The Hong Kong Polytechnic University · FACULTY OF MANAGEMENT

ISE548 Chap.5 Risk Assessment and Matrix Decisions

- one subject, every graph, every model, every mark
6 Chapters4-page Bible
Our own words - no uploaded lecturer files
Updated for this semester
Chapter 5 of 12 · ISE548

Risk Assessment and Matrix Decisions

A matrix is a routing device

Risk assessment identifies hazards, estimates likelihood and severity, and evaluates which exposures require action. This chapter defines scales before scoring, records existing controls and keeps the two axes visible beside any product or colour. It also explains organisational acceptability, consensus, implementation and event-driven review.

The aim is a reproducible decision trail, not a cell that appears precise because it has a number.

Every placement needs an event definition, scale anchors and evidence trail. Existing controls must be stated before deciding whether the score represents inherent or residual exposure.

The matrix then routes action through an organisational rule; it does not predict when an event will occur or prove that a low cell requires no review.

In this chapter

What this chapter covers

  • 01

    Assessment preparation and evidence

  • 02

    Likelihood scale definitions

  • 03

    Severity scale definitions

  • 04

    Matrix placement and acceptability

  • 05

    Control-driven score movement

  • 06

    Implementation and review triggers

Worked example · free

Re-score a pallet-mover contact

Q [8 marks]. An event is occasional and moderate. A speed limiter reduces impact. Show the before and after placement. The weighting shown is not an official University mark allocation.
Likelihood and severity meet in a three-by-three decision matrix.Likelihood and severity meet in a three-by-three decision matrix. The coloured cells encode the course's low, medium and high priorities, not universal physical quantities.LowLowMediumLowMediumHighMediumHighHighRemoteOccasionalFrequentMinorModerateMajorLikelihoodSeverity
Likelihood and severity meet in a three-by-three decision matrix. The coloured cells encode the course's low, medium and high priorities, not universal physical quantities.
  • 3Place the initial event with evidence.
  • 2Identify the dimension changed.
  • 3State residual exposure and trigger.
Occasional likelihood and moderate consequence place the event in the medium cell of the three-level matrix. A tested limiter reduces severity to minor while likelihood remains occasional, moving it to low. Contact still occurs, so collision injury, layout change or rising near misses should trigger review.
Sia tip — Move one axis only when one mechanism changed.
Glossary

Key terms

Likelihood
The assessed chance or frequency of an event under a stated scale and time basis.
Severity
The degree of harm or loss associated with the defined consequence.
Risk Matrix
A table combining likelihood and severity categories to route management attention.
Acceptability
An organisational decision about whether exposure may be retained under stated conditions.
FAQ

Risk Assessment and Matrix Decisions FAQ

Why can two teams give the same product different meanings?

A product can arise from frequent limited harm or rare severe harm. Those pairs need different evidence and treatment, so preserve likelihood and severity beside the product rather than allowing multiplication to erase the pattern.

Should every control lower both axes?

No. Prevention may lower likelihood, containment may lower consequence and detection may change neither matrix axis directly. Credit only the movement supported by the control mechanism and performance evidence.

When should a completed assessment be reviewed?

Review when new risk information appears, requirements or procedures change, an accident or serious incident occurs, or control evidence fails. Record the trigger and owner when the assessment is approved.

How should likelihood categories be anchored?

Define each category with an observation period, population and evidence source. Rare may mean fewer than one event in a stated number of operations, while occasional may rely on a documented frequency band. Avoid mixing feelings, annual counts and per-transaction rates in one scale. When evidence supports only broad distinctions, use fewer categories rather than inventing precision.

Why keep likelihood and severity beside the product?

The same product can arise from frequent limited disruption and rare severe harm. Those patterns need different controls, escalation and assurance. Retaining the pair shows which dimension a proposed action changes and prevents equal arithmetic from being mistaken for equal management meaning. If a matrix uses colours without a product, the same rule applies: preserve both axis judgements.

How should disagreement between assessors be handled?

Compare event definitions, operating states and evidence rather than averaging scores immediately. Operations may cite near misses while engineering cites recorded component failures; both can be correct about different indicators. Score plausible interpretations, identify whether they change the action and gather evidence targeted at the disputed boundary.

Record the minority assumption when it would produce a different treatment.

What belongs in a residual-risk record?

Keep the original event, consequence and scale, then name the implemented control, verification evidence and axis that changed. Add the new score, remaining exposure, accepting authority, action owner and review trigger. If the control affects only consequence, likelihood should not fall through optimism. A proposed control supports a forecast, while residual risk requires evidence of operation.

How can sensitivity improve a matrix decision?

Place the event under each plausible category definition and compare the required actions. If the same treatment follows, the decision is robust even though the exact cell is uncertain. If the action changes, identify the boundary evidence that would resolve the placement. Sensitivity prevents teams from debating a colour when the real question is whether more evidence would alter priority, timing or authority.

Study strategy

Exam move

Create one three-level likelihood scale and one three-level severity scale with evidence anchors. Score three events, then add one control to each and move only the supported axis. Finish every row with the action owner, target date and observation that would reopen the assessment.

Working through Risk Assessment and Matrix Decisions in ISE548? Sia is AskSia’s AI Management tutor — ask any ISE548 Risk Assessment and Matrix Decisions question and get a clear, step-by-step explanation grounded in how ISE548 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.

A+Everything unlocked
Unlocks this Bible + all 24 of your subjects at The Hong Kong Polytechnic University, plus 1,000+ more Bibles across the AskSia library.
Sia - your ISE548 tutor, unlimited, worked the way the exam marks it
The full 4-page Bible + practice bank with worked solutions
Chrome extension - sync your LMS so Sia knows your deadlines
Bilingual EN / Chinese on every Bible and every Sia answer
$0.99 Trial
30-day money-back · cancel in one tap · how it works