Knowledge / AEC-native data
Method statements and RAMS
Risk assessments and method statements describing how a task will be carried out safely, submitted before work starts and reviewed by the principal contractor.
A RAMS is the subcontractor's account of how a task will be done safely, submitted before the work starts, reviewed by the principal contractor's site team and safety adviser, and briefed to the operatives who will do it. A live project holds hundreds, revised when the sequence changes, and many begin as a generic template with the site name changed. The reviewer checks the method against the drawings, the programme, the site risk register and the permit to work system, and judges whether it describes this task rather than one like it. RAMS AI cannot make that judgement. What it can do is put the drawing, the programme and the register beside the document, so the competent person can question all four together, with every answer traceable to its source.
What teams need to ask of it
Does this RAMS describe the work as shown on the current drawings, or an earlier version of it?
Does the method match the sequence in the current programme revision, or a sequence that has since changed?
Which hazards on the site risk register does this RAMS not address?
Which RAMS are approved, which have expired, and which were revised after a change to the work?
Was this RAMS briefed, to whom, and on which date?
Which RAMS covered the activity in that location on the day of the incident?
Does the RAMS assume a permit that the site's permit to work system does not issue for this activity?
Why generic AI gets it wrong
A RAMS read alone almost always looks complete. It has a scope, a sequence, a hazard table, control measures and a sign-off box, and generic wording is written to sound thorough. A general-purpose tool reading it will find nothing missing, because the things that are missing are not in the document. They are on the drawing, which shows a slab edge the method does not mention; in the programme, which now sequences the work after the scaffold has been struck; and in the site risk register, which lists a live service the template never heard of. The tool has none of these, so it praises the document for its thoroughness, and a reviewer who trusts it signs off a method for a different job.
The second failure is the revision trail. RAMS are revised when the work changes, briefed in versions and superseded without ceremony, and the question after an incident is which version was approved and briefed on the day. A tool that reads the folder cannot tell rev B from rev C when both say approved, or notice that the briefing register refers to a version withdrawn the week before. Its answer names a document and sounds settled. In a review that a competent person is accountable for, and that an investigator may read later, a confident answer with the wrong revision is the failure that matters.
One question, answered
QUESTION
ANSWER
SOURCES
We onboard a focused slice of your project data, configure the ingestion paths, extract a structured context graph, and build a workflow against the resulting structured knowledge layer.
Related to
A RAMS is checked against the construction drawings that show the work, the construction programme that sequences it and the site diaries that record what actually happened on the day. Project knowledge search is how a safety adviser finds which RAMS covered an activity, and lessons learned capture is where the near misses, and the methods revised after them, are kept for the next project.
Does it replace the review by a competent person?
Which formats can it read?
How do we see it on our own RAMS?
Related pages
Construction programmes
The logic-linked schedule of activities, durations, milestones and float, held in P6, Asta or MS Project and issued as baselines and updates.
Construction drawings
The 2D graphical record of design intent: plans, sections, elevations, details, schedules and title blocks, issued as sheets with revisions.
Site diaries and daily reports
The daily record of labour, plant, weather, progress, deliveries, instructions and events on site, often the evidence base in disputes.
Project knowledge search
Asking questions across the whole project record, drawings, models, specs, contracts and correspondence, and getting answers that cite their source.
Lessons learned and institutional knowledge capture
Capturing decisions, defects, precedents and outcomes from completed projects into structured knowledge the next project can reuse.