ENGM60451 Chap.3 Infrastructure Risk and Project Evidence
Infrastructure Risk and Project Evidence
Infrastructure Risk and Project Evidence treats an estimate as a claim about scope, date, maturity and delivery conditions. The opening material links inaccurate cost and schedule estimates to public confidence while noting the effect of consultation and legal process. A credible comparison therefore aligns asset boundaries, price bases, design maturity, contingency and governance routes before ranking options.
Uncertainty should be named by source and connected to the choice, not hidden behind precision. Consultation can supply local knowledge and reveal implementation risk, while decision records must show how evidence was checked and balanced.
Benefits require the same discipline as costs. Demand, reliability and regional effects need a counterfactual, date and uncertainty range rather than an optimistic point value.
Governance should connect each material assumption to an owner and review trigger. When scope or evidence changes, the decision model must be rerun instead of allowing an old recommendation to survive behind updated narrative.
What this chapter covers
- 01
Estimate scope
- 02
Price basis
- 03
Design maturity
- 04
Consultation evidence
- 05
Uncertainty ranges
- 06
Decision review
Worked example · free
Audit two bridge estimates
- 2Align scope by identifying utility work, interfaces and exclusions in both proposals.
- 2Align price date, design maturity, risk allocation and uncertainty before comparing totals.
- 2Recommend the next evidence step and state the result that would reverse the ranking.
Key terms
- Estimate basis
- The scope, price date, assumptions and maturity underlying an estimate.
- Contingency
- Allowance associated with uncertainty and identified risk in the stated estimate basis.
- Decision trigger
- Evidence or a milestone that requires a choice to be reviewed.
Infrastructure Risk and Project Evidence FAQ
Why can two project estimates be incomparable?
They may use different scope, price dates, design maturity, contingency, interfaces or governance assumptions. Converting both to one headline unit does not repair those differences. A like-for-like comparison begins by normalising the estimate basis and exposing remaining uncertainty.
Is consultation only about preferences?
No. Consultation can reveal factual information about access, drainage, operations, local use and implementation risk. Those claims should be verified like other evidence. Preference and distributional judgement still matter, but they should not be confused with technical facts supplied through engagement.
How should uncertainty be written?
Name the source, plausible range and decision consequence. Ground conditions, demand, supply capacity, regulation and interfaces create different uncertainties. The important question is whether a plausible change would alter the preferred option, mitigation or timing, and what evidence will trigger review.
What should an estimate reconciliation contain?
Put the alternatives on a common basis before comparing totals. Record asset scope, interfaces, utilities, land, price date, inflation treatment, design maturity, risk allowance, financing assumptions and delivery stage. Keep exclusions visible rather than covering them with a generic uplift.
For every material uncertainty, describe how it could change cost, schedule, benefit or consent and identify the evidence that would narrow it. Consultation can reveal local operating and implementation constraints, but its claims still need verification. Link major assumptions to owners and decision gates.
When scope or evidence changes, update the numerical case and recommendation together so an earlier conclusion cannot survive merely because the surrounding narrative has been revised.
Exam move
Create an estimate-audit table with rows for asset boundary, interfaces, exclusions, price date, design maturity, contingency, risk allocation and approval route. Use it before comparing any two figures. Practise converting a confident headline into a bounded claim that states what is included and which uncertainty remains.
For consultation examples, classify each contribution as factual evidence, preference, value judgement or proposed mitigation, then explain how it enters the decision. Draw an evidence funnel from observations and estimates through assumption checks to a bounded recommendation.
Review common distractors: the lowest estimate is not automatically best value, delay is not automatically a technical failure, and a wider range is not automatically weaker analysis. Write one original project scenario in which a hidden exclusion reverses the initial ranking. Add a review trigger such as a design milestone, consent outcome or supplier test.
When working in a group, assign one person to compare bases across every quantitative claim rather than letting each section use its own definitions. Finish with a recommendation that is firm about the next action and honest about the evidence that could change the final choice. Add a reference-class drill that compares projects on one driver at a time and states why the comparator is relevant.
Build a counterfactual for each benefit claim so activity without the project is not credited to the intervention. Practise versioning assumptions: change one demand, scope or delivery value and trace every conclusion that depends on it. During review, ask which single assumption is most capable of reversing the ranking and commission a focused independent challenge. Record the response, not merely that assurance occurred.
End by writing a decision log with approver, outstanding evidence and reopening condition so later readers can judge the choice against what was known at the time.