AEROSPACE AND DEFENSE

Revenue-at-risk supplier recovery for a global aerospace manufacturer

Buyers chased late purchase orders by count, not by how much downstream revenue each one gated. Wexa built a recovery cockpit on Wexa that traces every late part through the bill of materials to the programs it blocks and ranks buyer effort by dollar leverage.

Download this case as PDF Book a demo

BOM where-used traversal in the context graph
agents: orchestrator, impact, risk, recovery, compliance critic
of supplier communication approval-gated
automated tests; export-control aware

CLIENT: Global aerospace and defense manufacturer (engines and auxiliary power units)
REGION: US
WHEN: 2026
ENGAGEMENT: Solution build
WEXA COMPONENTS: Wexa · declared ontology · context graph · 5-agent process flow · approval gate · lifecycle audit

01Case 15

The challenge

One critical supplier's past-due parts can gate hundreds of millions in engine and APU revenue
Days to weeks between the late signal and a recovery action
Export-controlled data and strict audit requirements

What Wexa built

Context graph of suppliers, parts, bills of material, purchase orders and programs, from a declared ontology
Impact agent: where-used traversal → gated revenue → ranked leverage list
Risk agent separates chronic from one-off supplier slippage
Recovery agent drafts the plan and the supplier email; never commits on the company's behalf
Compliance critic checks citations, schema and export classification; 10-stage governed lifecycle as the audit ledger
Operator console: cockpit, POs, suppliers, parts, programs, graph, recovery, approvals, runs, audit

02Case 15

How it runs on Wexa

FigureHow this use case runs on Wexa: client systems connected, ontology discovered, context graph in the Wexa memory layer, agents and process flows, and governance. Highlighted components are central to this use case.

03Case 15

Results

Measure	Before / without Wexa	With Wexa
Buyer hours a week on late-PO prioritisation (20 buyers)	200 hours	80 hours
Annual cost of the process	$528K	$211K
Program revenue at risk surfaced and ranked per cycle	–	$4.8M
Time to implement the AI solution	9 months	8 weeks
AI build cost (one-time)	$1.2M	$89K
AI operating cost per year (models, infrastructure, upkeep, platform)	$704K	$336K
Total AI cost in year 1	$1.9M	$425K

With Wexa

net annual value to the client after Wexa's own operating cost

faster to implement

lower AI operating cost

less AI spend in year 1

Working cockpit on Wexa with 12 screens; 24-week pilot plan to a live buyer pilot
Pilot targets agreed: −50% time to detect, ≥70% recommendation acceptance, ≥6 buyer-hours saved a week

Parts carry export classifications; a US-person gate and project-scoped graph queries are enforced.