What your thickness history has to prove before scope freeze
Turnaround scope is decided months before the window opens, from data nobody can re-take. A thickness history has to distinguish three states that look identical in a spreadsheet: measured and healthy, measured and thinning, and never measured at all. Every retained reading carries its date, technician, instrument and method; every missing reading carries the reason it is missing, because that reason is scope.
Chemical plants inspect differently from refineries. The asset base is many small code-stamped vessels and short alloy circuits rather than a few long carbon steel trains, and corrosion allowance is often near zero — a 3 mm Hastelloy C-276 reactor shell has no sacrificial wall, so a gauge accuracy of ±0.10 mm is a trivial fraction of a 20 mm refinery line but a decisive fraction of what is left here. Chloride stress corrosion cracking of austenitic stainless under wet insulation, described in API 571 and addressed by AMPP SP0198, is not found by thinning trends at all; it is found by stripping insulation, which is scaffold and labour that must be bought before the window. Glass-lined and clad equipment cannot be measured conventionally on the lining. Each of those constraints turns an excluded reading into a purchase order, and the exclusion record is where the purchase order comes from.
Source: Sources: API 510 Pressure Vessel Inspection Code; API 570 Piping Inspection Code; API 571 Damage Mechanisms Affecting Fixed Equipment in the Refining Industry; API 579-1/ASME FFS-1 Fitness-For-Service; API 580 and API 581 Risk-Based Inspection; ASME Boiler and Pressure Vessel Code Section V (nondestructive examination) and Section VIII Division 1; ASME B31.3 Process Piping; ASME PCC-2 for repair of pressure equipment; AMPP/NACE SP0198 for control of corrosion under insulation; ASTM C692 and ASTM G189 for insulation-related stress corrosion cracking evaluation; ASNT SNT-TC-1A and ISO 9712 for personnel qualification; OSHA 29 CFR 1910.119 Process Safety Management.
| Recorded exclusion reason | What planning has to buy before the window | What happens if it stays a blank cell |
|---|---|---|
| Insulation not removed at this CML | Strip-and-reinstate order, scaffold, insulation materials, CUI inspection slot | The CML reads as inspected; CUI on austenitic stainless in the 120-350°F band goes unexamined for another cycle |
| Surface temperature above couplant rating | High-temperature couplant, delay line probes, or a cooldown slot in the shutdown sequence | Trend gaps concentrate on the hottest and most aggressive services in the plant |
| Confined space entry not authorised | Entry permit, atmosphere monitoring, attendant hours, chemical clean and neutralise before entry | Internal readings never accumulate; the vessel is assessed on external UT alone with no corroboration |
| Glass or PTFE lining present — UT not applicable | Spark testing, holiday detection, lining thickness measurement by an appropriate method | The vessel appears un-inspected in a UT-centric register even though the correct method exists |
| Scaffold dismantled before crew arrived | Sequencing correction; scaffold hold or a second erection | Repeat cost every campaign, invisible because it never appears as a line item |
| CML could not be located on the equipment | Register correction, re-establish and photograph the CML with reference dimensions | A phantom CML is carried forward indefinitely and its coverage is counted |
| Reading rejected on review as implausible | Re-measurement while access still exists, before demobilisation | A wrong number enters the trend, or the point silently disappears from it |
Scope freeze is a data deadline that arrives before anyone feels ready
Most chemical sites run turnaround governance on a T-minus calendar: scope submission opens twelve to eighteen months out, scope freezes somewhere around six months, and after freeze new work enters only through a change process that everyone would rather avoid. The window itself is fixed by production commitments and cannot move. Everything downstream of scope freeze — long-lead materials, contractor mobilisation, scaffold design, permit volume, shutdown sequence — is calculated from the frozen scope list.
What is rarely acknowledged is that scope freeze is really a data deadline. The list is an assertion about the condition of several hundred pieces of equipment, and the evidence for that assertion is whatever the thickness history contains on the day of freeze. No new evidence can be created: the plant is running, the equipment is insulated and hot, and the locations you most want to know about are the ones you cannot reach. Whatever the record says at T-6 is what the decision is made on.
This is why a chemical plant's thickness history has a different job than a refinery's. A refinery integrity engineer is mostly trying to compute a rate. A chemical plant turnaround planner is mostly trying to compute confidence: how much of this unit do I actually have data for, how old is it, who took it, and where are the holes. A module that answers the rate question beautifully and the coverage question not at all is the wrong tool for the buyer at T-6.
Absent and excluded look identical, and only one of them is safe
Open any inspection workbook at a chemical site and you will find blank cells. Some of them are locations that were read and found healthy but not transcribed. Some are locations that were physically unreachable. Some are locations that no longer exist because the circuit was replaced in a previous outage and nobody updated the register. Some are locations the last contractor could not find. From the spreadsheet, all four are the same absence.
The consequence is asymmetric and always in the unsafe direction. A location with no reading contributes nothing to the corrosion rate, so the circuit's calculated rate is driven entirely by the locations that were readable — which are, systematically, the accessible, cool, uninsulated ones. The inaccessible, hot, insulated locations are also the ones where corrosion under insulation, dead-leg accumulation and localised attack concentrate. The data set is not merely incomplete; it is biased towards the healthy portion of the equipment.
Making exclusion a first-class record fixes this cheaply. The CML exists in the register with a scheduled campaign; the campaign closes with either a value or a coded reason. The reason has an owner and a resolution path. Coverage is then computable per circuit and per campaign — readings taken against readings scheduled — and a circuit whose coverage has been sixty percent for three cycles running is visible as a risk rather than as a tidy trend line drawn through the easy points.
Small vessels, thin alloys and a corrosion allowance that was never there
Chemical manufacturing asset registers look nothing like refinery ones. Instead of a handful of large carbon steel trains, there are dozens or hundreds of comparatively small code-stamped vessels: batch and semi-batch reactors, crystallisers, neutralisers, filters, dryers, scrubbers, and short runs of connecting piping in materials chosen for one specific chemistry. Alloy 20, Hastelloy C-276 and B-3, titanium, zirconium and tantalum appear where carbon steel would be consumed in weeks.
Those materials are selected precisely because they are expected not to corrode, and they are therefore specified thin. A vessel wall sized by pressure design with negligible added corrosion allowance has no margin to give away. This inverts the usual relationship between measurement uncertainty and decision-making. On a heavy-wall carbon steel line, a tenth of a millimetre of gauge and placement uncertainty is noise. On a 3 mm alloy shell, it is a material share of the wall you are arguing about, and it will decide whether a vessel goes on the scope list or off it.
Everything follows from that. Probe placement has to be repeatable, which means CMLs need physical reference dimensions and photographs, not just numbers. Surface preparation has to be recorded, because a reading through product residue or an oxide layer is not comparable with one on prepared metal. Material of construction has to be on the CML, not just on the vessel, because clad and dissimilar-material sections have different minimum thicknesses and a single vessel-level minimum applied across all its CMLs will be wrong somewhere.
Access is the cost, and the exclusion reason is the estimate
In a chemical turnaround the inspection labour is rarely the expensive part. The expensive parts are scaffold, insulation strip and reinstate, chemical cleaning and neutralisation, isolation and blinding, confined space entry provision, and the schedule float consumed by all of it. A single vessel internal inspection may require the vessel to be emptied, washed, neutralised, purged, gas-tested, blinded and entered under permit with an attendant, before a technician takes a single reading.
This means the exclusion reasons recorded in one campaign are the access estimate for the next one. Forty-one CMLs excluded for insulation not removed is not a data-quality footnote; it is a strip-and-reinstate package with a materials list and a labour estimate. Nine CMLs excluded for surface temperature is a consumables and procedure decision. Six excluded for permit unavailability is a shutdown sequencing problem that will recur unless the sequence changes. Aggregated by unit and by reason, these numbers land on the planner's desk at T-12 instead of arriving as discovery work at T-plus-three-days.
The reverse is also true and is where sites lose the most money. Because uncoded absences are invisible, the same access failure repeats every campaign, and the cost is never attributed to it. Nobody buys the scaffold because nobody has ever seen a number that says the scaffold is missing. A history module that cannot roll exclusion reasons up into a planning report is not participating in the process that actually determines turnaround cost.
The equipment a thickness gauge cannot help you with
Chemical plants contain a substantial population of equipment where ultrasonic thickness measurement of the process-side surface is meaningless or impossible. Glass-lined reactors are the clearest case: the integrity question is the continuity of the lining, assessed by spark testing and visual examination, and the steel substrate thickness tells you almost nothing about whether the vessel can hold the next batch. PTFE and rubber-lined vessels, refractory-lined equipment, and clad or weld-overlaid vessels each have their own examination logic.
The failure mode is quiet. In a register organised around thickness readings, these vessels accumulate no data, which makes them look like coverage gaps to some readers and like non-applicable entries to others. Either way, the actual question — was the lining examined, by whom, when, and what was found — sits outside the system in a separate report. The fix is to make the required examination method an attribute of the location, so that a glass-lined reactor with no spark test in three years raises the same flag as a carbon steel circuit with no thickness reading in three years.
Corrosion under insulation deserves the same treatment. It is not a thinning trend; it is an external attack that concentrates in the moisture-retaining temperature band, at penetrations, at damaged cladding, at supports and at low points. AMPP/NACE SP0198 frames it as a design, coating and insulation-system problem as much as an inspection problem. In the history, a CUI inspection is an event with a location, an extent of insulation removed, a coating condition finding and a photograph — none of which is a thickness value, and all of which belongs in the same record so that the vessel's story is in one place at scope freeze.
Freezing a version you can be audited against
Every turnaround produces at least one uncomfortable retrospective conversation: why was this vessel not in scope, or why did we spend three shifts on a vessel that turned out to be fine. Both conversations are answerable only if you can reconstruct the dataset as it stood at the moment the decision was made. If readings have been added, corrected and re-baselined since, the reconstruction is guesswork and the discussion becomes about people rather than about information.
A frozen version is a small feature with a large effect. At scope freeze, the system takes an immutable snapshot: every reading, every rate, every exclusion, every coverage figure as of that date and time. Later work is additive on top. When the retrospective happens, the snapshot answers whether the decision was reasonable on what was known, which is the only fair standard. It also answers the more useful forward-looking question: what did we learn in the window that we could have known before it, and which of those gaps is a repeatable process fix.
This matters beyond the turnaround itself. Under a process safety management regime, mechanical integrity decisions and their basis are auditable, and deficiencies identified must be corrected before further use or otherwise managed in a documented, safe manner. A deferral decision taken at scope freeze is exactly that kind of documented judgement. Being able to produce the frozen dataset, the exclusion list and the reasoning as a single package turns an audit finding into a five-minute retrieval.
Evaluating the module against your own T-minus calendar
Do not evaluate this software on a trend chart. Evaluate it against the questions your turnaround process asks on specific dates. At T-12, can it produce coverage by circuit and unit for the last three campaigns, with exclusion reasons aggregated into an access scope you could hand to a scaffold contractor? At T-6, can it freeze and label a version, and later show what changed after that version? Inside the window, can a field crew close an exclusion on a tablet the moment access is granted, so the record is right without a back-office pass?
Test the ingestion path with your own material, not a sample dataset. Chemical sites accumulate history in vendor-specific gauge exports, contractor workbooks, legacy databases and PDF reports, often with different unit conventions and different CML naming across units acquired at different times. Ask the vendor to import two of your worst files live, and watch whether unmatched CMLs are quarantined for review or silently created as new locations. Silent creation is how a register doubles in size and loses its history.
Finally, check that the location can carry more than a number. A CML on a glass-lined reactor should be able to hold a spark test result. A CML on an insulated stainless line should be able to hold a CUI inspection with photographs and an insulation-removed extent. If the data model only accepts thickness, everything else in your integrity programme will keep living in a shared drive, and the scope freeze conversation will keep happening across two systems. To walk through your own turnaround calendar and see how the exclusion data would roll up, request a consultation at info@atlantisndt.com.
Why does turnaround readiness depend on the thickness history rather than the plan?
Because scope is an argument about condition, and condition is what the history holds. At scope freeze, every proposed vessel entry, every circuit replacement and every deferral has to be justified from measured wall, measured rate and measured coverage. If the history cannot show which locations were actually read in the last two campaigns, the planner cannot defend either including a job or leaving it out, and the default becomes a discovery budget that consumes the window.
How does a near-zero corrosion allowance change what the data has to do?
It removes the margin that absorbs measurement error. A 20 mm carbon steel line with 3 mm allowance tolerates a 0.1 mm gauge uncertainty without consequence. A 3 mm alloy reactor shell designed with essentially no sacrificial wall does not: the same uncertainty is a meaningful share of what remains, so repeatability of probe placement, surface preparation and instrument setup stops being good practice and becomes the thing the remaining-life number rests on.
What is the difference between an excluded reading and a blank cell?
A blank cell is ambiguous and therefore treated as benign. An excluded reading is a decision with an owner, a reason code and a date, and it converts directly into work: a scaffold, a strip order, a permit, a different examination method. Chemical plants live or die on this distinction because access, not inspection labour, is the dominant cost, and access has to be purchased before the window rather than discovered inside it.
Which chemical-service damage mechanisms will a thickness trend never find?
Chloride stress corrosion cracking of austenitic stainless under insulation, caustic stress corrosion cracking, hydrogen blistering and HIC, embrittlement, and cracking of glass or PTFE linings all progress without measurable general wall loss. A history that only stores thickness quietly implies these were checked. Storing method against every examination — and letting a location carry a required method it has never received — is how the gap becomes visible before the window rather than after a leak.
How should the data be locked at scope freeze?
Snapshot it. The version of the dataset that scope was frozen against should be immutable and retrievable, with later readings added on top rather than overwriting it. When a job is challenged mid-window, or reviewed afterwards, the question is always whether the decision was reasonable on the information available at the time. Without a frozen version, that question cannot be answered and every deferral is re-litigated with hindsight.
Can the same history support a Fitness-For-Service assessment?
It supports the input, not the conclusion. An API 579-1/ASME FFS-1 Level 1 or Level 2 assessment of local thin area needs the measurement grid geometry, spacing, the minimum measured thickness, the extent of the thinned region and the future corrosion allowance — which means the raw grid, not a single reported minimum. A history that stores only the lowest number per CML forces a re-inspection before any assessment can start.
Built for any business that runs on operations
Most companies do not fail at their craft. They lose time, margin and goodwill in the gaps between the tools they use to run the place — a quoting spreadsheet that does not talk to the job sheet, a job sheet that does not reach accounts, and a compliance folder nobody can search when a client asks. Atlantis closes those gaps by putting the whole operation on one platform, so information is entered once and everything downstream stays in step.
What you can run on it
- Sales and CRM — leads, quotes, follow-ups and the pipeline that tells you what next month looks like.
- Projects and job costing — plan the work, track the hours and materials against it, and see the margin while the job is still live rather than at final account.
- Field and service teams — dispatch, schedules, mobile capture that works with no signal, and sign-off from site.
- Inventory and purchasing — stock, suppliers, reorder points and goods receipt, joined to the jobs that consume them.
- People — records, qualifications and licences with renewal reminders, timesheets, leave and payroll.
- Quality and documents — procedures and forms under revision control, with the audit trail an inspection or accreditation body actually asks for.
- Accounts — invoicing, expenses, multi-currency and the reporting your accountant stops chasing you for.
Affordable, accessible, fully customizable — and we mean each word
Affordable because the whole suite is included rather than sold to you a module at a time, and because implementation is done by people who have run operations rather than by a chain of subcontractors. Accessible because it runs in a browser and on a phone, works for a small team on day one, and does not need a specialist on staff to keep it alive. Fully customizable because your process is the thing that makes you competitive — the software should bend to it, not the other way round.
Industries we configure for
Service businesses and contractors, manufacturing and fabrication, trading and distribution, laboratories and testing houses, engineering consultancies, construction and facilities, and asset owners across energy, marine, aerospace and infrastructure. Inspection and testing is where we started, and it remains the sector we go deepest in — but the platform underneath is general-purpose, and most of what it does has nothing to do with inspection at all.
What happens when you get in touch
A short conversation, not a sales sequence. We ask how the business runs today and where it hurts, show you the platform doing that work, and send a written quote shaped to your region, your team size and the scope you actually need. No obligation, nothing to install first, and no pressure to decide on the call. Reach out and tell us what you are trying to fix.
Related: business management platform · inspection management software · choosing the right category of software · modules · by industry · asset integrity platform. Book a free consultation.