Making Bridge TMLs Re-Findable Before the Closure Window Opens
A bridge CML/TML registry fixes each thickness point to a span, member, face and access envelope rather than a pipe circuit, so the same steel is re-measured campaign after campaign. Before a lane-closure or in-water window opens it produces an access-grouped work list: what one under-bridge unit setup or rope drop can reach, and what each reading feeds in the load rating.
Steel bridge deterioration is not the broadly uniform wall loss a process plant plans around. Section loss concentrates under leaking deck joints, in the troughs of built-up members, at bearing seats, and inside pack rust that jacks faying surfaces apart. Pack rust reads thicker, not thinner, on an ultrasonic gauge, so an uncorrected grid mean flatters the member. AASHTO's Manual for Bridge Evaluation wants measured section properties before a rating is revised, and 23 CFR 650 Subpart C with the FHWA SNBI item set fixes the interval those readings must survive to. Between campaigns the coating system is renewed, the scaffold comes down and the chalk marks are gone, so a TML recorded only as "web near bearing" cannot be re-found. The registry's real product is a locator strong enough to survive a recoating, and a grouping of points by what one access setup can physically reach.
Source: Sources: 23 CFR 650 Subpart C, National Bridge Inspection Standards; FHWA Specifications for the National Bridge Inventory (SNBI); AASHTO Manual for Bridge Evaluation; AASHTO/AWS D1.5 Bridge Welding Code; FHWA Bridge Inspector's Reference Manual; ASME BPVC Section V Article 23 (SE-797, ultrasonic thickness measurement); ASNT SNT-TC-1A; OSHA 29 CFR 1926.62, lead in construction; MUTCD Part 6 for temporary traffic control.
| Locator element | Process-plant convention | Bridge equivalent that actually works | Why the substitution matters |
|---|---|---|---|
| Parent object | Equipment tag and corrosion circuit | Structure number, span, member mark from the shop drawings | Bridges have no circuits; invented circuits become meaningless within two campaigns |
| Position on the object | Clock position and distance from a weld | Rivet or bolt row count from a named splice plate or stiffener | Welds are often absent on riveted members; rivet lines are permanent and countable under paint |
| Vertical reference | Elevation or line number | Height above deck or below soffit, plus fascia or interior face | Determines whether the point is reachable by snooper, rope or scaffold |
| Condition context | Circuit damage mechanism from API 571 | AASHTO element number and defect, e.g. corrosion on a steel girder element | The regulatory report is element-quantity based while the reading is a point measurement |
| Baseline | Design nominal wall from the line list | Measured reference thickness at a protected location on the same member | Plan nominals carry mill tolerance and undocumented strengthening plates; percent loss against them is not defensible |
| Scope grouping | Circuit or line | Access envelope: everything reachable from one setup | Setup time dominates reading time on a bridge, inverting plant economics |
A bridge portfolio has no CML tradition to inherit
Process industries spent four decades building a vocabulary for condition monitoring locations. API 570 gives you circuits, API 574 tells you where to place a point on an elbow versus a straight run, and the numbering descends from an equipment tag assigned when the line was first drawn. A steel bridge inherits none of that. Its statutory condition record is element-based, a quantity of an element distributed across condition states, and elements are counted in linear feet or each. There is no native concept of a point that gets re-measured.
That mismatch is the first thing a registry has to absorb rather than fight. A section-loss reading is a point measurement, but the inventory it must reconcile with is a quantity. Software that insists on a parent equipment record with a child circuit forces inspection engineers to invent placeholder circuits so the form will save. Those inventions are undocumented by definition, and two campaigns later nobody can say whether circuit B-04-W meant the west fascia girder or the west half of the floor system.
The model that survives carries both identities on one record: an element and defect reference sufficient for the condition report, and an independent geometric address sufficient for the probe. One audience is the biennial report. The other is a technician standing on a platform in February trying to find last cycle's point.
The locator has to survive a recoating
Every new TML on a painted steel bridge starts with coating removal to bare metal and ends with a coating repair. Where the existing system contains lead, that pulls in containment, exposure monitoring and waste handling under OSHA 29 CFR 1926.62, and the repair has to be compatible with the surrounding system. So a TML is not free the way a chalk circle on insulated piping is free. It is a small capital item with a permanent maintenance obligation attached, which changes the whole economics: the value is in re-using the same point, not in adding density.
Which means the locator has to be written for someone who has never seen the structure. Span number and member mark from the shop drawings. Face, fascia or interior. Position within the cross-section: top flange, web, bottom flange, outstanding angle leg. Then a countable offset from permanent geometry rather than a dimension: rivet rows from a named splice plate, bolt count north of a stiffener, distance from the centreline of bearing. A photograph with a scale in frame and a recognisable datum feature, not a close-up of bare steel that could be anywhere on the bridge.
The acceptance test is blunt. Hand the record to a technician who has never been on that structure and ask whether they can place the probe within about 25 millimetres of the last reading, working from the record alone, after the bridge has been blasted and repainted. If the answer depends on someone remembering, the locator has failed and the trend it feeds is not a trend at all — it is a series of unrelated measurements on the same general member.
Access is the unit of scope, not the member
Under a bridge, the cost driver is not the reading. It is getting to the reading. An under-bridge inspection unit has to be positioned, the lane closure has to be set out to MUTCD Part 6, traffic control has to be manned, and the boom has its own reach envelope and clearance limits. Rope access needs anchor identification and a rescue plan. Water access needs a vessel, a channel notification and often a seasonal window. Each of those is a fixed cost paid per setup, not per point.
So the registry's work list must be ordered by access envelope: the set of TMLs reachable from one snooper position, one rope anchor, one scaffold bay, one lift location. Group by member and the crew repositions constantly, doing three points, striking the setup, moving forty feet, doing four more. On a real portfolio job, 180 points ordered by member has been the difference between roughly forty setups and fourteen. That ratio is the entire economics of a fixed window.
This inverts process-plant intuition, where reading time dominates and access is largely a scaffold that stays up. On a bridge, access is transient and metered. A registry that cannot express "these nineteen points share a setup" cannot plan a closure, and the planner ends up rebuilding the grouping by hand in a spreadsheet every single campaign — which is where the version that goes to the field diverges from the version of record.
Pack rust, pitting, and readings that lie
Built-up and riveted members corrode between their faying surfaces. The oxide occupies several times the volume of the steel it consumed, so it jacks the plates apart, distorting the member and sometimes shearing rivets. A digital thickness gauge pointed at that region can time to the far side of the laminated corrosion product rather than to the back wall of sound steel, and return a number at or above nominal on a member that has lost real section. Anyone reading a spreadsheet of numbers with no context will conclude the member is fine.
Heavy pitting on the back wall causes the opposite failure. The reflected energy scatters, the back-wall echo drops out, and the instrument either refuses to read or locks onto a spurious interface. The fixes are ordinary — a twin-crystal probe, a lower frequency, echo-to-echo mode through the coating, an A-scan display rather than a bare digital readout — but they only exist in the record if the record has fields for them.
So the registry must hold probe type and frequency, echo mode, couplant, surface preparation performed, and whether the coating was removed or the reading was taken through it. Above all it needs an explicit exclusion path: a reading the technician did not trust, retained with the reason. Discarding it silently on site is how a member with pack rust gets carried forward as sound steel, and the discrepancy surfaces years later when the rating is revisited and nobody can explain the gap.
What the load rating engineer actually needs
On a bridge, thickness is not primarily a remaining-life input. It feeds a load rating under the AASHTO Manual for Bridge Evaluation, and the rating drives whether the structure gets posted for reduced load under 23 CFR 650.313. That is a public consequence with commercial and legal weight, arrived at by a different engineer, often months after the crew has demobilised. The registry is the only bridge between them.
The arithmetic trap is the baseline. Percent section loss is meaningless without a defensible original thickness, and the plan nominal is not it. Original mill tolerance, plate substitutions during fabrication, and undocumented strengthening or retrofit plates added decades later all mean the drawing number may never have been the real number. The workable approach is a reference thickness measured on the same member at a protected location — inside a box, under a cover plate, above the splash zone — recorded as its own point in the registry and carried forward with the same locator discipline.
What the engineer needs exported is therefore a member-organised package: reference thickness, measured thicknesses with their position in the cross-section, and the extent over which the loss is present. Without extent, the engineer must conservatively assume the minimum applies across the full member length, and a bridge can be posted on the strength of missing information rather than on the strength of measured deterioration.
The window is fixed long before the scope is
Bridge access windows are booked against constraints nobody in the inspection group controls. Lane closure permits are issued by an agency with its own calendar. Night windows have noise limits. Construction season moratoria, holiday restrictions and event closures block whole months. In-water work is often confined to an environmental window set by a fisheries agency. Specialised access equipment and its operators are booked out well ahead. The consequence is that the scope has to be right when the window is requested, not when the crew mobilises.
So the readiness output the registry owes the planner is specific: which TMLs are due this cycle, which access unit each one sits in, which require coating removal and therefore a containment plan and waste handling, which need confined-space entry into a box girder, which technicians hold the qualification the owner agency requires, and what equipment each unit consumes. That is a scope package that can be priced and permitted, not a list of due dates.
What goes wrong without it is predictable and expensive. A point is found on the day to be behind a utility conduit installed since the last campaign, or inside a sealed box with no access hatch, or above a live rail with its own possession requirement. It gets skipped. The gap does not surface at handover — it surfaces at the next load rating, when the engineer discovers the member has no current data and has to rate it conservatively, or send a crew back for a single reading at full setup cost.
How to evaluate a registry for a bridge portfolio
Ask to enter one real TML end to end during the demonstration, on a riveted through-truss or a built-up plate girder rather than a tidy example. Can the record hold a span, a member mark, a face, a cross-section position, a rivet-count offset from a named datum, a reference thickness point, an element and defect reference, and an access envelope, all at once? If any of those has to go in a free-text notes field, that field will be the first thing to rot.
Then test the outputs in both directions. Can it produce an access-grouped field pack a crew can work from, sorted by setup, with photographs, and can that pack be worked offline in a location with no signal and returned without conflict? And can it produce a member-organised export a rating engineer can use directly, with reference thicknesses and extents rather than a flat list? A registry that only outputs one of those two shapes leaves someone rebuilding it by hand.
Finally, migration. Existing data lives in inspection reports, agency databases, consultant deliverables and legacy spreadsheets, with locators of wildly varying quality. A serious system will let you import a point with a weak locator, flag it as unverified, and upgrade it in place on the next campaign without breaking its history. A system that requires a complete locator before it will accept a record guarantees the historical data never gets loaded at all. Atlantis NDT configures the CML and TML registry around the structure types and access constraints in your own portfolio; walkthroughs and scoping consultations are available on request at info@atlantisndt.com.
Why can a bridge TML not be numbered the way a piping CML is?
Piping numbering falls out of an equipment tag and a corrosion circuit that existed before the first reading. A bridge has neither. Its condition record is element-based, counted in feet or each across condition states, while a thickness reading is a point. Forcing a circuit hierarchy makes engineers invent placeholder circuits nobody can decode two campaigns later. The registry should carry an element reference for the report and an independent geometric address for the reading.
What makes a bridge TML expensive to create in the first place?
Each new point needs the coating removed to bare metal and then restored. On a structure with a lead-bearing paint system that pulls in containment, worker exposure monitoring and waste handling under 29 CFR 1926.62, plus a coating repair that must match the existing system. A TML is therefore a small capital item with a recurring coating liability, which is exactly why re-using the same point beats adding new ones.
How does pack rust make a thickness reading read high?
Corrosion product forming between the faying surfaces of riveted or bolted built-up members expands and forces the plates apart. A gauge running in a single-echo mode can time to the far face of the laminated oxide rather than the sound steel interface, returning a value at or above nominal on a member that has lost significant section. Recording probe type, frequency and echo mode alongside the number is what lets a reviewer catch it.
How should the registry group work for a fixed closure window?
By access envelope, not by member or by span. Everything a single under-bridge inspection unit position, one rope anchor or one scaffold bay can reach becomes a work unit with its own duration, permit and crew. On a portfolio job, a member-ordered list of 180 points can require roughly forty repositions where an access-grouped list needs fourteen. Setup, traffic control and equipment movement dominate the shift, not the readings.
What does the load rating engineer need that a simple reading list does not give?
A rating needs reduced section properties, not thicknesses. That means each reading tied to its position within the member cross-section, a defensible reference thickness measured at a protected location on the same member, and the extent over which the loss applies. Handing over a spreadsheet of numbers without geometry forces the engineer to assume the worst case across the member, which can post a bridge that did not need posting.
What is the single most common cause of a lost TML between campaigns?
Recoating. The blast and repaint removes every chalk mark, paint dot and marker the previous crew relied on, and often the temporary tag as well. Any locator that depends on a mark applied during the last campaign fails on the next one. Durable locators reference permanent geometry, rivet and bolt rows, splice plate edges, stiffener spacing, bearing centrelines, supported by a photograph taken with a scale in frame.
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.