SIRE 2.0: what the transition actually changes for tanker operators

SIRE 2.0 went live on 2 September 2024 and fully replaced VIQ7 as OCIMF’s tanker inspection regime. Nearly two years on, the transition question is no longer “when do we switch” — every operator has switched — but something harder: which parts of the old vetting playbook silently stopped working, and what has to replace them.

The honest answer is that most of it stopped working. VIQ7 rewarded a specific organisational skill: the two-week preparation sprint before a booked inspection, run from a known questionnaire, rehearsed with the crew, closed out with fresh paint and fresh paperwork. SIRE 2.0 was explicitly designed to make that skill worthless. Understanding how it does so is the difference between operators who have adapted and operators who are collecting negative observations while still preparing for the old exam.

Scope. This is an independent guide to what SIRE 2.0 changes for operators. It is not an OCIMF publication, carries no OCIMF endorsement, and does not reproduce the CVIQ question library — that material is OCIMF’s and is available through its own channels. Nothing here replaces the current OCIMF documentation; where the two differ, OCIMF governs.

From a fixed questionnaire to a compiled inspection

The structural change sits in one acronym: CVIQ, the Compiled Vessel Inspection Questionnaire. Under VIQ7, every tanker faced the same published question set. Under SIRE 2.0, an algorithm — OCIMF calls it the Inspection Compiler — assembles a unique questionnaire for each inspection, drawing roughly 100 vessel-specific questions from a large question library and taking the ship’s own inspection history into account. No two inspections are alike, and the selection is deliberately non-predictable.

The library splits into four question types, and the split is worth internalising because it defines what can and cannot be prepared for:

• Core questions cover risks that can lead to catastrophic or severe outcomes. They appear in every inspection and make up roughly half of each CVIQ.
• Rotational questions cycle through in an unpredictable pattern — one tier appearing around every third or fourth inspection, a second tier around every sixth — so that the full library is covered over time without ever being fully present at once.
• Campaign questions are deployed fleet-wide for a period when OCIMF wants data on a specific topic, typically reacting to incident trends.
• Conditional questions are triggered by the vessel’s own particulars and the operator’s declared data.

Read that structure as an operator and the conclusion writes itself: the only way to be prepared for a CVIQ is to be prepared for the entire library, all the time. Cramming is not harder under SIRE 2.0 — it is mathematically pointless.

The human factor is now scored, not assumed

The second change is the one crews feel most directly. Every SIRE 2.0 question is answered through up to three response tools — hardware, process and human — and each can generate observations independently. A pump can be in perfect order (hardware), the procedure for operating it can be accurate and accessible (process), and the inspection can still record a negative observation if the officer operating it does not demonstrate understanding of the safety-critical steps (human).

Crew responses are graded against expectation — from exceeding expectations through “as expected” and “largely as expected” down to “not as expected.” And here SIRE 2.0 does something genuinely new: when a crew member’s performance is graded negatively, the inspector must identify which of nine performance influencing factors (PIFs) contributed. The nine cover recognition of safety criticality, custom and practice around procedures, procedure accessibility and accuracy, team dynamics and communications, stress, workload and fatigue, morale and motivation, workplace ergonomics, human–machine interface, and opportunities to learn and practise.

This matters for two reasons. First, it changes what an inspection observes: OCIMF’s guidance directs inspectors to watch normal daily activities rather than request staged demonstrations, which means a rehearsed script is not just useless but visible. Second, it changes what the data says about the operator. A pattern of “not as expected” gradings linked to the fatigue PIF across a fleet is no longer an anecdote — it is a queryable signal in a database that charterers screen against. The human element has become fleet-level, structured vetting data.

Evidence became infrastructure

The third change is the least discussed and, in our view, the most consequential: SIRE 2.0 runs on standing evidence, not on documents produced for the occasion.

Before an inspection can even be booked, the operator must have completed the Harmonised Vessel Particulars Questionnaire (HVPQ) for permanent vessel data, the Pre-Inspection Questionnaire (PIQ) for recent operational history, and an electronic certificate repository. During the inspection itself, the inspector works on a dedicated intrinsically safe tablet that operates offline on board, with a built-in camera and voice recorder under strict usage rules; photographs — whether supplied by the operator beforehand or captured by the inspector — go into the final report.

And around all of it sits the standing photographic obligation: a defined photoset of the vessel — safety-critical equipment, cargo areas, accommodation, engine spaces — refreshed approximately every six months. That photoset is not decoration. It is the objective condition baseline against which an inspector’s observations sit, and over successive cycles it becomes something the industry has never had in vetting before: a dated, standardised visual history of every tanker in the programme.

The scale is already substantial — OCIMF reports its database holding over 22,500 SIRE 2.0 inspection reports from more than 8,000 vessels inspected within the last twelve months. Vetting is no longer a snapshot business.

What this means day-to-day

Put the three changes together and the operational consequences fall out fairly cleanly:

Vetting became a workflow, not an event. If the question set is unpredictable, the photoset is due every six months, and the questionnaires must be current before booking, then “vetting readiness” is a permanent operating condition. Operators who have adapted run it like planned maintenance: owned, scheduled, audited.

The crew is the inspection. Half the exposure now runs through human responses observed during normal work. That shifts preparation from document review toward genuine familiarisation — officers and ratings knowing why a procedure exists, not what to say about it. It also makes the soft factors (rest hours, workload, morale) hard vetting variables, because the PIF framework records them as contributing causes.

Observation response quality compounds. Because the compiler sees the vessel’s history, a poorly closed-out observation is not buried in a filing cabinet — it shapes future question selection and is visible to every screening charterer. The corrective-action loop, long treated as paperwork, now has direct commercial weight.

Where “SIRE 2.0 CBT” fits — and what it cannot replace

A large share of the crews searching around this regime are looking for computer-based training. The place of CBT in SIRE 2.0 is specific, and worth stating plainly. Officers serving on tankers subject to SIRE inspections are expected to be trained for the regime, and CBT modules — OCIMF’s own training material on the human-factors approach, plus the commercial libraries built around the question structure — are an efficient way to deliver the why: how the compiled inspection works, what the three response tools are, how PIFs are recorded.

What CBT cannot deliver is the thing SIRE 2.0 actually scores: behaviour observed during real work. A certificate of module completion proves exposure, not competence — and under this regime, watching a senior officer operate equipment does not count as familiarisation either. What stands up is the sequence: CBT for the framework, then structured onboard familiarisation on the specific equipment and procedures, documented with dates, participants and task-level evidence, tested by whether the officer can explain the why under questioning during normal duties. Treat CBT as the entry ticket to preparation step 3 below, not as the preparation.

A preparation sequence that works

For operators still restructuring around the new regime, the sequence we see working in practice — and for the shipboard leg of it, walk the vessel area by area, across hardware, process and human factors, with the SIRE 2.0 inspection preparation checklist:

1. Put the photoset on the PMS calendar. Six-monthly, owned by a named role, executed to the defined standard — treated with the same discipline as a class survey window.
2. Keep HVPQ, PIQ and certificates permanently current, not booking-triggered. A stale PIQ is a self-reported observation.
3. Train against the library, not the last inspection. Rotate the full question library through onboard familiarisation so that every applicable topic is touched between inspections.
4. Run internal inspections in SIRE 2.0 format — three response tools, expectation-based grading, PIFs recorded — so the crew’s first contact with the method is not the real thing.
5. Treat every negative observation as a data point you will meet again, because you will: in the compiler, and in the charterer’s screen.
6. Watch your own patterns. If your internal data shows the same PIF recurring across ships, that is the finding — fix the system, not the ship.

The bigger signal

Step back from the mechanics and SIRE 2.0 is a preview of where all maritime assurance is heading: away from periodic, document-centred events and toward continuous, structured, evidence-backed records — photographs on a fixed cadence, human performance as recorded data, history feeding the next assessment. Tanker vetting got there first because charterers had the leverage to force it. The operators treating this as the template, and building their record-keeping accordingly, will find the next regime — whatever flag, class or charterer it comes from — already familiar.