Evidence library

What a digital maintenance system can — and cannot — tell you about human performance

By Georgios Zografos · Published

A planned maintenance system holds records of work, not observations of people. That limit is worth stating precisely, because the distance between 'this pattern of work is heavy' and 'this person is impaired' is where most human-factor software overclaims.

What independent evidence establishes

Published, citable sources. These are not Full Ahead claims.

  • IMO's fatigue guidance identifies workload, and maintenance and inspection specifically, among the factors relevant to fatigue at sea.
  • Fatigue guidance and rest-hour instruments address declared rest and fitness for duty. They do not define a measurement of individual performance derived from maintenance software.
  • No published instrument authorises a maintenance system to declare a seafarer unfit, impaired or fatigued.

What Full Ahead observes and records

Product behaviour verified in the running system.

  • What a maintenance record actually holds is: a job, its due position, an action, an acting account and role, and timestamps.
  • Where duration appears in Full Ahead it is session wall-clock duration — the elapsed interval between opening and closing a job record in the app. That is not man-hours, labour hours, wrench time, actual task duration, or proof of labour consumed, and the platform holds no field for those.
  • The current model holds no biometric, physiological, wearable, camera or medical fields. There is nothing to wear and nothing filmed.
  • Output is aggregated to vessel and role. Individuals are not scored or ranked.

What Full Ahead does not currently establish

Stated plainly, so nothing here is read as more than it is.

  • A maintenance system cannot observe sleep, alertness, health, stress or competence, and Full Ahead does not claim to.
  • Full Ahead does not claim that its work-pattern figures predict human error or degraded judgement, and does not claim a lead time.
  • Absence of biometric data describes what the product handles. It is not a legal conclusion about an operator's data-protection obligations, and it does not make the model medically or scientifically validated.
  • One signal in the wider model — a deferral event — has no producer in the current workflow and is not captured. Nothing presented publicly treats it as live evidence, and no capture workflow has been added merely to make a claim true.

What a technical manager can practically examine

Questions that can be answered from the maintenance record already held, with or without this product.

  • Ask of any human-factor claim in a software demonstration: which stored field produces this number, and what does that field literally record?
  • Ask whether a duration figure is wall-clock elapsed time in an application session or a measurement of work performed. They are different things.
  • Ask whether output is aggregated or attributed to a named person, and what happens to that output in a dispute.
  • Ask what the vendor states it has validated on live vessels, and ask for the evidence rather than the adjective.

What the product can currently support

Capability that exists today.

  • Deterministic arithmetic over recorded timestamps, aggregated to vessel and role.
  • A stated sufficiency floor: at least 12 signals across at least 3 distinct days for a vessel, below which the platform reports insufficient data instead of a number.
  • Consent-gated capture per workspace, scoped by row-level security to the operator's own organisation.
  • A maintenance decision record protected against alteration.

What remains unvalidated

Open, undemonstrated, and not claimed.

  • The work-pattern model has not been validated against live-vessel outcomes. The signals recorded to date are seeded internal Full Ahead data.
  • No customer fleet has yet generated a fatigue or attention assessment through this product.
  • No accuracy, sensitivity or false-positive rate is published, because none has been established.

References

Each reference is listed with the specific proposition it supports. No peer-reviewed status is claimed for this page.

  1. MSC.1/Circ.1598 — Guidelines on Fatigue

    IMO Maritime Safety Committee, 2019

    Supports: MSC.1/Circ.1598 identifies workload and maintenance among the factors relevant to fatigue; it does not define a software-derived measure of individual performance.

  2. IMO — Fatigue

    International Maritime Organization

    Supports: IMO's fatigue page sets out the scope of the current guidelines.

  3. STCW Code, section A-VIII/1 — Fitness for duty

    IMO International Convention on Standards of Training, Certification and Watchkeeping

    Supports: Fitness for duty and minimum rest are addressed in STCW Code section A-VIII/1.

Related