Why it matters
Without AI, mislabelled plans, inconsistent scales, and missing archive standards cause post-excavation delays, risk of non-compliant archives, and backlogs in HER submissions and curation.
From design to safety
Design
AI supports plan quality design
The AI reads draft written schemes of investigation, desk-based assessments, and proposed context sheets to verify plan numbers, scales, and legends before field capture proceeds, enabling consistent design decisions. It cross-checks proposed trench layouts against the Harris matrix, flags potential conflicts, and records deviations for inclusion in the design pack and archive-ready design packets.
- drawing registers
- plan numbers
- scales
Build
AI checks in-field drawings
During trench setup and recording, the system cross-checks live context sheets against plan legends and levels to ensure on-site plans align with trench maps and section drawings, reducing rework. It flags mismatches between on-site measurements and plan elevations and creates an incident log for corrective actions.
- levels
- legends
- scales
- plan numbers
Operation
AI validates ongoing archiving readiness
As recording proceeds, AI flags inconsistent annotations, cross-references context sheets with updated plans, and ensures archive standards are reflected in transfer notes, captions, and bundle metadata. It automates draft deposition notes and aligns them with master templates.
- drawing registers
- archive standards
- legends
Maintenance
AI maintains archive readiness
Periodic rechecks compare final context sheets to master plans, updating metadata, validating plan numbers against the archive deposition checklist, and recording version history for audit trails. It surfaces gaps in the master archive template and recommends versioned revisions.
- plan numbers
- archive standards
- scales
Safety
AI guards documentation integrity
In safety reviews, AI confirms that drawings reflect trench boundaries, levels, buried services notes, and UXO risk zones, preventing misfiles that could compromise welfare and risk assessments. It integrates with site safety records to verify that plans support welfare and emergency planning.
- levels
- plan numbers
- legends
Part by part
- 1
drawing registers
- Watch for
- mismatched context numbers or missing signatures that would affect provenance
- AI action
- reads context IDs and logs discrepancies for immediate corrective action
- 2
plan numbers
- Watch for
- duplicated or missing plan numbers that break traceability
- AI action
- validates unique IDs and routes anomalies to archive officers
- 3
scales
- Watch for
- scale inconsistencies between field sheets and generated plans
- AI action
- flags scale drift and prompts re-scaling before archiving
- 4
levels
- Watch for
- level mispostings or inconsistent elevations across sheets
- AI action
- cross-checks contour intervals and raises deviations for recollection
- 5
legends
- Watch for
- missing symbols or legend mismatch with plan sheets
- AI action
- audits symbol vocabulary and harmonizes with master legend template
- 6
archive standards
- Watch for
- incomplete deposition metadata or missing archive format
- AI action
- ensures metadata completeness and flags missing elements for senior approval
Worked example
Catching a plan mismatch before archival acceptance
Illustrative scenario with sample data.
A 0.8 ha development site yields mixed-epoch contexts; draft WSI requires rigorous plan labeling and archive metadata to pass archival checks. The team aims to produce interoperable records that survive HER review and future access requests, with every drawing tied to its context sheet and Harris matrix position.
Signal in
project: 0.8 ha; periods: Roman, medieval; context sheets: 120; master plan: 1; scales: 1:100; plan numbers: P-01 to P-25; archive standards: compliance checklist; draft legend: L1; deposition notes draft: DN-3; master archive template: MAT-2
System out
alert: plan P-04 has mismatched scale; action: re-measure to 1:100 and update plan, context sheet C-24; report: archive checklist updated; owner: archive officer; log: deviation D-12 recorded in design pack
Human sign-off: Archiving officer validates the action, signs off the corrected plan, and forwards to HER for submission; planning archaeologist signs off on the revised legend and deposition notes.
How to roll it out
Phase 1
Weeks 1 to 4
Baseline data models for plans, scales, and legends are established and prepared for automated checks.
Phase 2
Weeks 5 to 12
In-field integration tests verify AI reads drawn registers and context sheets against the master archive template.
Phase 3
Weeks 13 to 20
Full workflow for design through archiving is operational with automated audits and escalation routes.
Where people stay in charge
- Decisions about non-compliant archive metadata must stay with the archive officer and project manager.
- Final deposition changes require sign-off by the planning archaeologist or heritage consultant.
- The system must not override legal deposit rules or standard archive formats.
Questions owners and crews ask
What is AI drawing register and plan quality review before archaeological archiving?
It is an AI-assisted check that validates drawing registers, plan numbers, scales, levels, legends, and archive standards before plans are archived, ensuring provenance and compliance.
Which staff must confirm AI-flagged issues in plans?
The archive officer, project manager, and planning archaeologist review AI alerts and approve corrective actions before deposition.
How does AI integrate with context sheets and Harris matrix?
AI cross-references context IDs with plans, ensuring consistency across context sheets, plans, and the Harris matrix for traceability.
What records are generated for post-excavation archiving?
Automated deposition notes, revised plan sheets, updated legend files, and metadata logs accompany the final archive set.