Whole-building LCA from IFC, energy model validation, and life-cycle cost analysis: one auditable chain from geometry to GWP to NPV. Every result ships with its boundary, assumptions, and mapping table. No naked numbers.
Every deliverable ships with its boundary, assumptions, and mapping table. Re-runnable, not a static PDF.
FIG. 01 IFC → GWP → NPV EVERY RESULT SHIPS RE-RUNNABLE
Anyone can produce a carbon number. I produce numbers that survive your reviewer, your certification body, and your board.
No LCA runs on an unaudited model. Readiness is measured, scored, and disclosed; uncertainty language scales with model quality.
Every result carries its assumptions block: boundary, study period, functional unit, data sources, biogenic treatment. A number without its boundary is meaningless.
Every deliverable passes an independent review gate before it's called client-ready: claims traced to evidence, figures recomputed, consistency checked across the carbon, energy, and cost models.
Language never exceeds its verified basis. Design models are "reviewed"; only measured-data comparison meeting acceptance criteria earns "calibrated".
Methodology parameters (databases, study periods, discount rates, thresholds) live in versioned methodology documents, applied consistently across every engagement.
Requirements expressed as IDS specs, not prose. The open tooling for this is in progress; once it lands, your team re-runs the validation after every revision without waiting on me.
Fifteen years of this work taught me a method, but it ran on corporate software. I'm rebuilding it on tooling that doesn't need a licence you don't have: IFC, Blender Bonsai, EnergyPlus. Below is the honest state of each piece. The method is proven; the open implementation is under construction.
Where that leaves you. I'm taking one or two projects at a time right now, walking through the process together and hardening it as we go. If you want a polished platform, I'm not there yet. If you want the analysis done properly by someone who will show you every assumption, that part works today, and early clients shape what gets built next.
Model requirements as executable IDS rather than prose in a PDF. The specification method is proven work from my corporate practice. The open-tool version, the one that lets your team re-run the check without me, is what I'm rebuilding now.
Serves: Model Readiness Audit
The chain from a checked IFC to an LCA detailed enough to defend: quantity extraction, material mapping, boundary handling. This is the core of the rebuild and the piece that gates everything downstream. Screening-level results work today; the detailed chain is in progress.
Serves: Whole-Building LCA · GHG-LCCA
Scoring and per-system gap tables generated from the audit run rather than typed up afterward. A working prototype: live data from my own house below, and a full walkthrough on a sample building.
Serves: Model Readiness Audit · house data ↓ · sample walkthrough →
The intent: findings returned as BCF issues that open in your authoring tool at the exact element concerned, so closing a gap is a modelling task with an address. Not built. Today the fix-list names the IFC entities and properties in writing, and you locate them yourself.
Not yet available. Listed because it's what I'm building toward.
The honest way to show what I do is to point the whole pipeline at my own house, and show you the state it's actually in, including what isn't finished yet.
/data/house/; the site and my geometry model read the same files.
Geometry rebuild in progress; readiness, LCA, energy, and GHG-LCCA stages follow. Live status loads from data/house/status.json.
Provisional screening score with a per-system gap table. Loads from data/house/readiness.json.
Four retrofit paths compared on avoided carbon and 30-year NPV. Loads from data/house/scenarios.json.
The analysis is the service; the tooling is what keeps it honest at scale. The direction: model-based exchanges where you interact with your building data directly. Digital portfolios, live readiness dashboards, and re-runnable validations instead of static PDFs.
Today that means every deliverable I produce is machine-readable and re-runnable. Tomorrow it means a client space where your models, results, and decision history live in one place. In active development.
Today you get auditable files. Next, those same deliverables live in a space you log into: a portfolio of your buildings, each with its readiness, results, and history in one place. Here's the shape of it.
● Illustrative mockup · demo data, not a real clientFIG. 02 SAMPLE PORTFOLIO · 3 BUILDINGS ILLUSTRATIVE, NOT REAL CLIENT DATA
Illustrative only, with no real projects or client data. The real client space will run as a separate, authenticated application on its own subdomain, with a managed auth provider; client files never touch this marketing site. Same boundary, enforced by architecture.
I'm Jean Carrière. Fifteen years in building energy modeling and BIM/openBIM: IFC, IDS, and the integration work that connects a building model to its carbon, energy, and cost consequences.
The niche is deliberate: most LCA practices treat the model as an inconvenience; most BIM practices stop before the carbon math. I sit exactly at the join, and build the tooling to prove it.
Based in Ontario, Canada. Working across Canadian and European projects and standards.