizon = 90d

class "slow-mover":
  retire_when     = usage.last_180d == 0
  liquidate_via   = "secondary-market"
Where it lives

Plugs into your supply chain stack.

ERP

Inventory of record

NetSuite, SAP, Oracle — Allometry layers on top, no forklift.

WMS

Warehouse ops

Manhattan, Blue Yonder, native — pick, pack, transfer in your tool.

Procurement

PO + RFQ

Coupa, Ariba, native — POs fire with full demand context.

Field

Truck binding

ServiceTitan, FieldEdge, native — parts reserved at dispatch.

Real outcomes

"Eleven SKU exceptions caught before invoicing in the first 30 days. Stopped a $96K leak we didn't even know was leaking."

Director Supply Chain · Confidential design partner · pallet manufacturing
$96KLeak stopped · 30 days
11SKU exceptions caught
99.2%Invoice accuracy post-deploy
Confidential design partner · case study Pallet manufacturing · Quebec · 400+ customer addresses
See it on your data

Run the carry vs stockout math.

Give us your top 200 SKUs and 12 months of usage. We'll return the carry-cost reduction and stockout drop you'd see — modelled on your data.