World Model Readiness When the plan meets the floor

Module · When the plan meets the floor

The Operations Team AI Check

Operations is where a clean plan meets a messy floor, and the gap between them is filled by people who have run this route, this line, this shift a hundred times. AI planners are good at the plan and blind to the floor. This module checks the five things that decide whether an AI planner strengthens your operation or overrides the people who keep it running: who can override the system, how exceptions get handled, whether hard-won judgment is written down, whether you measure plan against actual, and whether your alerts still mean anything.

Question 1 of 5 · Humans can override

When the AI planner is wrong, can the people on the floor override it?

A dispatcher who can see the plan is about to fail but cannot change it becomes a spectator to a slow crash. Override authority, held by the people closest to the work and respected when they use it, is what keeps an AI planner a tool rather than a boss.

Question 2 of 5 · Exceptions have a path

When something happens the plan did not foresee, does your team know what to do?

Every operation runs on exceptions: the late truck, the sick driver, the machine that throws a fault at 3am. An AI planner optimises the normal case and goes quiet on the strange one. Whether your team has a clear path for the exception, or improvises every time, is the difference between a hiccup and an incident.

Question 3 of 5 · Tribal knowledge captured

Is the knowledge that makes your operation run written down anywhere?

Your best dispatcher knows which customer always underestimates volume, which route floods in October, which supplier means it when they say noon. None of that is in the system. An AI planner cannot learn what was never captured, and neither can the next hire.

Question 4 of 5 · Plan versus actual

Do you measure what the AI planned against what actually happened?

A planner that is never checked against reality drifts into confident nonsense. Comparing the plan to the actual outcome, shift by shift, is how you find out whether the AI is helping or just producing tidy numbers that the floor quietly ignores.

Question 5 of 5 · Alerts still mean something

When the system raises an alert, does your team still take it seriously?

An AI operation generates alerts, and an operation drowning in alerts trains its people to ignore all of them. The alert that finally matters arrives in a sea of ones that did not. Whether your team still reacts, or has learned to swipe them away, decides whether the system protects you or just performs vigilance.

For the statistics · one click each

Three questions for the public picture

These do not affect your score. They feed the anonymised, aggregated statistics; groups under 8 respondents are never shown.

How much of your operational planning does AI now drive?

None, all human
Suggests, humans decide
Plans, humans approve
Plans and executes
We are not sure

Can your floor staff override an AI plan when they know it is wrong?

No, they cannot
Only informally
Yes, clearly
Yes, and respected
No AI planner yet

How much of your team's operational knowledge is actually written down?

Almost none
Some of it
Most of it
Nearly all of it
We do not know

Your context

Used to calibrate the report. Company size and sector remain in the anonymized dataset; your email does not.