The situation

Token 47 is a number with no information attached. A person holding it does not know whether they have twenty minutes or ninety, so they sit where they can see the board. They skip food, skip the pharmacy, and skip the second counter they need, because leaving the chair means losing their place in a line that has no shape they can see.

The same number drives the other side of the problem. Everyone arrives at nine, the counters fill, and the department spends the morning managing a crowd that could have been spread across two hours.

What the guidance does

Two pieces have to arrive together. The first is a queue estimate: how many people are ahead, how fast the counter has been serving, and how much time that implies. The second is walk time from the map. Subtract the walk from the estimate and you get a time to start moving, which is the sentence a person actually needs: leave now, you will arrive three minutes early.

With estimates published, arrival windows spread. A person with an 8:26 turn can be told to come at 8:20 rather than to sit from 7:45.

What changes

Waiting turns into free time. People collect reports, eat something, or sit somewhere quieter. Peak crowding at the counter drops because the queue stops being the only place a person can wait.

Where this stands

In design. The first version reads the current token from staff entry or from the counter board, and calculates the estimate from observed service rate. That version needs no connection to hospital systems, which is what makes it shippable in a pilot.

Reading tokens directly from the hospital information system is a separate step that depends on per-vendor integration, and it is the piece that makes the estimate exact. That is on the roadmap rather than in the pilot scope.