Why heritage breaks standard BIM
Modern BIM assumes a building is parametric — walls are rectangles with defined material layers, openings have standard head and jamb conditions, structural elements meet at sharp intersections. Heritage buildings, especially the vernacular stock that Vision 2030 is preserving across Diriyah, AlUla, and historic Jeddah, don't obey any of those assumptions.
A traditional Najdi mud wall in Diriyah isn't a parametric rectangle. It's a stratified mass — mud-brick core, lime-render skim, palm-trunk reinforcement, period repair patches at eye-level where someone in 1956 patched a crack with cement. Modelling it with out-of-the-box Revit families idealises away the things that make it valuable.
The scan campaign
Heritage buildings need denser scan coverage than typical commercial work because the geometry has irregularity that idealised BIM families can't represent. On a recent restoration engagement — a centuries-old castle in Old Jeddah with no usable record drawings — we ran the full scan-to-BIM campaign with these constraints:
Equipment selection
Faro Focus and Leica RTC360 are the two scanners that handle heritage well — both deliver ≤5 mm point density at typical scan distances, both have integrated photographic capture for colourised point clouds (useful for downstream material identification). RTC360 wins on speed in tight interior spaces (vaults, narrow passages); Faro Focus wins on outdoor coverage. Most heritage campaigns use both.
Survey control
The discipline that separates a useful point cloud from an expensive paperweight is scan registration. Every scan station has to register tight against neighbouring stations or the registered cloud accumulates drift. Heritage buildings with thick walls, vaulted ceilings, and few sight-lines through the structure require more scan stations than typical work — we plan for 30–40 stations per 1,000 m² versus 15–20 for modern commercial.
Coverage check on site
The single biggest scan-campaign failure mode is finding a coverage gap after demobilising. On heritage work we register the cloud daily on-site, not after we leave — any gap shows up at 6 PM the same day, scannable at 7 AM the next morning. Going back for a second campaign costs more than getting it right first time.
From cloud to HBIM
The point cloud is the forensic record. The HBIM model is the workable design baseline. Both need to ship as final deliverables — the cloud stays as the verification reference for every later design decision.
Custom family library
Heritage elements need custom Revit families authored against the cloud reference: mashrabiya screens with their specific wood-joinery pattern, traditional door hardware geometry, capital + corbel + cornice profiles that aren't in any out- of-the-box library. We typically build 30–60 custom families per heritage project, each parametric where it makes sense, each tagged with provenance metadata (period, material, condition state).
Period stratigraphy
Most heritage buildings have multiple construction periods stacked on top of each other. A wall might be 1820s mud-brick core, 1940s lime-render skim, 2010s cement patch repair. We model these as separate elements with period parameters, so the model can answer "which courses are original?" and "which are 20th-century repair?" — questions conservation decisions depend on.
Condition mapping
Every visible defect (crack, dampness stain, salt damage, lost render) gets mapped onto the model surface with a condition parameter (severity, intervention priority, photographic evidence link). The model becomes a queryable condition database, not just a geometric record.
What conservation logic adds
Pure technical scan-to-BIM is the wrong methodology for heritage. Minimal intervention, reversibility, and authenticity of materials are the principles that govern conservation — and the BIM model has to encode them, not just the geometry.
- Minimal intervention shows up as preservation-priority parameters: which elements must stay untouched, which can be consolidated, which may be replaced (with documented justification).
- Reversibility shows up in detail design: any new intervention must be removable without damage to historic fabric. The model documents the interface.
- Authenticity of materials means new work uses period-appropriate substrates where structurally possible — and the model specifies them.
HBIM done right means the building becomes a queryable dataset. What material is that wall? Where is the original timber lintel? Which courses are 1820 and which are 1956 repair? Three years later, the conservation architect asks those questions and the model answers without doubt.
Where this is going across Saudi heritage work
Three trends to track through 2027:
First, the Diriyah Gate programme is normalising HBIM as the default for Najdi vernacular restoration. What used to be a niche specialism is now baseline expectation for any AlUla / Diriyah / historic Jeddah engagement.
Second, photogrammetry is supplementing laser scan for high-altitude façades and inaccessible interiors. UAV-based photogrammetry registered against terrestrial scan control gives full-envelope capture without scaffolding.
Third, condition monitoring through repeat scans is starting to appear in the conservation brief — scan the building before restoration, again after restoration, again annually for the first three years. The cloud-on-cloud difference quantifies structural movement and weathering.
Heritage is patient work. The discipline rewards getting the cloud right first time, building the custom family library properly, and treating the model as a conservation dataset rather than a deliverable to tick off the schedule. Koraysh Castle remains our flagship HBIM showcase for exactly these reasons.