Risk, Uncertainty and Decision Analytics

Introduction and Context

Construction projects are temporary, capital-intensive and exposed to interacting technical, commercial, social, environmental and political conditions. A project may fail to achieve its objectives even when the design is technically sound and the contract is properly administered. Ground conditions may differ from the investigation; productivity may be lower than planned; prices may rise; approvals may be delayed; extreme weather may interrupt work; or a regulatory change may alter the project’s economics.

Risk management provides a systematic way of dealing with such conditions. It does not mean attempting to eliminate every uncertain event. Rather, it enables decision-makers to understand uncertainty, reduce avoidable exposure, allocate responsibility fairly, prepare responses and make transparent choices. At postgraduate level, risk management should be viewed not merely as a register of threats but as an integrated governance and decision system.

A useful distinction is that risk concerns uncertainty that can affect objectives, whereas uncertainty describes incomplete knowledge about future conditions, parameters or relationships. Risk may involve threats or opportunities. Decision analytics supplies quantitative methods for comparing alternatives under uncertainty, including expected value, decision trees, scenario analysis, Monte Carlo simulation and real-options reasoning.

Learning Objectives

After studying this chapter, the reader should be able to:

1. Explain the nature of risk, uncertainty, ambiguity and opportunity in construction.

2. Establish a risk governance structure and maintain an effective risk register.

3. Identify, assess, allocate and treat technical, commercial, safety, environmental and political risks.

4. Apply probability–impact analysis, expected monetary value, decision trees and Monte Carlo simulation.

5. Interpret scenario analysis and distinguish sensitivity from probabilistic analysis.

6. Use real-options thinking to evaluate flexibility, staged commitment and adaptive decisions.

7. Critically evaluate limitations, bias, data quality, ethics and governance in risk analysis.

1. Foundations of Construction Risk

1.1 Risk, uncertainty and ambiguity

Risk exists when alternative future states can be described sufficiently to assign probabilities, whether from historical evidence, models or expert judgement. Uncertainty is broader and includes incomplete knowledge, measurement error and model limitations. Aleatory uncertainty arises from inherent variability, such as daily weather or fluctuating productivity. Epistemic uncertainty arises from lack of knowledge, such as insufficient information about rock quality; it may be reduced through investigation.

Ambiguity occurs when different interpretations, models or definitions are plausible. For example, parties may disagree about whether a design change is a variation or a contractor obligation. Ambiguity is not resolved simply by assigning a more precise probability. It requires clarification, investigation, contractual interpretation or governance.

A risk event has a cause, an uncertain event or condition, and an effect on one or more objectives. For example:

• Cause: incomplete geotechnical investigation.
• Event: unexpected weak strata are encountered.
• Effect: additional excavation, support and delay.

Risks should be expressed in this cause–event–effect form because it supports ownership and response design.

1.2 Threats and opportunities

Traditional construction risk management concentrates on threats: cost overruns, delay, accidents, defects and disputes. A mature system also identifies opportunities, such as an alternative construction method, early procurement, design standardisation, digital fabrication or favourable market prices. The same uncertainty can create both outcomes. For instance, early bulk purchase may protect against price escalation but create storage and obsolescence exposure.

Risk appetite describes the amount and type of risk an organisation is willing to accept in pursuit of objectives. Risk tolerance defines acceptable variation around targets. Safety-critical activities normally have very low tolerance for fatality risk, whereas a contractor may tolerate moderate price variability if the expected margin is attractive.

1.3 Risk categories

A project risk breakdown structure provides a systematic classification. Typical categories include:

• Technical: design errors, constructability, interfaces, ground conditions, performance failure and technology maturity.
• Commercial and financial: inflation, exchange rates, interest rates, insolvency, claims, payment delays, inaccurate quantities and price volatility.
• Programme and resource: productivity, labour availability, equipment breakdown, logistics and long-lead materials.
• Health and safety: falls, lifting operations, traffic interfaces, temporary works, hazardous substances and occupational exposure.
• Environmental and social: pollution, biodiversity, noise, waste, climate hazards, community opposition and land access.
• Legal, contractual and regulatory: ambiguous clauses, permits, changes in law, insurance gaps and dispute resolution.
• Political and country: expropriation, sanctions, civil unrest, policy change, corruption and restrictions on imports or currency conversion.
• Organisational and information: weak competence, poor communication, fragmented procurement, cyberattack and unreliable data.

Categories must not become isolated silos. A design change may generate procurement delay, cost growth, safety exposure and a dispute. Risk interactions are often more important than individual risks.

2. Risk Governance and the Risk Process

Risk governance establishes who is accountable, how decisions are escalated and how risk information is challenged. It should operate across the project life cycle, from feasibility through operation and handover.

2.1 Governance structure

A robust governance arrangement normally includes:

• Project sponsor or investment committee: sets strategic objectives, risk appetite and approval thresholds.
• Project director: integrates risk with scope, cost, schedule, procurement and stakeholder decisions.
• Risk manager or risk lead: maintains methodology, facilitates workshops, consolidates analysis and reports trends.
• Risk owners: have authority and accountability for managing specified risks.
• Control owners: implement particular preventive or mitigative actions.
• Independent assurance: reviews assumptions, models, controls and compliance.
• Steering committee: resolves escalated risks and authorises contingency or changes.

Risk governance should specify reporting frequency, escalation thresholds, independence requirements, documentation standards and the relationship between risk management, change control, safety management and internal audit.

2.2 The risk management cycle

The process is iterative:

1. Establish objectives, context, constraints and risk criteria.

2. Identify sources, causes, events, consequences and opportunities.

3. Assess probability, impact, time profile, velocity and interdependence.

4. Select responses and allocate ownership.

5. Implement controls and contingency actions.

6. Monitor indicators, reassess exposure and communicate decisions.

7. Capture lessons and update estimates, assumptions and the risk register.

Risk should be addressed at stage gates. A risk that is inexpensive to investigate during concept design may become costly to change after procurement or construction.

2.3 The risk register

A risk register is a controlled management record, not a static list. A useful entry contains:

• unique identification number and category;
• cause, risk event and consequence;
• affected objectives and interfaces;
• inherent probability and impact;
• existing preventive and mitigative controls;
• residual probability and impact after controls;
• risk owner and action owner;
• response strategy, action, deadline and required resources;
• trigger or early-warning indicator;
• contingency and escalation threshold;
• status, trend, review date and closure evidence.

Qualitative ratings should use defined scales. For example, probability may range from rare to almost certain, while impact may be rated from insignificant to catastrophic. Each level needs written criteria, such as cost percentage, days of delay, injury severity or environmental consequence. Without calibration, a score of “4” has little shared meaning.

Risk registers have limitations. They can encourage false precision, duplicate correlated risks and focus attention on visible events rather than systemic conditions. The register should therefore be supplemented by assumptions logs, interface registers, opportunity registers, issue logs, quantitative models and periodic facilitated reviews.

3. Identification, Assessment and Allocation

3.1 Identification methods

Identification techniques include document review, structured interviews, expert elicitation, brainstorming, prompt lists, design reviews, constructability reviews, hazard and operability studies, failure-mode analysis, historical data, site reconnaissance, stakeholder mapping and lessons-learned databases. Workshops should include designers, constructors, operators, commercial specialists, safety professionals and affected stakeholders.

A good workshop distinguishes causes from consequences. “Delay” is usually an impact, not a sufficiently defined risk event. The analysis should ask: delay caused by what, affecting which activity, under what conditions, and controlled by whom?

3.2 Probability–impact analysis

For a discrete risk, a basic expected monetary value (EMV) is:

EMV = pC

where p is the probability of occurrence and C is the monetary consequence. For a threat, C is generally positive when expressed as a loss; for an opportunity, the benefit may be represented with a negative cost or separately as an expected benefit. The calculation assumes that the probability is meaningful, the consequence is appropriately defined, and the event is sufficiently discrete. It is not a forecast of the actual outcome of one project; it is an average value across repeated comparable situations.

If several mutually exclusive outcomes have consequences Cᵢ and probabilities pᵢ, then:

EMV = Σ pᵢCᵢ, with Σ pᵢ = 1.

For a qualitative matrix, a risk score is often calculated as:

Score = likelihood rating × impact rating.

This is an ordinal prioritisation device, not a monetary expectation. Multiplying ordinal labels can imply mathematical precision that the scales do not possess. A low-probability catastrophic event may deserve priority even if its score is below a frequent minor event. Safety and legal criteria may therefore override the numerical ranking.

3.3 Controls and residual risk

Preventive controls reduce the probability of an event; mitigative controls reduce its consequence after occurrence. For example, additional geotechnical investigation is preventive, while a contingency excavation method is mitigative. Risk assessment should distinguish inherent exposure before controls from residual exposure after controls. Controls must be tested for effectiveness rather than merely recorded as planned.

3.4 Allocation and contracting

Risk should be allocated to the party best able to influence its probability or consequence, insure it efficiently, absorb it at reasonable cost or obtain information about it. Allocation solely to the party with the greatest bargaining power is often inefficient and may produce inflated prices, defensive behaviour and disputes.

Contract mechanisms include clear scope definition, allowances, provisional sums, price-adjustment clauses, differing-site-condition provisions, insurance, performance security, liquidated damages, relief events, compensation events, target-cost pain/gain sharing and force-majeure provisions. Allocation must be consistent with actual control. A contractor cannot fairly price an unforeseeable statutory change that only the client or government can control without an appropriate adjustment mechanism.

4. Decision Analytics

4.1 Expected monetary value and expected utility

Suppose an option has outcomes xᵢ with probabilities pᵢ. Its expected monetary value is:

E[X] = Σ pᵢxᵢ.

EMV is appropriate when outcomes are measured consistently, probabilities are credible, and the decision-maker is approximately risk-neutral or is comparing financial expectations as one input. It can be misleading for safety, reputation, insolvency or irreversible outcomes.

Expected utility addresses risk attitude:

E[U(X)] = Σ pᵢU(xᵢ),

where U(x) is the decision-maker’s utility function. A risk-averse utility function is concave: the utility gained from an additional monetary unit decreases as wealth increases. Utility analysis requires explicit assumptions about preferences and should not be used to monetise human life or legal rights in a way that undermines non-negotiable duties.

4.2 Worked example: expected value

A contractor is choosing between two temporary works strategies. Strategy A costs £1.20 million with certainty. Strategy B costs £0.90 million initially; with probability 0.25, instability causes an additional £1.60 million cost, and with probability 0.75 no additional cost occurs.

The expected cost of B is:

E(B) = 0.75(£0.90m) + 0.25(£2.50m)

= £0.675m + £0.625m

= £1.300m.

The expected cost of A is £1.20 million, so A has the lower expected monetary cost by £0.10 million. However, the decision should also consider safety, delay, reputational effects, constructability and the reliability of the 0.25 probability. If instability could cause unacceptable harm, expected cost alone cannot justify B.

4.3 Decision trees

A decision tree represents decisions as squares, chance events as circles and outcomes at terminal branches. It is evaluated from right to left. At a chance node, calculate the probability-weighted value of branches. At a decision node, select the alternative with the preferred value, subject to constraints and risk criteria.

Consider a geotechnical investigation costing £80,000. Without investigation, there is a 0.30 probability of an adverse condition costing £600,000 and a 0.70 probability of a £100,000 treatment. The expected cost without investigation is:

0.30(£600,000) + 0.70(£100,000) = £250,000.

If investigation reduces the adverse-condition probability to 0.10 and treatment remains £100,000 when conditions are favourable, the expected post-investigation treatment cost is:

0.10(£600,000) + 0.90(£100,000) = £150,000.

Adding investigation gives £230,000, which is £20,000 lower than proceeding without it. The expected value of information is therefore £20,000 under these simplified assumptions. The calculation must also include investigation duration, reliability, possible redesign and the value of avoiding safety exposure before approval.

4.4 Scenario and sensitivity analysis

Sensitivity analysis changes one input at a time, or uses controlled variations, to identify drivers of an output. A tornado diagram can rank the influence of quantities, rates, productivity, duration or escalation. Sensitivity does not assign likelihood and may ignore interactions.

Scenario analysis evaluates coherent combinations of assumptions, such as baseline, high-inflation, delayed-approval and severe-weather scenarios. Scenarios are especially useful for strategic and political uncertainty, where precise probabilities are not defensible. A scenario should describe its assumptions, internal logic, indicators and management response.

4.5 Monte Carlo simulation

Monte Carlo simulation repeatedly samples uncertain inputs from specified probability distributions and calculates the resulting project cost, duration or other output. If cost is modelled as:

C = Σ QᵢRᵢ + D + K,

where Qᵢ is quantity, Rᵢ is unit rate, D is defined risk allowance and K is fixed cost, simulation samples the uncertain quantities, rates and risk events and recalculates C for each iteration. The output is a distribution rather than a single estimate.

A generic procedure is:

1. Define the output and model boundaries.

2. Identify uncertain inputs and select distributions using evidence.

3. Represent correlations and conditional relationships.

4. Sample inputs using random or quasi-random values.

5. Run many iterations and store outputs.

6. Examine percentiles, mean, spread, tail outcomes and sensitivity.

7. Validate assumptions and use results to set contingency and decisions.

Common distributions include triangular distributions when minimum, most likely and maximum estimates are available, and empirical or parametric distributions when reliable data exist. A triangular distribution is not automatically superior; its shape should reflect evidence and expert reasoning.

For a simulated output, the P50 is the value not exceeded in 50% of simulations, while P80 is not exceeded in 80%. Percentiles are not guarantees. A P80 cost estimate still has a 20% modelled probability of being exceeded. Contingency may be expressed as the difference between a selected percentile and the deterministic base estimate, but this depends on what the base estimate already includes.

Correlations are essential. Labour productivity and duration, quantities and costs, or weather and productivity may move together. Treating correlated variables as independent can understate or distort uncertainty. Double counting is another serious risk: a risk register allowance and an uplift in the cost distribution may represent the same exposure.

Simulation output should be subject to model assurance, data lineage, version control, independent review and back-testing where possible. A sophisticated model built on weak assumptions produces precise-looking but unreliable results.

5. Real-Options Thinking

A conventional discounted cash-flow analysis often assumes that a plan is selected now and executed passively. Construction decisions frequently create flexibility: investigate before committing, stage a project, preserve the option to expand, switch suppliers, defer procurement or abandon an uneconomic phase.

A real option is the value of managerial flexibility under uncertainty. A staged development may cost more than immediate full commitment but allow the sponsor to proceed only if demand, approvals or ground information are favourable. The value of the option increases when uncertainty is high, future decisions can be changed, downside exposure is limited and information arrives before irreversible commitment.

Real-options thinking does not imply that every flexible feature should be purchased. The premium for flexibility, delay cost, interface effects and operational consequences must be compared with the value of improved decisions. Options may be contractual, physical or organisational. Examples include break clauses, framework agreements, modular design, expandable foundations, dual sourcing and pilot construction.

A simple decision rule is to compare the value of an adaptive strategy with the value of a fixed strategy after including the cost of flexibility. More advanced valuation may use decision trees, binomial lattices or option-pricing concepts, but financial option formulas should not be transferred uncritically to construction because construction assets are often illiquid, non-tradable and affected by multiple correlated uncertainties.

6. Applied Case: Illustrative Urban Rail Interchange

Background

An illustrative public authority is procuring a £420 million urban rail interchange involving deep excavation, utility diversion, complex interfaces and a compressed opening date. The project is delivered through a target-cost arrangement with multiple design and construction organisations.

Problem

Early reviews identify uncertain ground conditions, utility records of inconsistent quality, inflation in structural steel, restricted possession windows, community concerns about noise and a possibility of revised accessibility regulations. A simple deterministic estimate shows a £420 million cost and a 48-month programme, but the sponsor needs confidence levels and a defensible treatment plan.

Analysis

The team creates a risk breakdown structure and separates causes from events. A geotechnical risk is expressed as: incomplete borehole coverage causes unexpected groundwater and weak strata, resulting in additional support, pumping and delay. The risk owner is the design-and-construction manager, while the sponsor owns residual funding exposure.

A qualitative matrix identifies groundwater, possession access and utility diversion as high-priority risks. A quantitative model then represents excavation quantities, steel prices, productivity, possession availability and discrete design changes. Correlations are included between possession availability and productivity. The model excludes risks already embedded in the base estimate and documents this boundary.

Scenario analysis examines normal regulation, delayed regulation and a severe political-pressure scenario in which the opening date is brought forward. A decision tree compares immediate full excavation with a staged investigation and pilot excavation. Although the staged approach costs £1.8 million and adds six weeks initially, it substantially reduces the probability of a major groundwater claim. The sponsor selects the staged approach because it purchases information before the most irreversible commitment.

Approach and solution

The contract includes a clear differing-site-condition mechanism, transparent measurement rules, early-warning obligations and a jointly governed contingency. Utility surveys are expanded, temporary works receive independent checking, and possession risks are monitored through a rolling 90-day look-ahead. Steel procurement uses an agreed price-adjustment mechanism rather than transferring all market risk to a party unable to control it.

The risk register is reviewed monthly, while critical safety and possession risks are reviewed weekly. Triggers include groundwater levels, unresolved utility clashes, steel-index movements and approval dates. Escalation thresholds are linked to forecast cost, schedule float and safety consequence rather than to risk scores alone.

Outcome

The staged investigation identifies a high-permeability layer before the main excavation. The design is modified, reducing expected pumping and delay exposure. The project still experiences steel-price growth, but the contractual mechanism and early procurement limit dispute. The case demonstrates that risk management is most valuable when it changes decisions, not when it merely records risks.

Learning points

1. Investigation can have positive expected value even when it increases immediate cost.

2. Allocation should follow control and information, not simply bargaining power.

3. Risk registers, quantitative models and contract mechanisms must be consistent.

4. Flexibility has value when it permits learning before irreversible commitment.

5. Governance must monitor triggers and actions, not only numerical scores.

7. Limitations, Ethics and Contemporary Practice

Risk analysis is vulnerable to optimism bias, anchoring, availability bias, confirmation bias and strategic misrepresentation. Promoters may understate cost and duration to obtain approval, while contractors may overstate uncertainty to protect margin. Independent challenge, reference-class forecasting, documented assumptions and separation of estimating from approval can reduce these problems.

Probabilities may be based on sparse data or expert judgement. Experts should state confidence, evidence and disagreement rather than provide unsupported point estimates. Model uncertainty should be distinguished from variability. Where probability is genuinely unknowable, scenarios, robust decisions and adaptive monitoring may be more defensible than artificial precision.

Safety, environmental protection, human rights and legal compliance should not be traded away through expected-value calculations. A low-probability fatality cannot be made acceptable merely because its expected financial value is small. Ethical governance requires transparency, competence, protection of whistle-blowers, avoidance of conflicts of interest and meaningful engagement with affected communities.

Digital risk platforms can connect registers with cost, schedule, BIM, procurement and field data. Artificial intelligence may support pattern detection, document review, predictive maintenance and anomaly identification. However, AI outputs inherit data bias, may confuse correlation with causation and can conceal assumptions in opaque models. Human accountability, explainability, cybersecurity, data provenance and independent validation remain necessary. AI should support—not replace—professional judgement and statutory responsibility.

Climate change increases the importance of non-stationary hazards. Historical weather frequencies may not represent future conditions, so resilience decisions should consider climate projections, adaptation pathways and the consequences of failure. Contemporary best practice also treats supply-chain fragility, cyber risk, modern slavery, biodiversity and whole-life carbon as interconnected project risks.

Best-Practice Checklist

• Define objectives, risk appetite and decision criteria at project initiation.
• Use a common risk breakdown structure and controlled terminology.
• Record causes, events, consequences, owners, controls, triggers and actions.
• Calibrate qualitative scales and avoid treating scores as precise probabilities.
• Integrate risk with estimating, scheduling, procurement, safety and change control.
• Test controls and distinguish inherent from residual exposure.
• Use quantitative analysis where decisions justify the effort and data support it.
• Model correlations, avoid double counting and document exclusions.
• Present ranges, percentiles and assumptions rather than false certainty.
• Apply independent assurance and challenge optimism or strategic bias.
• Allocate risk to the party best able to manage it, with fair adjustment mechanisms.
• Preserve flexibility when information is expected to improve before commitment.
• Review triggers, trends and action effectiveness throughout the life cycle.

Chapter Summary

Construction risk management is an integrated governance process for dealing with threats and opportunities under uncertainty. Effective practice begins with clear objectives, risk appetite, structured identification and a live risk register. Risks should be described through causes, events and consequences, assessed using calibrated criteria, controlled through preventive and mitigative measures, and allocated according to control and information.

Probability–impact analysis supports prioritisation, while expected monetary value and expected utility support alternative evaluations under stated assumptions. Decision trees clarify sequential decisions and the value of information. Sensitivity analysis identifies influential variables; scenario analysis explores coherent futures; and Monte Carlo simulation produces distributions of cost, duration or other outcomes. These methods are only as reliable as their data, assumptions, correlations and governance.

Real-options thinking recognises that staged commitments, investigation, modularity and contractual flexibility can have value when uncertainty is high. Quantitative analysis must be complemented by safety, legal, environmental and ethical judgement. Modern risk governance increasingly incorporates climate change, supply-chain resilience, cybersecurity, digital data and AI, while retaining human accountability and independent assurance.

Key Terms

Risk: Uncertainty that affects project objectives.

Uncertainty: Incomplete knowledge or variability about future conditions.

Ambiguity: Uncertainty arising from competing interpretations or models.

Aleatory uncertainty: Inherent variability in a process or environment.

Epistemic uncertainty: Uncertainty caused by limited knowledge.

Risk appetite: Amount and type of risk an organisation is willing to pursue or retain.

Risk register: Controlled record of risks, assessments, owners, controls and actions.

Risk owner: Person accountable for managing a specified risk.

Risk breakdown structure: Hierarchical classification of project risk sources.

Inherent risk: Exposure before considering controls.

Residual risk: Exposure remaining after controls.

Expected monetary value: Probability-weighted average monetary consequence.

Expected utility: Probability-weighted value based on a utility function.

Decision tree: Graphical representation of sequential decisions and chance events.

Sensitivity analysis: Study of output response to changes in input variables.

Scenario analysis: Evaluation of coherent sets of alternative assumptions.

Monte Carlo simulation: Repeated random sampling used to generate an outcome distribution.

Real option: Value arising from flexibility to defer, stage, expand, switch or abandon.

Contingency: Allowance for specified or modelled uncertainty.

Risk trigger: Observable condition indicating that a response or escalation is required.

Review Questions

Short Answer Questions

1. Distinguish risk, uncertainty and ambiguity in a construction context.

2. What information should a well-constructed risk-register entry contain?

3. Differentiate preventive controls from mitigative controls.

4. State two limitations of probability–impact matrices.

5. Why are correlations important in Monte Carlo simulation?

Long Answer Questions

1. Explain how risk governance should be structured across a major infrastructure project.

2. Compare expected monetary value, expected utility and qualitative risk scoring.

3. Describe the procedure for developing and assuring a Monte Carlo cost model.

4. Explain how contractual risk allocation can create either cooperation or disputes.

5. Discuss real-options thinking and its relevance to staged construction investment.

Analytical / Application Questions

1. A project option costs £2.4 million with certainty. An alternative costs £1.7 million initially and has a 0.20 probability of an additional £4.0 million cost. Calculate the expected cost of the alternative and recommend an option, identifying non-financial factors that could change the decision.

2. Design a risk register entry for unexpected groundwater during a deep excavation. Include cause, event, consequences, controls, owner, trigger, response and escalation criteria.

3. A sponsor must choose between immediate full commitment and a six-week investigation that costs £250,000. Explain how you would use a decision tree and value-of-information analysis, and identify circumstances in which a staged option should be selected even if its expected monetary value is not the highest.

Multiple Choice Questions with Answers

Question 1

Which statement best distinguishes epistemic uncertainty from aleatory uncertainty?

A. Epistemic uncertainty is inherent variability, whereas aleatory uncertainty is lack of knowledge.

B. Epistemic uncertainty arises from limited knowledge, whereas aleatory uncertainty is inherent variability.

C. Epistemic uncertainty applies only to safety, whereas aleatory uncertainty applies only to cost.

D. There is no meaningful distinction between them.

Correct Answer: B. Epistemic uncertainty arises from limited knowledge, whereas aleatory uncertainty is inherent variability.

Question 2

A risk has probability 0.20 and a monetary consequence of £500,000. What is its expected monetary value?

A. £100,000

B. £250,000

C. £400,000

D. £2,500,000

Correct Answer: A. £100,000

Question 3

Which item is primarily a mitigative control?

A. Additional site investigation before design finalisation

B. A contingency pumping system activated after groundwater ingress

C. A constructability review before tender

D. A contract clarification issued before award

Correct Answer: B. A contingency pumping system activated after groundwater ingress

Question 4

What does a P80 cost estimate mean?

A. The estimate will never be exceeded.

B. The estimate is 80% higher than the base estimate.

C. The modelled probability of not exceeding the estimate is 80%.

D. The estimate has an 80% profit margin.

Correct Answer: C. The modelled probability of not exceeding the estimate is 80%.

Question 5

Why should correlations be represented in a Monte Carlo model?

A. They ensure every input has a normal distribution.

B. They prevent the need for sensitivity analysis.

C. They reflect variables that tend to change together and affect output uncertainty.

D. They guarantee that the mean output equals the deterministic estimate.

Correct Answer: C. They reflect variables that tend to change together and affect output uncertainty.

Question 6

Which principle is most appropriate for allocating project risk?

A. Allocate every risk to the contractor.

B. Allocate risk to the party best able to control or manage it.

C. Allocate risk to the party with the smallest organisation.

D. Allocate risk randomly to avoid negotiation.

Correct Answer: B. Allocate risk to the party best able to control or manage it.

Question 7

What is the principal purpose of a decision tree?

A. To replace all professional judgement with a formula

B. To represent sequential decisions, chance events and outcomes

C. To calculate only qualitative risk scores

D. To eliminate uncertainty from a project

Correct Answer: B. To represent sequential decisions, chance events and outcomes

Question 8

Which statement correctly compares sensitivity and scenario analysis?

A. Sensitivity analysis usually changes inputs to identify influential drivers; scenario analysis evaluates coherent sets of assumptions.

B. Sensitivity analysis assigns legal responsibility; scenario analysis calculates insurance premiums.

C. Sensitivity analysis always uses probability distributions; scenario analysis never uses assumptions.

D. They are identical methods with different names.

Correct Answer: A. Sensitivity analysis usually changes inputs to identify influential drivers; scenario analysis evaluates coherent sets of assumptions.

Question 9

A real option in construction is best illustrated by:

A. Ignoring future information after approving the project

B. Committing all funds immediately regardless of conditions

C. Staging investment so that later commitment depends on information obtained earlier

D. Replacing a risk register with a fixed contingency

Correct Answer: C. Staging investment so that later commitment depends on information obtained earlier

Question 10

Which statement about expected monetary value is correct?

A. It predicts the exact outcome of a single project.

B. It is a probability-weighted average and may be insufficient for safety-critical decisions.

C. It is valid only when all outcomes are certain.

D. It automatically accounts for risk aversion and ethical constraints.

Correct Answer: B. It is a probability-weighted average and may be insufficient for safety-critical decisions.