Project Room
Project snapshot
Baseline data for quantified consequences
About Project Room
Use this section to define the basic project context: sector, lifecycle stage, delivery model, financing model, next decision gate, timing, and baseline values.
Project Room does not produce a diagnosis by itself. It frames the evidence review, gives context to the AI Diagnostic Interview, supports quantified consequence ranges, and supplies basic project information for the final report.
Political and technical conditions entered here are treated as part of the project system. They become relevant when combined with evidence, accepted variables, mechanisms and decision gates.
Evidence & Confidence
Evidence is separated from risk severity. A risk may be high with low confidence when key documents are missing.
Evidence register
About Evidence & Confidence
Use this section to record what evidence exists and how reliable it is. You do not need to upload documents at this stage; classify the evidence base and note important gaps.
Evidence quality affects confidence, not risk severity directly. Missing evidence should not be interpreted as low risk. It means the diagnostic, scenarios and report should be read with lower confidence.
The saved evidence context is passed to the AI Diagnostic Interview and later informs confidence in the Dynamic Risk Architecture and Report Builder.
AI Diagnostic Interview
Conversation
Optional document context
Attach non-confidential text notes or public document excerpts to inform the interview. Selected files are read only after user selection and are not stored by the workspace.
How to use AI Diagnostic
Use this tab as a structured conversation with the AI. Describe the project condition, uncertainty, pressures, decision gates, or concerns in your own words. The AI will propose variables, mechanisms, assumptions to challenge, and possible measures.
Indicative information to share
- Current lifecycle stage and next decision gate.
- Political urgency, deadlines, public commitments or election timing.
- Engineering maturity, cost estimate basis, demand evidence and alternatives analysis.
- Fiscal exposure, guarantees, subsidies, availability payments or deferred obligations.
- Procurement route, contract readiness, risk allocation and change mechanisms.
- Land, permits, utilities, environmental/social approvals and stakeholder issues.
- Governance, coordination, implementation capacity and interface complexity.
- Operational model, O&M funding, operator readiness and system integration.
- Technical or political measures already being considered.
- Evidence gaps, disagreements, assumptions or unknowns.
The AI proposes; the user decides. Proposed variable updates are reviewed in the Variables tab before they update the model.
Variables
Variables begin as Not assessed. A placeholder is not a diagnosis and is not treated as an accepted score until the user accepts, edits, or manually moves the value.
AI Variable Proposals
Each AI Diagnostic iteration adds or updates variable proposals here. Nothing is applied to the model until you accept or edit the proposal. New iterations build on prior accepted assumptions instead of resetting them.
Systems Architecture Panel
Mechanisms are interpreted as feedback structures, not isolated factors. Active loops explain how risk is generated, amplified, delayed or dampened over the project lifecycle. The panel uses accepted variables from previous steps; apply or edit pending AI proposals in Variables to reduce Not assessed inputs.
Active feedback mechanisms
Loop logic
Risk migration pathway
Assumption challenges
Dynamic Risk Architecture
The 11 risks are scored by probability, impact, trend and confidence. Scores are explainable through variables, mechanisms, lock-ins and dampeners.
Quantified Consequences
Indicative scenario-based ranges. These are not forecasts; they translate the current risk architecture into consequence bands for decision review.
Top risk cluster
Decision posture
Scenario Lab
This step is for discussing decision alternatives with the AI and testing how technical and political measures change the risk architecture. Political strategies are treated as endogenous levers: they can reduce, amplify, defer or migrate risk.
AI Strategy Discussion
AI-proposed strategy portfolio
Scenario interpretation
Alternative measures to discuss or test
Measures are shown as prompts for discussion and testing, not as a static checklist. Test one measure alone, or ask the AI to combine several technical and political measures into a portfolio.
The AI should explain why a measure is useful, what variable it moves, which loop it affects, and what side effects it may create before the engine tests the scenario.
Risk movement across the 11 risks
The table shows change from the current trajectory, not isolated scenario scores.
Quantified consequence movement
Indicative consequence ranges are scenario-based and should be reviewed against project evidence.
Scenario report
Variables, loops and side effects
Report Builder
Reports are generated manually. The explicit engine remains the source for risk scores and quantified consequence ranges. The AI may draft narrative, decision framing and management focus, but it does not override accepted assumptions or scoring.