Building a steel plant outage scope that survives a fixed window
A steel plant outage is scoped after its window is already fixed by production and hot metal logistics. Inspection scheduling has to work backwards from that: crew-hours by method and certification, radiography treated as an exclusion resource rather than a task, permits and gas-freeing as predecessors, and post-weld heat treatment modelled as the long pole in every repair contingency.
Integrated steel does not stop the way a refinery stops. A blast furnace is banked rather than shut, stoves and refractory hold a temperature clock, and the caster, reheat furnaces and mills each have their own downturn rhythm — commonly a short weekly or fortnightly line stop, plus a small number of longer annual outages and, on a multi-year cadence, a reline. The date is set by order book and hot metal balance long before anyone knows what the inspection scope is. Everything after that is compression. Non-destructive examination of BOF or EAF shells, water-cooled panels, gas mains, ladle and torpedo shells, crane runway girders, back-up rolls and Section I boiler pressure parts all competes for the same hours, the same scaffold, the same isolations and the same small number of certified technicians. A scheduling module that cannot express those shared constraints will produce a plan that looks complete and cannot be executed.
Source: Sources: ASME Boiler and Pressure Vessel Code Section I for power boilers, Section V for non-destructive examination, Section VIII Division 1 including the UCS-56 post-weld heat treatment requirements, and Section IX for welding procedure and performance qualification; National Board Inspection Code NB-23 for repairs and alterations; ASME B31.1 for power piping; AWS D1.1 for structural steel repair welding; CMAA 70 for crane service classification and ASME B30.2 and B30.10 for overhead cranes and hooks; OSHA 29 CFR 1910.119 for covered process areas and 29 CFR 1910.1029 for coke oven emissions; NRC 10 CFR Part 34 and equivalent Agreement State regulations for industrial radiography.
| Outage scope item | What fixes its duration | Predecessor commonly missed | Effect when it is missed |
|---|---|---|---|
| Internal examination of gas mains, dust catcher and cleaning plant | Gas-freeing, purge and continuous atmosphere monitoring, not the examination itself | Nitrogen purge sequence and gas crew availability booked as a resource | Crew stands at the manway for a full shift; the examination slips a day and takes scaffold time with it |
| Ultrasonic and magnetic particle examination of BOF or EAF shell and trunnion ring | Cooling to a temperature where couplant and contact examination are valid | Cooldown curve treated as scheduled duration rather than as dead time | Technicians are mobilised for a shift they cannot work, and the shell scope is compressed into the back half of the window |
| Radiographic examination of boiler pressure parts and power piping welds | Exclusion boundary clearance, which idles every other trade inside the radius | Night-shift allocation and a documented boundary agreed with area owners | Radiography is repeatedly cancelled by day-shift activity and ends up on the critical path it was meant to avoid |
| Weld repair of a rejectable indication in a Section VIII component | Post-weld heat treatment heat-up rate, hold time and controlled cooling | A contingency chain from the examination task, with PWHT and re-examination attached | A single unexpected indication consumes more of the window than the entire inspection scope |
| Crane runway girder, hook and structural examination | Crane release from production and access equipment, not examination time | Interlock with hot metal movement schedule and a released area | The crane is never actually free; the examination is deferred to the next outage and the interval quietly lengthens |
| Refractory-related examination on stoves, ladles and torpedo cars | Refractory cooling and controlled reheat rates | Reheat curve booked as window hours belonging to operations, not maintenance | The outage overruns at the end, when the pressure to restart is highest and inspection findings are least welcome |
The window is fixed before anyone knows the scope
In most process industries the outage duration is negotiated against the work identified. In integrated steel it usually runs the other way. The downturn date and length come out of the production plan, the hot metal balance and the order book, and they are communicated to maintenance as a fact. The inspection scope is then developed inside a box whose dimensions were decided elsewhere, and the planner's real job is triage rather than estimation.
This is why steel plants overrun for a characteristic reason: not because the work was badly estimated, but because the plan was built as a list of tasks with durations rather than as a network of shared constraints. Twelve examinations that each take four hours do not take forty-eight hours if they need the same scaffold, the same isolation, the same crane and the same two certified technicians. A scheduling module has to make that collision visible while there is still time to do something about it.
The corollary is that scope must be classified by what happens if it does not get done. Work that is legally or jurisdictionally required, work that protects a known damage mechanism, work that is opportunistic because access is rarely available, and work that could wait — these are four different things, and a module that cannot sort the outage backlog into them forces the sorting to happen in a meeting at two in the morning.
A banked furnace is not a shut furnace
Short outages on the iron side rarely involve a cold furnace. The furnace is banked, stoves hold heat, and refractory temperature imposes a clock that no amount of planning goodwill can extend. That clock is the outer boundary of everything else, and it is not visible in a task list. It has to be modelled as a hard finish constraint that the whole downstream plan is scheduled against, with a defined point at which remaining work is abandoned rather than attempted.
Cooling and reheating are the second half of the same problem. Where a vessel, stove, ladle or duct does have to come to a temperature at which contact examination is valid, that cooldown consumes window hours and is frequently omitted from the plan because it is nobody's task. Technicians get mobilised for a shift in which the steel is still too hot for couplant. Controlled reheat afterwards, governed by refractory limits, does the same at the other end.
The practical fix is small and rarely implemented: represent cooldown and reheat as scheduled activities with owners, and schedule inspection crews against the window that remains. It moves an argument that normally happens during the outage to a planning meeting six weeks before it, which is the entire value proposition of outage scheduling software in this industry.
Radiography is a scheduling resource, not a task
Nothing else in an outage has radiography's ability to stop unrelated work. The exclusion boundary required for industrial radiography under 10 CFR Part 34 or the corresponding Agreement State regulation is a physical area from which everyone must be excluded, and in a congested plant that area routinely overlaps scaffolds, walkways and other crews. A single shot can idle far more labour than it consumes.
Treated as a two-hour task on a bar chart, radiography loses every conflict. It gets bumped by whatever is happening in the area, then bumped again, and finally executed at the end of the window when there is no time left for the repair that a rejectable indication would trigger. That is the sequence behind a surprising number of outage overruns, and it is entirely predictable.
The correct model is an exclusive resource with a footprint and a calendar, scheduled deliberately into night shifts or into planned area clearances, with the affected work explicitly linked so the plan shows what stops. Once a module can express that, the planner can also make an informed choice to substitute an alternative technique where the code and the joint configuration allow it, trading examination cost for window hours — which is the trade that actually matters.
The repair loop is the real critical path
An inspection scope is not the risk to the window; the findings are. The chain from a rejectable indication to a returned-to-service component runs through excavation, re-examination of the excavation, welding to a procedure qualified under Section IX, any required post-weld heat treatment, and final examination. On carbon steel pressure components, the post-weld heat treatment requirements in Section VIII Division 1 mean a hold at temperature proportional to thickness, on top of controlled heat-up and cooling. That is shifts, not hours.
Where the component is jurisdictional, the National Board Inspection Code governs the repair or alteration, which brings an authorised inspector into the sequence along with documentation that must be complete before restart. Scheduling this after the fact means discovering on the last night of the outage that the paperwork, not the welding, is holding the plant down.
The countermeasure is to plan the contingency before the outage. For each examination with a realistic chance of a finding, attach the repair chain as a modelled branch with its own resources and duration. The planner does not have to believe every branch will fire; the value is in knowing how many can fire concurrently before the restart date moves. A module that cannot hold contingent work as anything other than an emergency work order cannot support that conversation.
Crew-hours by method, not tasks on a bar chart
Resident NDT capability at a steel plant is sized for normal operation, which means outage demand can be several times the standing crew. The gap is filled by contract technicians mobilised for the window, and the binding constraint is not people but certified people: Level II ultrasonic for shell and weld work, magnetic particle for structural and crane items, radiography crews with their own licensing and source logistics, and Level III oversight for technique approval and interpretation disputes.
Expressing the scope in crew-hours by method turns a vague worry into a number. If the plan needs four hundred ultrasonic crew-hours across a five-day window and the mobilisation covers two hundred and forty, that is visible six weeks out, when it can still be solved by adding a crew, resequencing or cutting scope. Expressed as tasks, it is visible on day two of the outage, when it can only be solved by cutting scope.
Certification currency belongs in the same calculation. A technician whose method certification or annual vision check expires mid-outage is not available for the whole window, and discovering that at the gate is worse than not having booked them. The scheduling module should refuse to allocate work to a technician whose qualification lapses before the planned finish, and should show the planner why.
Gas, isolation and permits as predecessors
Blast furnace gas is toxic and combustible, and by volume it is roughly a fifth to a quarter carbon monoxide. Any internal examination of gas mains, dust catchers, cleaning plant, holders or downstream burners depends on isolation, purge, gas testing and a permit regime that involves a dedicated gas crew. Coke oven areas add their own regulated exposure controls, and by-product and gas-handling areas may fall under process safety management requirements depending on inventories.
None of this is paperwork in the scheduling sense. It is a predecessor with a duration and a scarce resource attached. If the permit and isolation system lives entirely apart from the inspection schedule, planners will publish start times that the permit office cannot honour, and the first two hours of every shift will be spent discovering that. Linking the isolation and permit request to the inspection work item, with its own lead time, removes an entire class of daily delay.
Confined space entry, hot work adjacent to combustible dust, and simultaneous operations control all behave the same way. The test for a scheduling module is whether it can hold these as linked prerequisites with owners and lead times, so the readiness report says not only what is planned but what is permitted to start.
Readiness gates and how to evaluate the module
The most useful artefact a scheduling module can produce for a steel outage is a readiness percentage at fixed milestones — six weeks, four weeks, two weeks and the day before. Readiness is not a feeling. It is the proportion of the scope for which access is confirmed, scaffold or access equipment is requested, isolation and permits are identified, previous inspection data and drawings are attached, and a certified technician is allocated. Scope that fails any of those at two weeks is scope that will not be executed well.
When evaluating vendors, ask for that report against a real scope, not a demonstration dataset. Then ask three specific things: can radiography be modelled as an exclusion resource that visibly stops other work; can a repair contingency including post-weld heat treatment be attached to an examination in advance; and can the plan be expressed and re-levelled in crew-hours by method and certification. A module that does all three will change the outcome of the next outage. A module that does none of them is a calendar with a good user interface.
Atlantis NDT builds this layer on an Odoo foundation, so the outage plan, the contractor mobilisation, the certification register, the examination records and the repair documentation are the same dataset rather than five systems arguing. If you want to see it against your own downturn scope and mobilisation plan, a working session with an ASNT Level III is available on request through info@atlantisndt.com.
Why does a steel outage scope have to be built backwards from the window?
Because the window is not negotiable. It is set by order book, hot metal balance and the cost of lost production per hour, and it is often fixed months before the inspection scope exists. That inverts normal planning: instead of asking how long the work takes, the planner asks what fits, in what order, under which shared constraints. A scheduling module that only sums task durations cannot answer that question.
Why is radiography treated as a resource rather than a task?
Because a radiographic shot clears an area. In a congested plant the exclusion boundary required under 10 CFR Part 34 or the equivalent state rule can idle dozens of people across several trades. The correct model is a shared, exclusive resource with a geographic footprint and a time slot, almost always on night shift. Scheduled as an ordinary task with a two-hour duration, it will be bumped repeatedly and end up controlling the outage.
How should post-weld heat treatment be represented in the schedule?
As a contingency chain hanging off the examination that might find the defect, not as a separately planned job. Excavation, welding to a qualified procedure, PWHT with its heat-up rate, hold at temperature and controlled cooling, then re-examination, is a sequence measured in shifts. Modelling it in advance lets a planner see immediately how many concurrent repairs the window can absorb before the restart date is at risk.
What does scheduling in crew-hours by method change?
It exposes the real bottleneck. An outage scope expressed as three hundred tasks says nothing about whether it is executable. The same scope expressed as ultrasonic crew-hours, magnetic particle crew-hours and radiography crew-hours, against the certified people actually mobilised for those days, shows the shortage weeks earlier. Since most outage NDT is contract labour, that lead time is the difference between adding a crew and cutting scope.
Why do gas and isolation belong in the inspection schedule?
Because they determine when the work can start. Blast furnace gas is roughly a fifth to a quarter carbon monoxide, and every internal examination downstream of it depends on a purge, a gas test and a permit issued by a crew that is itself a scarce resource. If isolation and gas-freeing are held in a separate permit system, the inspection plan will be built on start times that the permit office has no knowledge of and cannot honour.
How do repairs to jurisdictional pressure equipment change the plan?
They add an inspector and a documentation path. Repairs and alterations to jurisdictional boilers and pressure vessels follow the National Board Inspection Code, which means an authorised inspector's involvement, a qualified repair procedure, and paperwork that has to be complete before the equipment returns to service. Scheduled late, this becomes a restart delay rather than an engineering one, which is a far more expensive kind of delay.
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.