How Receiving Windows Lose Executability Before Failure Becomes Visible

The Moment It Breaks

The shipment looked fine.

The schedule was stable. The window was open. Nothing had moved.

Then it rolled.

For most export teams, the sequence feels like:

Everything was on track. Then it wasn’t.

That conclusion is wrong.

Figure 1 - The Illusion of Sudden Failure

Failure appears sudden. The system leading to it is not.

The Window Was Never Stable

The standard assumption is that execution risk develops over time.

The data does not support that.

Rolled bookings don’t lose their window.

They start without one.

• Median initial receiving window (rolled): 82 hours
• Median initial receiving window (non-rolled): 120 hours
• 91.2% of rolled bookings begin under 120 hours

The margin is already constrained at the starting point.

Figure 2 - Structural Window Distribution

Execution risk is embedded at first observation - not accumulated.

The Shape of Decay

Execution does not decline evenly.

It follows a distinct pattern.

Phase 1 - False Stability

Everything appears intact.

Phase 2 - Hidden Decay

Small losses accumulate invisibly.

Phase 3 - Compression

The rate of change accelerates.

Phase 4 - Collapse

Recovery becomes impossible.

Figure 3 - Execution Decay Curve (Core Model)

Executability declines gradually, then collapses as the system enters the final 72-hour window.

The Signal Arrives After the Decay

The system does not warn early.

It confirms late.

• Median detection lag: 8.5 hours
• 40% detected with <24 hours remaining
• 7% detected after ERD

Detection does not precede failure.

It follows it.

Figure 4 - Detection Lag vs Remaining Time

By the time the signal is visible, recovery is already constrained.

The Receiving Window Is the System

Export teams operate using the loading window.

Execution is governed by the receiving window.

Loading window:

• ETD
• Vessel schedule
• Booking

Receiving window:

• ERD → CY Cut
• Terminal acceptance

They are not synchronized.

Figure 5 - Loading vs Receiving Window

Expectation is stable. Execution is not.

The Boundary Moves

The CY Cut is treated as fixed.

It isn’t.

• 31.2% of CY Cut changes occur after ERD
• 45% on rolled bookings

A boundary that moves is not a boundary.

Figure 6 - CY Cut Drift Timeline

The constraint moves after execution begins.

Risk Is Concentrated

Risk is not evenly distributed.

• ONEY: 15.7%
• MSCU: 0.3%

52× difference.

This is not randomness.

This is structure.

Figure 7 - Carrier Risk Spread

Execution risk is concentrated, not market-wide.

The Quiet Schedule Problem

At 5 days before ERD:

• 25.3% show ERD movement
• 36.6% show CY Cut movement

The rest appear stable.

They are not.

Figure 8 - Quiet vs Late Movement

Silence is not stability. It is latency.

Execution Decay Is the Mechanism

Execution does not fail gradually.

It fails when:

• the window is already narrow
• changes cluster
• detection lags
• decisions compress

Figure 9 - Execution Timeline Collapse

The signal arrives after the decision moment has passed.

The System Reveals Too Late

The exporter does not react too late.

The system reveals the problem too late.

Rolled bookings don’t feel sudden because they are unpredictable.

They feel sudden because: the system shows failure after it has already become inevitable

Failure is not when the window collapses. It’s when executability is already gone.