Brownfield before the walkdown

Every revamp or turnaround on an operating asset starts in the archive. Before anyone walks the unit, someone has to work out what the drawings say is there: which P&IDs are current, which lines and instruments fall in scope, where the tie-ins go. On a unit that has run for decades, the archive rarely gives a clean answer.

The usual response is to send people into the field. The walkdown becomes the place where the team learns what exists, and scope gets priced from what a sample showed. This post argues for the other order: read the archive into a structured register first, so the walkdown verifies a list instead of discovering one.

What the archive looks like

Ask for the drawings of an operating unit and what arrives is usually some mix of:

• Scans of old prints. Sepia and blueline prints scanned years ago at whatever resolution was available, with stamps, fold lines and handwritten notes in the margins.
• Several generations of CAD. Sheets redrawn in different CAD systems over the years, some exported to PDF, some surviving only as plots.
• Superseded revisions beside current ones. Two or three copies of the same drawing number at different revisions, with nothing outside the title block to say which one governs.
• Red-lines that never made it back. Field markups from past changes in a binder in the control room or a folder on a shared drive, never drafted into the masters.
• Vendor packages. Skid drawings to the vendor's conventions, with tag numbering that may or may not match the plant's.
• Missing or stale registers. No line list, or one kept in a spreadsheet by someone who has since retired, and an instrument index that stopped matching the field several changes ago.

An asset that has changed hands adds another layer: drawing numbers, tag formats and legend conventions from each previous owner, sometimes on the same unit.

Why scope priced off a walkdown sample is risky

Walkdowns are expensive and constrained. The unit is running, so access depends on permits, escorts and operations. Piping runs at height, under insulation and behind other equipment. Many lines carry no field marking of their line number, so identifying one means tracing it physically against a drawing that may not be current. Time on the unit is limited, and so is the number of people who can be there. So teams sample. They walk the areas the scope obviously touches, check a representative set of lines and instruments, and extrapolate the rest.

The risk sits in the extrapolation. Everything the sample did not see is uncertainty, and it lands somewhere:

• On a lump-sum revamp, the contractor either qualifies it out of the bid or absorbs it as growth in the field.
• On reimbursable work, the owner pays for it as change.
• On a turnaround, scope found during the event competes for a fixed window, and late discoveries can push work onto the critical path.

Take an illustrative revamp that touches four units, with a walkdown budget that covers two of them in detail. Scope for the other two is priced from the archive as it stands, plus whatever the walked units suggest. Any tie-in, line or instrument the archive misrepresents in those two units stays invisible until construction finds it.

A walkdown sample tells you about the lines it saw. The bid carries the ones it did not.

Read the archive first

The alternative is to digitize the archive before the walkdown and turn it into a structured register. Every P&ID is read into a typed graph of equipment, lines, loops and instruments, connected across sheets, with each record traced to the sheet and region it came from. Engineers verify it tag by tag, low-confidence items route to review, and every decision is logged.

From that graph come the registers the archive is missing:

• a line list with size, service, piping class and insulation parsed from each line number, and every sheet each line crosses
• an equipment list and an instrument index, with loops grouped and cross-sheet references resolved
• a drawing register that separates current revisions from superseded ones, with the superseded sheets kept as history

For a contractor, this can happen at bid when the client makes the archive available, or at kickoff when it does not. For an owner, it can happen well before a turnaround is scoped.

The walkdown then changes character. The team goes to the unit with a register and a set of specific questions, not a blank notebook. They confirm what the register says, correct it where the field disagrees, and record each finding against the entry it confirms or corrects. Field time goes to the items that matter for the scope, not to rediscovering what the drawings already said.

Find the contradictions on paper

Reading a whole archive into one graph does something a sampled walkdown cannot: it puts every document next to every other. Contradictions that stayed invisible while each drawing was read on its own become explicit. These are the kinds that typically surface:

• the line list gives a line as 4" and the P&ID shows it at 6"
• two revisions of the same sheet disagree about whether a valve is normally open or normally closed, and it is unclear which revision governs
• an off-page connector names a sheet that does not exist in the archive
• the same tag appears on two different pieces of equipment
• an instrument on the P&ID has no entry in the instrument index, or the other way round
• a vendor package drawing shows a tie-in the plant P&ID does not

Each contradiction is documented with both sources cited, down to the region of each sheet, before anyone goes to the field. Some resolve on paper, because one document is plainly superseded. The rest become targeted walkdown questions with a location attached, so the field team knows exactly what to look at.

Not every contradiction is a drafting error. Where the field turns out to differ from the drawings, the cause is often a red-line that never made it back, and sometimes a change that never went through management of change. The two are resolved differently, and an owner would rather learn which is which before the revamp than during it.

For a contractor, a documented contradiction is a clarification raised during the bid, with drawing evidence behind it, rather than a claim argued after award. For an owner, it is a known correction to process safety information rather than a surprise during construction.

What the operator keeps

The revamp ends, the contractor demobilizes and the scope is closed out. What remains matters as much as the project did:

• A verified register that matches the field, because the walkdown confirmed it and the project's own changes were recorded against it.
• A graph the next change can start from. The next MOC on these units traces lines through a graph instead of through binders, and the next turnaround is scoped from the register instead of the archive.
• Traceability. Every tag resolves to the drawing, revision and region it came from, and every verification carries a name and a date.
• History. Superseded revisions are kept and linked, so the question of what the drawing showed before the revamp still has an answer.
• A firmer base for compliance work, such as an LDAR component inventory or a process safety information review, that depends on knowing what is installed.

Without this, what the project team learned leaves site with the people who learned it. With it, the operator keeps the work.

Walk the plant with a list

Walkdowns are not going away, and they should not. Nobody should commit to scope on a brownfield unit without walking it. The argument is about order. Read the archive first, document what it contradicts, and send the field team out to confirm a list. The walkdown spends its time on the questions that matter, the scope is priced on a counted base instead of a sample, and the operator finishes the project with a register that matches the plant.

For how we read the drawings themselves, see Reading a P&ID like an engineer. For how this fits project and operating work, see construction and EPC and energy.