S03 · stakeholder simulation

"95% accurate per flight": negotiating a forecast target

People, data and outputs in this scenario are simulated. Each stage is recorded when you commit; the reveal comes after.

4stages
35 minabout
0stages committed

The brief

You are the data scientist assigned to a new passenger-demand forecast at a synthetic airline (simulated company). The head of network planning (simulated) opens the kickoff with: "The new forecast has to be 95% accurate for every flight, otherwise we keep the spreadsheet." The finance controller (simulated) is also on the call and wants "a number we can budget against". The spreadsheet's accuracy has never been measured.

Facts for later stages come from the course's synthetic flight table; the people and their opinions are simulated.

Stage 1 of 4

Planning uses the forecast to set weekly capacity by route and to decide ground-staff rosters two days out. Finance uses a monthly total. Nobody has written down what "accuracy" means.

Which questions do you ask first? Select all that apply.

Which decision is made from the forecast, at what level (flight, route, week) and how far ahead.How "accuracy" is defined: which error measure, over which unit, and how over- and under-forecasts are weighed.What the spreadsheet achieves today on the same definition, so the target has a reference.What an over-forecast and an under-forecast cost in each decision, to see whether errors are symmetric.Who will be held responsible if the target is missed, so that the target can be worded safely before the project starts.