S03 · stakeholder simulation

Three stakeholders and one overbooking limit

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

5stages
40 minabout
0stages committed

The brief

The synthetic airline wants a booking limit (how many tickets to sell) for a 180-seat domestic flight, because some booked passengers do not show up. You own the analysis. Three stakeholders, all simulated, with simulated opinions and numbers:

Head of Revenue (commercial): "Last year 10% of passengers were no-shows. Sell 195 tickets; empty seats are lost revenue."
Finance business partner: "Denied-boarding costs are above budget. Keep expected involuntary denied boardings at or below 0.25 per departure."
Platform lead (engineering): "Until Q3 we can only publish limits from the nightly batch. No intraday re-scoring."

Hypothetical economics for this exercise: average fare 90, fares non-refundable (the airline keeps a no-show's fare), and each denied boarding costs 600 (refund, compensation, rebooking and goodwill combined).

Stage 1 of 5

You have 15 minutes with the three stakeholders before you start the analysis.

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

What does one denied boarding cost in total, who owns that number, and does it vary by route or season?Has the no-show rate been stable, or did fare rules, fees or check-in change?When must limits be published, and can they change after publication?Can we guarantee zero denied boardings while still overbooking, so that no passenger is ever bumped?Which neural network architecture would the stakeholders prefer for the no-show model?