Construction Risk,
Quantified
Integrated QRA and SRA for capital projects
A register-driven cost and schedule risk platform. Excel in front, a Python simulation engine behind, and a live dashboard on top.
Tropic Ridge runs in two stages: a pre-construction QRA/SRA baseline that sizes contingency before the bid goes in, then monthly active management once construction starts — which is what the live dashboard shows.
Tropic Ridge Wind Farm is a synthetic demonstration project • The methodology is the one run on a live infrastructure engagement
What It Does
Every published P-value traces back to a named risk in a named cell
Pre-construction: the bid-stage baseline
Register to EMV to two Monte Carlo engines to a joint P80. This is the workbook, the blog post and the deck.
One Register, Two Engines
The risk register stays in Excel, where the risk manager works. Both the cost and schedule simulations read from it directly — no re-entry, no second source of truth.
Full Critical Path Replay
The schedule engine parses the programme and re-solves the critical path inside every iteration. Risks landing on float get absorbed; risks on the driving path move the finish date.
Cost and Schedule, Together
Percentiles do not add. The platform takes the P80 after summing each iteration's cost and schedule outcome — not by adding two separate P80s, which overstates contingency.
EMV Cross-Check
Expected monetary value is the first-pass answer and the standing reasonableness check. Every simulation mean is reconciled back against it.
Construction: monthly active management
Once the project is building, the baseline becomes a monitoring cycle. This is what the live dashboard shows.
Auditable, Not a Black Box
Every run is seeded and stamped with its iteration count, inputs and engine version. The same register produces the same figures next month, or the difference has to explain itself.
Active Management Cycle
A contingency study is a photograph. The monthly cycle re-runs the register and attributes every movement in the P80 to the risk that caused it.
Two Views, One Model
The dashboard opens on a project manager view — movement, drivers and what changed this month. An analyst view sits behind it with the distributions and run settings underneath those numbers.
How It Works
From risk register to a defensible contingency figure
Maintain the register
Risks, quantification, correlations and activity mapping live in one Excel workbook, owned by the risk manager. It runs a full Monte Carlo QRA standalone, without a Python install.
Simulate both arms
The Python model adds the schedule side, reading the P6 programme and replaying the critical path across 100,000 iterations, with dual-coded risks sharing a single severity draw.
Report, then hand over
S-curves, tornado charts and criticality rankings publish to spreadsheet and notebook. That baseline is what the bid carries — and what the monthly dashboard cycle tracks against once construction starts.
Built With
Excel + xlwings lite
Risk register and standalone QRA engine
Vectorised Python
Monte Carlo and full CPM schedule analysis
Next.js + Vercel
Risk analyst and project manager dashboards
Quarto
Reproducible reporting straight from the model
Every production run: 100,000 seeded iterations, cross-validated between two independently built engines
Is your contingency a round number?
Start with the bid-stage deck to see how the baseline is built, then walk the dashboard to see how it is managed once construction is under way.
Synthetic demonstration data • No client information