Engineering / InfrastructureLong (years)Detectability: Moderate

Fixed-interval offshore pipeline inspections

An operator standardized offshore inspection at a fixed 5-year cadence across a mature pipeline network.

Assumptions have half-lives.

Decision summary

Year
2017
Failure mode
Cadence lock-in: fixed schedules outlived the assumptions that justified them.
Silent failure window
~18 months: the system was drifting into a higher-risk regime, but the next scheduled inspection was still “on plan.”

The original logic

The network’s historic degradation rates were slow, inspection resources were constrained, and a uniform cadence simplified planning, contracting, and regulatory reporting.

Key assumptions

  • Degradation drivers would remain stable (temperature, chemistry, flow regime, CP performance).
    Confidence at decision: High
    Expected lifetime: 3–5 years
  • Inspection technology and coverage would be “good enough” to catch meaningful changes at 5-year intervals.
    Confidence at decision: Medium
    Expected lifetime: 5 years
  • Operational changes (tie-ins, flow changes, inhibition programs) would be documented and reviewed centrally.
    Confidence at decision: Medium
    Expected lifetime: 2–3 years

What changed

Two incremental operational shifts compounded: a change in produced-water handling increased water cut variability, and a third-party tie-in introduced cyclic flow and temperature swings. Local teams adjusted inhibitors, but the changes were not escalated into the inspection plan.

Outcome

A localized internal corrosion cluster developed in a segment that had been low-risk historically, prompting an unplanned shutdown, expedited inspection campaign, and replacement spool installation.

Early warning signals (missed)

  • Rising inhibitor consumption and “small” dosing adjustments that became the new normal
  • Repeated pigging debris anomalies (wax/scale mix) with no change to risk ranking
  • Increasing variance in temperature/flow data at the segment inlet

How AssureAI would have helped

  • Assumption owners + validation cadence: “stability of degradation drivers” would require periodic confirmation.
  • Signals tied to assumptions: variance in operating envelope becomes a tracked drift signal, not just a trend chart.
  • Decision record diffing: operational changes automatically prompt a “review due” event on the inspection decision.

Non-obvious lessons

  • Uniformity is an operational convenience, not a risk argument.
  • Small parameter changes matter when they persist.
  • Silence is not stability—especially in mature systems with accumulating modifications.
Fixed-interval offshore pipeline inspections — Decision Graveyard