Sub-function
Network Design
Owns network study work end to end inside Supply Chain. Runs 3 lines across 17 stations, drains the repeatable share into no shared tower, and holds 4 human gates.
Estate rating
BB
67 / 100
Deliberately capped low. Network modeling is fully agentic, but opening or closing a facility is a capital decision with employment consequences that only a board can take.
Agents
8
Lines
3
Units / wk
7,324
Timetable vs actual
6.1h → 6.5h
Cost / unit
$12
was $30
Run-rate saving
$7.1M
annualized at this volume
The 8 agents running this sub-function
Grouped by what they are for, not by where they sit. The agent org is process-shaped.
Network Design Orchestrator
Owns the network design lane end to end. Sequences the other agents, holds the timetable, and decides what surfaces to a human.
ceiling A2 · escalates to Escalates to VP Supply Chain Strategy when the unit falls outside the decision envelope or confidence drops below the floor.
Network Modeling Agent
Executes the network design step it owns, posts its evidence, and hands the unit to the next stage.
ceiling A2 · escalates to Escalates to VP Supply Chain Strategy when the unit falls outside the decision envelope or confidence drops below the floor.
Cost-to-Serve Agent
Executes the network design step it owns, posts its evidence, and hands the unit to the next stage.
ceiling A2 · escalates to Escalates to VP Supply Chain Strategy when the unit falls outside the decision envelope or confidence drops below the floor.
Footprint Scenario Agent
Executes the network design step it owns, posts its evidence, and hands the unit to the next stage.
ceiling A2 · escalates to Escalates to VP Supply Chain Strategy when the unit falls outside the decision envelope or confidence drops below the floor.
Service Level Modeling Agent
Executes the network design step it owns, posts its evidence, and hands the unit to the next stage.
ceiling A2 · escalates to Escalates to VP Supply Chain Strategy when the unit falls outside the decision envelope or confidence drops below the floor.
Carbon Modeling Agent
Executes the network design step it owns, posts its evidence, and hands the unit to the next stage.
ceiling A2 · escalates to Escalates to VP Supply Chain Strategy when the unit falls outside the decision envelope or confidence drops below the floor.
Network Strategy Agent
Authors positions on network design. Frames the question, models the options, and writes a recommendation a human can ratify or reject.
ceiling A2 · escalates to Escalates when two options sit inside the confidence band and the evidence cannot separate them.
Network Design Challenger Agent
Adversarial reviewer for network design. Argues the opposite case on every position and flags where the evidence does not carry the claim.
ceiling A2 · escalates to Escalates when a ratified position is still running against a break condition it flagged.
Estate rating breakdown
Seven control dimensions. Judgment readiness runs against autonomy on purpose.
People
What the humans stopped doing, and what they do now.
Redeployed into scenario work and strategy authoring review.
Lines running in this sub-function
Each line is a workflow. Click a line to walk it station by station.
Strategy positions this sub-function holds
An agent that only executes is a robot. These are the calls it made, the argument against each one, and what would prove it wrong.
Network Study
In challengev1Where should the autonomy ceiling for network design sit over the next four quarters, and what evidence would justify moving it?
Hold the ceiling at A2 for network design. Deliberately capped low. Network modeling is fully agentic, but opening or closing a facility is a capital decision with employment consequences that only a board can take. Raise it only after two consecutive quarters where the override rate stays under five percent and every override has a written cause.
confidence
84%
The challenge — Network Design Challenger Agent
Network Design Challenger Agent argues the recommendation leans on 2 quarters of data from a period with no volume shock. The confidence band overlaps the alternative, and the position does not say what it would take to be wrong. Recorded as a dissent, not a block.
Break conditions
- ·Override rate rises above 14 percent for two consecutive months
- ·Cost per network study stops falling while volume keeps rising
- ·A control failure in this lane reaches a customer or a regulator
Evidence gaps
- ·No independent test of the control evidence behind the touchless rate
| Alternative considered | Cost | Risk | Verdict |
|---|---|---|---|
| Hold the current position | $161k run rate | Known and priced. Cedes ground if comparators move faster. | live |
| Raise the ceiling one level now | $813k to build controls | Override rate is 12 percent; raising the ceiling before that settles imports the error into production. | rejected on evidence |
| Move the exception tail to the spine | $148k transition | Loses local context. Rework risk on the cases that are hardest to recover. | under review |
Network Design Operating Position
Ratifiedv4Should network study work stay in the shared transaction spine, or come back inside the function where the context lives?
Keep the repeatable network study volume in the spine and pull the exception tail back into the function. The spine is cheaper per unit; the tail is where the context that prevents rework actually sits.
confidence
67%
The challenge — Network Design Challenger Agent
Network Design Challenger Agent argues the recommendation leans on 3 quarters of data from a period with no volume shock. The confidence band overlaps the alternative, and the position does not say what it would take to be wrong. Recorded as a dissent, not a block.
Break conditions
- ·Override rate rises above 14 percent for two consecutive months
- ·Cost per network study stops falling while volume keeps rising
- ·A comparator publishes a materially lower unit cost on a like-for-like basis
Evidence gaps
- ·No external comparator on a like-for-like unit definition
- ·Exception cases under $15k are sampled, not fully measured
| Alternative considered | Cost | Risk | Verdict |
|---|---|---|---|
| Hold the current position | $531k run rate | Known and priced. Cedes ground if comparators move faster. | not selected |
| Raise the ceiling one level now | $714k to build controls | Override rate is 11 percent; raising the ceiling before that settles imports the error into production. | rejected on evidence |
| Move the exception tail to the spine | $791k transition | Loses local context. Rework risk on the cases that are hardest to recover. | selected |
Ratified 9 months ago. Measured since: cost per network study fell 17 percent against a forecast of 30 percent.
Policy register
The rules the agents above are bound by. Version, owner, approver, and the agents each rule constrains.
| Ref | Rule | Scope | Owner agent | Approved by | Version | Status |
|---|---|---|---|---|---|---|
| SC10-POL-200 | Network Design Decision Envelope Agents in this lane may act without a human when the network study sits inside the stated value, risk and confidence envelope. Outside it, the unit holds at a gate with a named approver and a running clock. binds 5 agents · US · Canada | envelope | Network Design Orchestrator | VP Supply Chain Strategy | v2.1 | under review |
| SC10-POL-201 | Network Design Evidence Standard Every autonomous decision writes inputs, the rule version applied, the model and prompt version, the output and a reversal path. An action with no evidence record is treated as a control failure, not a fast decision. binds 5 agents · APAC | evidence | Network Design Orchestrator | VP Supply Chain Strategy | v5.9 | under review |