how to mark propeller polishing in AI voyage fuel baselines?

Propeller polishing fuel baseline marker: Mark propeller polishing as a discrete, date-stamped maintenance event tied to the correct vessel, propeller asset, and voyage window so AI voyage fuel baselines attribute performance changes to the right cause.

How Propeller Polishing Is Applied

To prevent AI fuel analysis from learning the wrong driver, treat propeller polishing like a controlled maintenance input that is consistently recorded and then linked to the voyages that occur after the work is completed. In practice, fleet teams should capture the propeller polishing date, record the scope and location (which propeller(s), whether it was full polishing or spot work), and then define a baseline eligibility window that starts after completion and ends when the next relevant underwater service voyage comparison event occurs.

Use these operational rules to standardize how the event is marked in your voyage fuel baseline dataset:

Create a dedicated “underwater maintenance event” record with fields for vessel ID, propeller asset ID, completion timestamp, work type (polishing), and crew or contractor reference as applicable.
Link the event to voyage segments using a deterministic rule: assign the event to voyages whose start time is on or after the propeller polishing completion time, and exclude voyages that overlap the work window.
Store a “propulsion efficiency reset” flag and rationale only when the baseline model should treat propulsion performance as re-initialized due to the polishing and any associated adjustments (for example, if speed trials or acceptance checks were performed after the work).
Capture supporting metadata for model governance: coating or condition notes, measurement method (if any), and the underwater service voyage comparison context (for example, “pre-polish vs post-polish” periods) so analysts can validate that the model is not attributing unrelated changes to polishing.
Use an audit-friendly data model where the event is immutable after approval and amendments create a new version with change reason; this supports data confidence during AI readiness reviews.

Operational Impact

Better attribution and fewer false correlations: Technical Managers can trust that changes in fuel rate or propulsion performance are evaluated alongside the correct propeller polishing date, reducing the risk that the AI baseline blames polishing when the driver was another factor.
Improved maintenance discipline and repeatability: Fleet and technical teams can compare underwater service voyage comparison windows consistently across vessels, enabling more reliable KPI tracking for polishing effectiveness and timing.
Stronger AI readiness and governance: CIO and IT governance teams get clear event lineage, so propulsion efficiency reset logic and baseline eligibility windows are reproducible, reviewable, and defensible during model validation.

Important to know: Define one standard “baseline eligibility window” policy (for example, start at completion time and include only voyages fully after completion) and enforce it in master data validation before AI fuel baselines are refreshed; otherwise, overlapping voyage timestamps will contaminate the dataset and reduce data confidence.