Crew readiness: qualifications, permits, and hours

A turnaround puts several hundred people onto a unit that normally hosts a dozen, most of them contractors, many of them new to the site, working around the clock for weeks. The integrity of the work they produce depends on three facts about each of them at each moment: they are qualified for the task in front of them, they are permitted to be where they are, and they are rested enough to do it safely. Most sites can prove all three eventually, from a filing cabinet of certificates, a stack of canceled permits, and a contractor's timesheets. The question this paper addresses is whether you can prove them at dispatch time, when the answer still changes what happens.

Why crew readiness is a turnaround risk, not an HR chore

The consequences of getting readiness wrong land on the integrity program, not on the HR function. A thickness survey performed by a UT technician whose Level II qualification expired on day nine of the event is a survey you cannot rely on, and depending on your jurisdiction and your own written practice it may be one you cannot use at all. A confined space entry made without an attendant is a fatality exposure. A welder on the sixteenth consecutive twelve-hour shift is a rework exposure and a safety one.

Each of these is a record problem before it is a people problem. Certificates arrive as PDFs from contractors, permits are paper forms in a box at the unit boundary, and hours live in each contractor's payroll system. Nobody is looking at the three together for one person on one shift, and the person doing dispatch at 05:30 is working from a crew list and memory. Crew readiness is the practice of putting those three facts on one record, per person, and making that record the thing dispatch reads. The full crew picture is described on the intelliCREW page.

What qualifications must be current before a technician touches the asset?

Qualification is specific to the task, and the record should be too. A single "qualified: yes" flag is worthless; the record needs the certifying body, the scope or level, the certificate number, and the expiry date. The common set for an inspection and repair turnaround:

• Pressure equipment inspection. API 510 for pressure vessels, API 570 for piping, API 653 for tanks. These are individual certifications with recertification cycles, and the inspector of record for a vessel must hold the relevant one.
• NDT method qualification. Level I, II, or III by method (UT, RT, MT, PT, ET, and the tube techniques) under the employer's written practice, typically based on SNT-TC-1A or ANSI/ASNT CP-189. Note that this is an employer certification: a technician changing employers mid-event may need to be requalified under the new written practice.
• Welding inspection. A recognized welding inspector credential where welds are being visually accepted, plus procedure and performance qualification records for the welders themselves.
• Coating inspection. A recognized coating inspector credential for surface preparation and DFT acceptance.
• Access and safety. Rope access levels, confined space training for each role, site orientation, and any client-specific permits such as hot work authorization.
• Medical and fit-testing. Respirator fit tests and medical clearance where the work requires them.

The record has to be checked against the work, not against the person in general. A technician who is Level II UT and Level I RT can perform and evaluate UT thickness readings but can only assist on radiography. Dispatch that assigns by name rather than by qualification will get this wrong under time pressure.

What does OSHA 1910.146 require on a confined space entry permit?

Vessels, columns, tanks, and many exchangers are permit-required confined spaces, and the entry permit is the document that proves the entry was controlled. Under 29 CFR 1910.146 the permit must be completed before entry is authorized and must record:

• The space to be entered, the purpose of the entry, and the date and authorized duration.
• The authorized entrants (by name or by a tracking method that identifies who is inside), the attendants, and the entry supervisor, with the supervisor's signature authorizing the entry.
• The hazards of the space and the measures used to isolate the space and to eliminate or control the hazards: lockout, blinding, purging, ventilation, and so on.
• The acceptable entry conditions and the results of initial and periodic atmospheric tests, with the tester's initials and the time. The standard defines a hazardous atmosphere to include oxygen below 19.5 percent or above 23.5 percent, flammable gas above 10 percent of its lower flammable limit, and toxic contaminants above their permissible exposure limits.
• The rescue and emergency services available and the means of summoning them.
• The communication procedures between entrants and attendant, and the equipment provided (testing, ventilation, communications, PPE, rescue).
• Any other permits issued for work in the space, such as hot work.

The permit is canceled by the entry supervisor when the entry is complete or when a condition arises that is not allowed under the permit, and canceled permits are retained for at least one year for the annual program review. The three roles have distinct duties and distinct training: the attendant does not enter, the supervisor verifies conditions before authorizing, and the entrants know the hazards and the signs of exposure. For an inspection program the permit is also evidence: it establishes who was inside the vessel and when, which is the same question the inspection record answers.

How do you track hours and fatigue across a 24-hour turnaround?

Turnarounds run continuous shifts for weeks, and the people most at risk are the ones who are hardest to replace: the senior inspector, the only Level III on site, the lead welder. Fatigue is a quality risk (missed indications, misread gauges, dropped steps in a procedure) as much as a safety one.

A workable approach follows the site's fatigue risk management policy; in refining and petrochemicals API RP 755 is the usual reference for shift length, consecutive days, and rest between shifts, and many sites adopt equivalent limits for contractors. Whatever the limits are, the mechanics are the same:

• Record hours per person per day against the event, from the actual badge-in and badge-out or from a confirmed daily log, not from a planned roster.
• Run the consecutive-day and rolling-hours totals and make them visible at dispatch.
• Treat exceptions as approvals, recorded with the approver and the reason, so that an extended shift is a decision someone owns.
• Record unavailability (rest days, sickness, training, travel) on the same record, so dispatch is not calling people who cannot come.

The point is not the limit. It is that the person assigning tomorrow's work can see today's totals for every candidate before they assign.

How should dispatch match people to work packs?

Dispatch is where readiness becomes real. Given a work pack (a vessel internal, a piping thickness survey, a bundle tube inspection), the dispatcher needs to find people who hold the required qualification, hold current site and confined space training, are within their hours, and are not already committed. Done from memory and a crew list, this fails under load. Done from a record, it is a filter:

1. Start from the work pack's required qualifications and roles (for a vessel entry: inspector of record with API 510, UT Level II, attendant, entry supervisor).
2. Filter the crew to those whose relevant certificates are current through the planned date of the work, not just today.
3. Exclude anyone over the hours or consecutive-day limit, or marked unavailable.
4. Assign, and record the assignment as a dispatch so the office view of who is where matches the field.
5. Reassign when the plan moves, with the change recorded.

The dispatch record then becomes part of the inspection record's provenance: the readings on this vessel were taken by this technician, holding this qualification, on this permit, on this shift. That chain is what an auditor, or an incident investigator, will ask for.

What should the readiness record look like to an auditor?

Post-event, the questions are predictable. Who inspected this vessel, and were they certified at the time? Who entered this column on this date, and was a permit open? Did anyone exceed the site's hours limits, and who approved it? The record that answers each of these in minutes rather than days has a few properties:

• Per-person records for qualifications (body, scope, number, expiry), training, skills, and unavailability, kept by the site rather than reconstructed from contractor files after the fact.
• Permits tied to the asset and the people, with the entry supervisor's signature and the cancellation time, retained for at least the required year.
• Hours by person by day, with recorded exceptions.
• Dispatch history linking people to work packs and shifts.
• An audit trail that shows when each of these records was created or changed and by whom, so a certificate expiry that was edited after the event is visible as an edit.

Compliance evidence for the program as a whole belongs in intelliCOMPLY; the point here is that the readiness record is the raw material that evidence is made from.

How intelliSPEC handles it

intelliCREW carries the crew as first-class records: member, team, dispatch, unavailability, inspector qualification, skills, and training. Qualifications are records with their own fields rather than attachments, so a dispatch can be filtered on what a person holds and when it expires. intelliPROJECTS's crew and daily progress records tie crews to the event day by day, and delay events record the consequences when a crew was not available.

Permits are forms, and because the platform is metadata-driven a site builds its confined space entry permit in the form designer with the fields 1910.146 requires, including the atmospheric test entries and the entry supervisor's authorization. The signature widget records per-signer identity even on a shared tablet at the vessel manway, so the supervisor's authorization and the attendant's acknowledgment are individually attributable. The field app works fully offline with no time limit on the local store, which matters at a manway with no signal.

Every record change is captured in the hash-chained audit trail with actor, timestamp, and request context, and audit events are never auto-deleted, so a qualification expiry date edited after the fact is visible as an edit. Access to crew records is controlled through roles and node-level entitlement over the asset graph, and the platform's identity and access controls are described in the Trust Center.