Case Study: Dimensional Control for Shipbuilding

TL;DR: This dimensional control shipbuilding case study follows a Western Australian shipyard erecting a 78 m steel patrol-class hull from six pre-fabricated blocks, where two adjoining blocks were drifting out of tolerance and threatening a costly weld-and-grind correction on the erection berth. ISS established a tracker-based control network across the build hall, measured every block's key control points against the digital ship model, and drove the butt-joint fit-up to within 1.5 mm before tacking. The erection joint closed first time, rework was eliminated on that seam, and the block landing held the yard's erection schedule.

Key takeaways

• A laser-tracker dimensional control survey caught a 6.2 mm cumulative drift across the block-to-block butt joint before erection, while it could still be corrected by jacking and re-fairing rather than by cutting a completed weld out on the berth.
• Measurement used a Leica Absolute Tracker AT960 (±0.015 mm + 6 µm/m) for the close-range block geometry and a Leica MS60 MultiStation to carry a stable control network the full length of the build hall, all reduced against ISO 1101 geometric tolerancing principles because no single Australian Standard prescribes hull fit-up tolerances.
• Every block was checked as-built against the yard's 3D ship model, so deviation was reported as a colour-mapped comparison to the design surface rather than as isolated tape measurements that hide cumulative error.
• Driving the fit-up to within 1.5 mm at the butt joint — against the yard's working tolerance of 3 mm — meant the seam closed first time, with no excess weld metal, no back-gouging, and no distortion correction on the critical-path erection berth.
• The full scope — control network, six block surveys, fit-up guidance and an as-built report — was delivered for roughly AUD $24,500 against a rework exposure the yard costed at AUD $90,000-plus once a berth-bound weld correction, NDT re-test and schedule slip were counted.

The yard and the problem

The client operates a steel shipbuilding facility on the WA coast, building patrol-class and commercial vessels under a rolling delivery contract. The vessel in this case study is a 78 m monohull erected from six pre-fabricated hull blocks — bow, three parallel mid-body blocks, engine-room block and stern — each built up to 14 m long on its own jig before being craned onto the erection berth and joined block-to-block. On a build like this, the geometry that matters is not any single block in isolation; it is whether the blocks stack into a fair, straight hull once they meet.

The production manager flagged the issue during block fabrication, not on the berth. Two adjacent mid-body blocks had each passed their own internal tape checks, yet the foreman could see the deck-edge sheer line was not going to line up cleanly when they met. The concern was the butt joint between them: the transverse bulkheads, the shell plating run, and the deck and tank-top stringers all had to close at the same seam, and small errors built up across each block were accumulating toward the joint.

The yard had two questions. First, what is the actual deviation of each block against the design hull, in millimetres, before anything is craned onto the berth? Second, can the fit-up be corrected on the jig — where blocks are accessible and adjustable — rather than discovered on the erection berth, where a correction means cutting steel at height on the critical path?

Why dimensional control beats tape measures

Traditional shipyard checking relies on tapes, plumb-bobs, optical levels and reference to centreline pin-marks. Those tools are fine for a single frame or a local check, but they have a blind spot: error accumulates. A 1 mm slip at frame 40, another at frame 60, another at the bulkhead, and the seam at frame 80 is out by an amount no single tape measurement ever showed. By the time the blocks meet on the berth, the cumulative drift is baked in.

Dimensional control treats the whole block as one measured object referenced to a common datum. Instead of asking "is this frame in the right place relative to the last one", it asks "where is every control point relative to the design hull". That reframing is the entire value of the method on a shipbuilding fit-up: it surfaces accumulated error while the block is still on the jig and adjustable, instead of after it is welded into the hull.

So ISS did not measure the joint. It measured the blocks — fully, against the model — and let the comparison reveal where the joint would land.

Establishing control across the build hall

The first task was a stable control network spanning the length of the build hall, because a tracker only sees a few metres of a 14 m block from one position and must be moved and re-tied without losing accuracy. ISS set permanent reference nests on the hall columns and floor, then used a Leica MS60 MultiStation to tie them into a single local network, with the laser tracker bundle-adjusted into the same frame at each setup.

The network was held to a sub-2 mm closure across the full hall, so the tracker could leapfrog along a block — and from one block to the next — while every measured point stayed in one consistent coordinate frame. The network was deliberately referenced to the ship's own build datum (centreline, baseline and a defined frame zero) rather than to a site grid, because the only frame that matters in shipbuilding is the vessel's, not GDA2020 — this is plant geometry, not a cadastral or earthworks survey where MGA2020 and AHD would govern.

That control frame is what lets each block, measured separately on its own jig, be compared as if all six were already assembled. Without it, six accurate block surveys would still be six separate measurements that could not be reliably stacked.

Measuring the blocks against the model

With control established, ISS surveyed each of the six blocks with the Leica Absolute Tracker AT960, capturing the key control points the yard's engineers had nominated: shell plate seams and butts, deck-edge sheer points, transverse bulkhead faces, tank-top and stringer lines, and the principal frame positions. Each block was probed against the yard's 3D ship model and the result delivered as a colour-mapped deviation surface — green where the steel matched design, warming to red where it drifted.

The two suspect mid-body blocks told the story immediately. Individually each sat within roughly 3 mm of model across most of its length. But the deviations ran in the same direction at the adjoining ends, so the cumulative drift across the planned butt joint was 6.2 mm — twice the yard's 3 mm working fit-up tolerance, and exactly the kind of stacked error that tape checks had missed. The deck-edge sheer line, the foreman's original worry, was the worst-affected feature.

Crucially, this was caught with both blocks still on their jigs. The deviation map showed not just that the joint was out, but which block, which end, and in which direction — turning a vague "it won't line up" into a specific, correctable instruction.

Equipment and method

All instruments carried current ISO/IEC 17025-traceable calibration certificates. Because there is no Australian Standard that prescribes hull block fit-up tolerances the way AS 1418.18 governs crane runways, results were assessed against the yard's own build specification, the OEM ship model, and ISO 1101 geometric tolerancing principles, with the yard's 3 mm working tolerance as the pass/fail line. Work in the build hall was covered by the yard's permit system and ISS's own SWMS, with surveying sequenced around fabrication so no measurement held up a welding crew and no welding obscured a control point before it was captured.

Driving the fit-up

The deviation map converted the problem into a plan. Rather than accept the drift and correct it on the berth, the yard adjusted the more-deviated mid-body block on its jig, with the ISS tracker live so each jack movement and re-fairing step was verified in real time rather than checked after tacking. The team worked the deck-edge sheer line and the shell run back toward the design surface, watching the colour map shift from red to green as the steel moved.

The blocks were then craned to the erection berth and presented to each other. Because both had been driven to the model before they left the jigs, the butt joint closed to within 1.5 mm across the full seam — well inside the 3 mm working tolerance. The transverse bulkhead, shell plating and deck stringers all met cleanly, with a consistent root gap ready for welding.

The seam was welded as a normal production joint, with no excess weld metal to grind, no back-gouging to chase distortion, and no out-of-position correction work at height on the erection berth.

The outcome for the yard

The commercial logic is direct. The full ISS scope — hall control network, six block surveys, live fit-up guidance and an as-built report — came to roughly AUD $24,500. The exposure it removed was a berth-bound weld correction on the butt joint: cutting out an out-of-tolerance seam, re-fairing at height, re-welding, re-running non-destructive testing on the repaired joint, and absorbing the schedule slip while the erection berth sat occupied. The yard's own production team costed that scenario at over AUD $90,000 once berth time and re-test were included — and that assumed the error was caught at fit-up rather than after the hull moved on.

Just as important, the method changed how the yard plans the rest of the build. With every block now measured against the model on its jig, the production manager has a documented as-built record for the hull and a repeatable check that catches accumulated drift before it reaches the berth. Dimensional control moved from a one-off rescue on a problem joint to a standing quality gate between fabrication and erection — the point of the whole exercise being to keep error correction on the jig, where it is cheap, and off the critical-path berth, where it is not.

Frequently asked questions

What is dimensional control in shipbuilding?

Dimensional control is the practice of measuring hull blocks against the design ship model using a common, surveyed datum, so that accumulated fabrication error is found and corrected before blocks are joined. Instead of tape-checking one frame against the next, every control point is referenced to the vessel's build datum, which surfaces the cumulative drift that block-to-block fit-up depends on.

Why not just use the yard's tapes and optical levels?

Those tools are accurate locally but cannot stack measurements across a 14 m block without accumulating error. A series of small 1 mm slips along a block is invisible to tape checks yet adds up to a butt joint that will not close. In this case study the cumulative drift at the joint was 6.2 mm even though each block passed its own internal checks — only a model-referenced survey revealed it.

What accuracy is achievable on a hull block this size?

The Leica AT960 tracker holds about ±0.015 mm plus 6 µm/m, far tighter than any shipyard fit-up tolerance. On a real block the limiting factors are the steel itself — plate flatness, weld distortion and thermal movement — and the control network, which here was held under 2 mm across the build hall. The practical, defensible target was the yard's 3 mm working tolerance, and the joint was driven to within 1.5 mm.

Is there an Australian Standard for hull fit-up tolerances?

No. Unlike crane runways, which AS 1418.18 governs, there is no Australian Standard prescribing hull block fit-up tolerances. ISS works to the yard's build specification, the OEM 3D ship model and ISO 1101 geometric tolerancing principles, with ISO/IEC 17025-traceable measurement. The method and traceability are what make the result defensible, not a single referenced clause.

When in the build should dimensional control happen?

Before erection, while blocks are still on their jigs and adjustable. The entire value of the method is catching accumulated error where it is cheap to correct — by jacking and re-fairing on the jig — rather than on the erection berth, where a correction means cutting completed steel at height on the critical path. The earlier each block is measured against the model, the cheaper any correction is.

Industrial Spatial Solutions delivers tracker-based dimensional control for shipbuilders, fabricators and offshore-module yards across WA and nationally, using Leica laser trackers and total stations to compare as-built steel against the design model and drive sub-millimetre fit-up before erection. If you are erecting a hull from blocks, building large fabricated modules, or simply want a quality gate between fabrication and assembly, call ISS on 0407 057 015 to scope the work around your build schedule and request a fixed-price quote.