This blog article is part of our Peak Event Operating System Series (3 parts)
You’re reading: Part 2 — Same Passenger Numbers. Totally Different Day. Here’s Why.
Previous: Part 1 — Your Next Peak Event Is Already on the Calendar—So Why Does It Still Hurt?
Next: Part 3 — Peak Days Don’t Fail All at Once—They Fail in This Order.
In Part 1, we built the foundation: a peak event readiness playbook with baseline plans, trigger thresholds, decision roles, and a rehearse → run → review loop.
Now we tackle the reason peak days still go sideways even when teams plan hard:
Most peak planning is still based on headcount.
But peak events aren’t just “more passengers.” They often bring different passengers—and that changes everything in airport operations.
Two days can have the same total passenger numbers and still feel completely different in the terminal. Airport planning must take into account how queues may build in different locations, processing times can shift, staff end up stepping in more often, baggage volumes can spike, and knock-on effects show up sooner and spread further.
That’s because the real driver isn’t only volume. It’s behavior: how people arrive, move, carry baggage, and interact with the process.
Headcount planning tends to assume the “average passenger.” Peak events are rarely average.
During peak travel periods, the passenger mix often shifts toward more families and first-time flyers, more group arrivals, heavier baggage and oversize items, greater special assistance demand, and tighter connection banks.
The result: throughput changes, dwell patterns change, and bottlenecks appear in unexpected places.
You don’t need to predict emotions. You need to forecast how passenger types consume capacity.
Passenger behavior forecasting predicts how different passenger types will use airport resources during a peak event. It focuses on: arrival timing, dwell time, processing time, baggage intensity, and where friction points form.
Wave shape is the time profile of demand (hour-by-hour arrivals to check-in/security/boarding), not the daily total. Two days can have the same total passengers but very different wave shapes.
Friction points are steps where flow slows due to decisions, checks, or constraints. Common friction points: document checks, bag drop, security divestment, boarding, bussing, transfer screening.
Resource multipliers are passenger segments that consume more capacity per person. Examples: families, group arrivals, first-time flyers, heavy baggage passengers, special assistance travelers.
Knock-on effects are secondary failures caused by a primary bottleneck. Example: check-in congestion → late arrivals at security → compressed boarding → missed turnaround targets.
A practical way to model peak events is to “shape the day” before it happens. Instead of anchoring on daily passenger totals, you describe how demand arrives over time, where flow is likely to slow, which passenger segments will consume more capacity per person, and how one bottleneck can trigger another. That shift matters because peak-day performance is usually decided by timing and concentration, not by totals.
Peak events change the type of demand the airport is serving. Heavier baggage tends to load check-in, bag drop, and baggage systems; more families and first-time flyers increase the number of interactions and slowdowns at security and decision points; group arrivals create short, intense pulses at curbside and processing; and full flights reduce recovery options when anything drifts.
The goal of this translation is simple: turn “what kind of day is it?” into “where will pressure build first, and what will it look like?”
On peak days, queues rarely grow linearly. They accelerate when demand arrives in tighter waves or when processing speed drops slightly at a friction point—and because there’s less slack, the system stops self-correcting. That’s why the most useful early signal is often queue growth rate rather than queue length: by the time a queue looks visibly bad, you’ve usually missed the easiest window to intervene.
Part 1 gives you the structure; behavioral forecasting makes it realistic. When you understand wave shape and resource multipliers, your baseline plan matches how the day will actually load the system. When your triggers focus on friction points, you act earlier than “flight delays.” And when your rehearsal mirrors the most likely failure mode for that event, the playbook becomes something the team can run under pressure.
Peak Event Playbook Series:
- Part 1: Your Next Peak Event Is Already on the Calendar—So Why Does It Still Hurt?
- Part 2: Same Passenger Numbers. Totally Different Day. Here’s Why.
- Part 3: Peak Days Don’t Fail All at Once—They Fail in This Order