Why it matters
Without AI, stabilisation risks irreversible damage from moisture, ionic attack, mis-timed drying, or incompatible consolidants, leading to costly re-treatment, loan delays, and damaged institutional trust.
From design to safety
Design
AI-guided stabilisation plan
The AI reads initial condition reports, X-ray findings, material IDs, moisture data, porosity, consolidant compatibility, reversibility notes, caliper measurements, and prior treatment histories to propose a stabilisation strategy, PEG choice, and controlled drying sequence that prioritises reversibility and minimal intervention while documenting each rationale in the treatment proposal and the PACR-compliant records.
- X-ray
- PEG treatment
- waterlogged wood
- freeze-drying
Build
Systematic treatment setup
The AI supports selecting materials, setting dosing thresholds, and scheduling bench tests; it generates batch records, QA sign-offs, and a materials ledger; it creates a go/no-go matrix for on-site operations, including glovebox use and ventilation settings.
- PEG treatment
- freeze-drying
- X-ray
- waterlogged wood
Operation
Real-time monitoring and guidance
During treatment, AI analyzes imaging, weight changes, moisture readings, calorimetric data, and consolidant absorption to adjust steps, trigger alerts, and append a live treatment log with timestamped entries, photograph series, and colourimetric maps over the drying horizon.
- X-ray
- waterlogged wood
- PEG treatment
- freeze-drying
Maintenance
Calibration and data integrity
Regular calibration of sensors (moisture probes, thermocouples, balance scales), verification of data streams, and cross-checks against ISO standards ensure readings remain consistent; versioned logs enable provenance tracking and preventive care planning with quarterly audits and component replacement schedules.
- X-ray
- finds packaging
- waterlogged wood
- PEG treatment
Safety
Chemistry and handling governance
The system flags COSHH risks, ensures proper containment, routes approvals to the accredited conservator and safety officer, documents PPE requirements, waste streams, and emergency procedures before any chemical use; it enforces spill containment and signage during workflow transitions.
- PEG treatment
- freeze-drying
- finds packaging
- X-ray
Part by part
- 1
waterlogged wood
- Watch for
- Sustained swelling, delamination, loss of structural integrity, rapid shrinkage on drying
- AI action
- Detect moisture gradient and trigger staged drying protocol with progressive consolidation
- 2
iron finds
- Watch for
- Rust bloom progression, chain-flagging of corrosion fronts, loss of surface detail
- AI action
- Monitor corrosion potential via surface colorimetry and humidity, initiate corrosion-inhibiting steps
- 3
PEG treatment
- Watch for
- Phase separation, uneven penetration, inappropriate viscosity
- AI action
- Recommend consolidant concentration and penetration timing based on porosity data
- 4
freeze-drying
- Watch for
- Collapse risk, micro-cracking, thermal shock
- AI action
- Optimize chamber temperature/pressure ramp and sublimation rate
- 5
finds packaging
- Watch for
- Mechanical stress during transport, rehydration risk, label mismatch
- AI action
- Suggest packaging configurations and alignment with condition reports
- 6
X-ray
- Watch for
- Internal cracks, laminations, voids migration
- AI action
- Highlight defects and track dimensional changes during drying
Worked example
Stabilisation of a waterlogged oak trough with iron fittings
Illustrative scenario with sample data.
A private studio receives a waterlogged wooden trough with iron fittings. AI proposes PEG consolidation, controlled freeze-drying, and protective packaging with imaging checks.
Signal in
conditionReport.date=2026-09-12; material=oak; moisture=32%; ironFittings=true; Xray=internalCracks; porosity=0.42; PEG=notApplied; weight=12.5kg; porosityMap=0.42@surface,0.38@mid,0.30@core; reversibilityScore=8/10
System out
treatmentProposal: PEG consolidation at 2% w/v, freeze-dry to 5 mTorr, packaging plan with corrugated separators, imaging schedule including post-drying X-ray, conditionReport update; safety sign-off.
Human sign-off: Head conservator approves the treatment proposal and safety plan; client notified with updated condition report and loan schedule; PACR records updated.
How to roll it out
Phase 1
Weeks 1 to 4
Baseline assessment, data capture, and system calibration across imaging, moisture, porosity, and chemical compatibility sensors; establish data pipelines to the PACR archive.
Phase 2
Weeks 5 to 8
Pilot treatments on low-risk items; validate AI recommendations against manual bench tests, adjust thresholds, and document deviations with signed approvals.
Phase 3
Weeks 9 to 12
Full-scale deployment with continuous monitoring, automated logs, and routine condition reporting; archive all decision rationale and material histories in the project file.
Where people stay in charge
- All decisions on chemical stabilisation and solvent use remain with the accredited conservator and safety officer.
- No irreversible treatments without written approval and a documented risk assessment.
- Data-handling and archival decisions must follow PACR accreditation and institutional policy.
Questions owners and crews ask
What data does the AI read during waterlogged stabilisation?
The AI reads condition reports, X-ray scans, moisture readings, porosity, reversibility scores, and consolidant compatibility data to inform treatment planning and monitoring.
Who signs off on AI-driven treatment plans?
The accredited conservator or head of conservation signs off, with safety officer approval when needed.
How are records kept for provenance and loan packages?
Versioned condition reports, treatment proposals, imaging analyses, porosity maps, and packaging records are stored in the central collections database with audit trails and PACR-compliant metadata.
Can AI alter a treatment if data changes mid-project?
Yes, the AI proposes adjustments, but final changes require human validation and sign-off before implementation.