Independent SHAREhub process-safety awareness. Selected learning maps to a numbered process-safety syllabus topic; this is not the NEBOSH qualification, approved provision, specialist design instruction or work authorisation. GB guidance is distinguished from corresponding NI requirements; establish applicability for the actual jurisdiction and establishment.
This reader supplies the complete teaching. Use the interactive LMS course for assessed activities, saved progress and completion.
LESSON 1
Separate the source from the risk
Distinguish a hazard, event and risk statement.
Existing SHAREhub photographic illustration. The written fictional case supplies the condition and decision evidence.
Separate the source from the risk
A hazard is a source of potential harm. Risk concerns the likelihood and consequences of a defined harmful event under stated conditions. Naming a substance or writing “high risk” does not explain the event or the control question.
Describe the activity and loss of control first: a transfer enters the wrong receiving system, for example. Then identify credible causes, consequences and affected people. The description should be detailed enough to examine safeguards without inventing plant facts. A matrix colour becomes meaningful only after the scenario and assessment basis are clear.
What is the difference between a hazard and a risk?
A hazard has the potential to cause harm, while risk is the chance and severity of harm in the actual circumstances.
This distinction matters because many weak assessments list hazards but never explain how people could be harmed. A forklift is a hazard; the risk may be pedestrian impact at a blind corner during peak loading. Silica dust is a hazard; the risk may be a dry-cutting task in a poorly ventilated room with no water suppression or suitable respiratory protective equipment.
A useful assessment links the source of harm to the person, task, exposure route, initiating event, existing controls and remaining risk. That is what turns a document into a decision tool.
Connect cause, event and consequence without claiming certainty.
Existing SHAREhub 2D teaching schematic. Use alongside the explanation and fictional case in this lesson.
Write a scenario that can be tested
A useful scenario explains how a hazardous source could reach an unwanted condition and cause harm. Separate the initiating cause from loss of control and its consequences. This helps the team locate safeguards at the right point in the pathway.
Credibility needs process knowledge and evidence, not the most dramatic story. State assumptions and uncertainties. An unidentified route can justify examination of a wrong-destination scenario without proving that a specific vessel is damaged. Keep the scenario specific enough to test and honest enough to revise when better information arrives.
How should credible scenarios be described?
A credible scenario should describe a realistic event sequence from initiating cause to possible harm and controls.
Scenario thinking prevents shallow risk assessments. Instead of writing 'fire risk', describe the actual chain: solvent spill in a bund, vapour spreads to a drain, extraction is off, a non-rated portable tool is used and ignition occurs. That scenario immediately suggests different controls from a generic fire row.
For major accident hazards, credible scenarios should consider inventory, energy, containment, ignition, toxic release, escalation, people affected and emergency response. For everyday occupational safety, they should still show the task, person, trigger, harm and control standard.
Block
Question to ask
Example
Hazard source
What can release, move, fall, ignite, contaminate or expose someone?
Petrol, silica dust, suspended load, energised equipment or fragile roof.
Initiating event
What starts the sequence?
Dropped container, blocked extraction, wrong valve opened or guard removed.
Existing SHAREhub photographic illustration. The written fictional case supplies the condition and decision evidence.
Use HIRA as a disciplined process
Hazard identification and risk assessment connects task scope, hazards, credible scenarios, existing controls and further action. Include relevant people and conditions, such as maintenance interfaces or abnormal work, rather than relying on a checklist detached from the process.
Record the assessment basis and who must act. Review when relevant conditions change or evidence challenges the assumptions. A signed form is not a substitute for implementation. The method must be suitable for the complexity and significance of the decision; a simple task sheet may not answer a major-hazard design question.
How should HIRA be completed step by step?
A good HIRA follows a clear sequence from scope through hazard identification, risk evaluation, control selection and review.
The step-by-step route below works for simple tasks and complex risk studies. The level of detail changes with risk. A low-risk office task may need a short assessment. A confined space, lifting operation, chemical process or construction excavation needs deeper planning, competent input and stronger evidence. HSE managing risks and risk assessment
The most common failure is starting with a template rather than the work. The team should observe the task, talk to workers, check history and understand the environment before scoring risk.
Step
What to do
Output
1. Define scope
State the task, process, area, lifecycle stage and people affected.
Recognise uncertainty and avoid numerical confidence without a basis.
Photorealistic illustrative UK workplace scene. Review checks: Avoid / redesign: Move pedestrian access and picking away from the loading route so the crossing is no longer needed. Control the task: Install suitable fixed separation and protected crossing points; design gates and routes to keep people out of the truck envelope. Verify before release: If the hazardous step remains, retain a task assessment and select the stronger feasible remaining controls.
Treat a rating as an aid
A risk rating helps organise discussion but depends on definitions, evidence and assumptions. Different scenarios can share a colour while needing very different controls. A low score does not make a required safeguard optional.
Record uncertainty and examine existing and further controls. Do not reduce a score simply because a procedure is mentioned or because no accident has occurred. If likelihood or consequence information is inadequate, identify the competent assessment needed. This course does not provide a numerical acceptance threshold or authorise risk acceptance.
How should risk rating be used sensibly?
Risk rating should support judgement and prioritisation, not hide uncertainty behind a number.
Example: a wrong-circuit maintenance task may be judged unlikely on a 5×5 matrix, but contact with live conductors could be fatal. A low arithmetic score cannot replace secure isolation, proving dead, competence and a hold point when the circuit identity is uncertain. HSE risk guidance calls for suitable, proportionate assessment and actual control; the matrix is a local aid, not a legal approval threshold. HSE managing risks and risk assessmentHSE electrical isolation case learningHSE managing risks and risk assessment
A five-by-five matrix can help rank actions, but it is not a substitute for competence. Two assessors may score the same risk differently if the scenario is vague, the controls are uncertain or exposure frequency is unknown. Good HIRA records the assumptions behind the score.
For high-consequence hazards, do not let a low likelihood score create complacency. Major accident hazards, fatal risks and irreversible health effects often require robust controls even when events are rare.
Distinguish broad identification from focused deviation questions.
Existing SHAREhub photographic illustration. The written fictional case supplies the condition and decision evidence.
Choose HAZID and what-if deliberately
A HAZID can help identify hazards and major scenarios across a defined activity or lifecycle stage. What-if questions explore how a proposed deviation or failure could matter. Both depend on suitable information, competent participation and a clear boundary.
Neither is an automatic substitute for every detailed technique. Record what the review can answer and where more focused analysis is needed. A broad workshop may reveal a concern about reverse flow without demonstrating every cause, safeguard performance or consequence. The useful result includes justified follow-up, not a claim of exhaustive coverage.
How are HAZID and what-if analysis different?
HAZID is usually a broader hazard review, while what-if analysis is a flexible challenge method that can be used inside or outside a HAZID.
In practice the two methods overlap. A HAZID workshop may use checklists and what-if questions to explore hazards. A separate what-if review may be used for a smaller change, temporary job or operating issue. The important point is not the label; it is whether the method is proportionate to the decision and whether it has enough technical challenge.
A weak review jumps straight to controls. A stronger review first describes the scenario and asks whether the current safeguards are independent, available and suitable for the consequence.
Include enough context to examine consequences and controls.
Photorealistic illustrative UK workplace scene. Review checks: Avoid / redesign: Remove the bulk solvent transfer where a closed supplied module or no-transfer design meets production needs. Control the task: Separate the transfer route from occupied areas and protect drains and ignition sources. Verify before release: Unproven throughput or displaced hazard holds redesign.
Ask a useful what-if question
“What if something fails?” is too vague to examine. Name the deviation, relevant activity and condition: what if flow continues after the intended receiving condition is reached? Then explore causes, consequences, existing safeguards and remaining questions.
Avoid phrasing the question so it assumes the answer or credits a safeguard without examination. Include operational knowledge and record assumptions. A workshop discussion is not approval for a trial. If the answer requires specialist calculations or unavailable process data, retain the action rather than inventing a reassuring conclusion.
What makes a HAZID scenario useful?
A useful scenario connects a cause, top event, consequence, safeguards and uncertainty in plain language.
The scenario should be specific enough for people to test it in the field. 'Fire risk' is weak. 'Solvent transfer hose leaks during night unloading; vapour enters a drain; portable lighting is not rated; ignition occurs near the loading bay' is useful because it tells the team what to check.
Good scenarios also identify escalation. A small leak may become a vapour cloud, fire, explosion, toxic exposure, drainage pollution or emergency response problem depending on location and controls.
SHARE example scope: HSE guidance supports a proportionate hazard study and the broad method; the specific rows below are editorial prompts to test against the actual process, evidence and competent study. HSE method context.
Flexible hose coupling leaks after wrong gasket is fitted. SHARE example—verify for the actual process.
Top event
Spill.
Loss of containment from transfer line during tanker offload. SHARE example—verify for the actual process.
Consequence
Injury or fire.
Flammable vapour cloud reaches ignition source and causes flash fire or explosion. SHARE example—verify for the actual process.
Safeguards
Procedure.
Pre-transfer checklist, earthing, exclusion zone, rated equipment, operator attendance and emergency shutdown. SHARE example—verify for the actual process.
Uncertainty
To be confirmed.
Confirm drain isolation standard and whether contractor lighting is suitable for hazardous area. SHARE example—verify for the actual process.
Action
Review controls.
Engineering to verify drain protection and operations to revise unloading checklist before next campaign. SHARE example—verify for the actual process.
Write a cause, credible consequence, affected people and actual safeguard; a label such as “operator error” is not a complete scenario. HSE COMAH process-safety assessment guidance
Applied case
Fictional exercise: A workshop asks what happens if a shared utility is lost during transfer. No dependency information is available.
Question
Specify the deviation and operating context.
Unknown
A missing dependency becomes an action, not an assumed safeguard.
Test your decision
What is useful?
Assume every protective function is independent.
Test the utility loss during live operation.
Record the scenario and obtain the relevant dependency information for review.
Compare the explanation
Record the scenario and obtain the relevant dependency information for review.
A question can identify missing evidence without creating the hazardous condition.
Existing SHAREhub photographic illustration. The written fictional case supplies the condition and decision evidence.
Understand design intent before deviation
A HAZOP examines deviations from intended operation using a structured team process. A node defines a manageable part of the process with a stated intent. Without that intent, a deviation such as “more flow” has no clear reference.
The boundary, current information and operating mode matter. Start-up or maintenance may have different intents and conditions from steady operation. Awareness learners can recognise the structure and supply task knowledge; they do not become competent study leaders by learning a guideword. A completed worksheet is useful only within its actual scope and quality.
How should HAZOP nodes and design intent be defined?
Nodes should be defined so the team can understand the intended operation and challenge deviations without confusion.
SHARE method illustration: The cited HSE material supports the broad method and its selection; the prompts and worked entries below are SHARE teaching examples, not a formal study or a claim about an unseen standard. HSE Corus investigation: HAZOP method
Poor node definition makes the study vague. A node that covers a whole plant may hide detail; a node that is too small may waste effort and lose system interactions. The facilitator should choose nodes that reflect process purpose, equipment boundaries and the decision being made.
Design intent should be practical, not decorative. It should say what the line, vessel, transfer, reaction or step is supposed to do and the conditions needed for it to remain safe.
Line from Tank T-101 outlet to Reactor R-201 charging nozzle. SHARE example—verify for the actual process.
Intent
Transfer solvent.
Transfer solvent at controlled flow to R-201 within temperature, pressure and nitrogen blanket limits. SHARE example—verify for the actual process.
Boundary
Whole charging process.
From tank isolation valve to reactor inlet valve, excluding reactor agitation sequence. SHARE example—verify for the actual process.
Information
Drawing reviewed.
Current P&ID, cause and effect, operating procedure, alarm list and relief basis available. SHARE example—verify for the actual process.
Scope note
Normal operation.
Normal charging, start-up, shutdown, low-flow, blocked-line and maintenance handback conditions considered. SHARE example—verify for the actual process.
A node needs a clear design intent and boundary so causes and consequences do not vanish between study sections. HSE HAZOP and plant-change guidance
Applied case
Fictional exercise: A worksheet says “more” but does not state the parameter or intended condition.
Node
A bounded part of the process under study.
Intent
The expected function and conditions provide the reference.
Test your decision
What is missing?
Proof that the plant is necessarily overpressurised.
The defined node, design intent and relevant parameter.
Only a larger spreadsheet.
Compare the explanation
The defined node, design intent and relevant parameter.
Separate causes, consequences, safeguards and actions.
Photorealistic illustrative UK workplace scene. Review checks: Avoid / redesign: Remove the bulk solvent transfer where a closed supplied module or no-transfer design meets production needs. Control the task: Separate the transfer route from occupied areas and protect drains and ignition sources. Verify before release: Unproven throughput or displaced hazard holds redesign.
Follow a HAZOP deviation through
A guideword and parameter help generate a deviation, such as no flow or reverse flow. The team then examines credible causes, consequences, existing safeguards and necessary actions. These entries should not be merged into one vague statement.
Challenge whether a safeguard actually addresses the cause or consequence being discussed and under which conditions it is available. Record unresolved technical questions for responsible follow-up. Completing every visible row does not prove the study is exhaustive, the actions are implemented or the plant is authorised to operate.
How do HAZOP guidewords help the team think?
Guidewords provide a disciplined way to challenge deviations such as no flow, more pressure, less cooling, reverse flow or other material.
SHARE method illustration: The cited HSE material supports the broad method and its selection; the prompts and worked entries below are SHARE teaching examples, not a formal study or a claim about an unseen standard. HSE Corus investigation: HAZOP method
Guidewords are not magic words. They are prompts that help a competent team challenge the process from different angles. The chair should use judgement: some guidewords will be highly relevant to a node, while others may be quickly dismissed with a reason.
The team should avoid jumping from guideword to solution. First understand the deviation, then credible causes, then consequences, then whether safeguards are adequate.
Guideword
Plain-English question
Example deviation
No or not
What if the intended action, flow, utility or protection is absent?
No cooling water to reactor jacket. SHARE example—verify for the actual process.
More
What if there is too much flow, pressure, temperature, level or reaction?
High pressure in transfer line. SHARE example—verify for the actual process.
Less
What if there is too little flow, cooling, agitation or concentration?
Low nitrogen purge flow. SHARE example—verify for the actual process.
Reverse
What if flow, sequence or movement reverses?
Reverse flow from reactor to storage tank. SHARE example—verify for the actual process.
As well as
What else could be present or occur at the same time?
Water contamination as well as solvent charge. SHARE example—verify for the actual process.
Other than
What if the wrong material, route, valve, source or destination is used?
Wrong chemical charged to batch reactor. SHARE example—verify for the actual process.
Early or late
What if the step happens too soon or too late?
Heating starts before purge is complete. SHARE example—verify for the actual process.
Apply guide words to parameters such as flow and pressure; record why a deviation is credible or screened out. HSE HAZOP and plant-change guidance
Applied case
Fictional exercise: A reverse-flow row lists “operator trained” as its only safeguard with no defined response.
Deviation
Guideword plus parameter starts the examination.
Action
An unresolved issue needs ownership and close-out evidence.
Test your decision
What needs examination?
The claimed response and whether it can control that specific pathway.
Nothing because the safeguard cell is filled.
A universal answer supplied by the guideword.
Compare the explanation
The claimed response and whether it can control that specific pathway.
HAZOP structure helps questions; it does not automatically validate safeguard claims.
Distinguish failure mode from effect and detection.
Existing SHAREhub photographic illustration. The written fictional case supplies the condition and decision evidence.
Trace a component failure mode
FMEA starts with defined items or functions and asks how each could fail, what effects follow and how the failure is detected or controlled. A valve failing open is a mode; an unwanted transfer may be an effect. Keep those relationships clear.
The analysis needs a suitable boundary and information about service and dependencies. It does not become a complete major-accident assessment merely because many components are listed. Common causes and combinations can need other analysis. Awareness supports recognition of the method and its limits, not an unsupported reliability calculation.
How does FMEA support process safety decisions?
FMEA supports process safety by identifying credible failure modes before they become hidden causes of major accident scenarios.
FMEA examines an item or system in a structured way: what is the function, how could it fail, what would cause that failure, what would the effect be, how would it be detected and what action is needed? It is useful for equipment packages, control loops, valves, pumps, utilities, loading systems, transfer systems and safety critical tasks.
FMEA is not normally the only major hazard study method. It is strongest when connected to HAZOP, maintenance strategy, spares, proof testing, pre-start-up safety review and management of change. A failure mode that appears minor at component level may become serious if it defeats containment, cooling, isolation, detection or emergency shutdown.
FMEA building blocks
FMEA element
Meaning
Example question
FMEA elementFunction
MeaningWhat the item is intended to do.
Example questionWhat must this valve, sensor, pump or procedure achieve?
FMEA elementFailure mode
MeaningHow the function could fail.
Example questionCould it fail open, fail closed, stick, leak, drift, give wrong signal or be unavailable?
FMEA elementCause
MeaningWhy the failure could occur.
Example questionCould corrosion, fouling, wrong calibration, human error or loss of utility cause it?
FMEA elementEffect
MeaningWhat the failure could lead to.
Example questionCould it cause overfill, overpressure, loss of cooling, release or ignition?
FMEA elementDetection
MeaningHow the failure would be found.
Example questionWould operators, alarms, proof tests or inspections detect it in time?
FMEA elementAction
MeaningWhat must change.
Example questionIs design, maintenance, testing, training or operating procedure improvement needed?
Sources and references
Specific official context: COMAH assessment criteria; functional safety guidance; control-systems guidance. Method tables, scenarios and invented numbers are teaching applications, not approved designs. Competent scenario data, layer independence/performance, full-function lifecycle and licensed standards approval remain required.
Applied case
Fictional exercise: A table calls “loss of containment” the failure mode of every component without explaining the path.
Mode
How does the item fail to perform its function?
Effect
What follows at the component or wider system level?
Test your decision
What would improve it?
Add more identical rows.
Assume component analysis covers every combination.
Identify the actual item failure mode and trace its effects in the process.
Compare the explanation
Identify the actual item failure mode and trace its effects in the process.
Recognise the evidence boundary for LOPA and integrity decisions.
Photorealistic illustrative UK workplace scene. Review checks: Avoid / redesign: Remove an unnecessary transfer or hazardous stock through a validated supply or process redesign. Control the task: Separate delivery operations from people and incompatible plant; control the credible spill and escalation area. Verify before release: If the hazardous step remains, retain a task assessment and select the stronger feasible remaining controls.
Do not count labels as independent layers
LOPA examines a defined initiating event and credited protection layers using a justified method and data. Independence, effectiveness and assurance matter. Two differently named functions may share a sensor, utility or final element, so counting names does not establish separate protection.
Safety-integrity decisions belong within a competent functional-safety lifecycle. A SIL is not a general quality badge for a device or a rating chosen by an awareness learner. The useful question is whether the claimed function, required performance and dependencies have a traceable technical basis. Do not invent failure frequencies or assign risk-reduction credit from this exercise.
How does LOPA test whether protection layers are enough?
A LOPA is a semi-quantitative test of one defined cause–consequence path. Credit only effective, independent, auditable layers whose performance and common-cause dependencies are evidenced; a number cannot repair a weak scenario.
A LOPA should start with a clear scenario, usually developed from HAZOP or another hazard study. The scenario should describe the initiating event, enabling conditions, consequence and existing safeguards. The team then decides which safeguards can legitimately be credited as independent protection layers.
The important word is independent. A high-level alarm and a high-high trip may not be independent if they share the same level transmitter, power supply, logic solver or maintenance practice. An operator response is not independent if the same operator action creates the initiating event or if there is not enough time to diagnose and respond.
LOPA is semi-quantitative. It helps structure judgement, but it does not remove uncertainty. The study should record the basis of the numbers and the reasons for excluding weak or dependent safeguards.
LOPA logic
Step
What is decided
Quality check
StepDefine scenario
What is decidedOne cause, one top event and one consequence path.
Quality checkIs the scenario specific enough to test?
StepEstimate initiating event
What is decidedFrequency of the event starting the sequence.
Quality checkIs the data relevant and conservative enough?
StepIdentify IPLs
What is decidedSafeguards that prevent or mitigate the consequence.
Quality checkAre they effective, independent and auditable?
StepApply PFD values
What is decidedProbability that each layer fails on demand.
Quality checkAre proof tests and human response assumptions justified?
StepCompare with target
What is decidedWhether mitigated frequency meets the risk criterion.
Quality checkIs the target suitable for the consequence?
StepClose the gap
What is decidedAdditional design, SIS, procedure or management controls.
Quality checkAre actions owned, funded and verified?
HSE control-systems guidance warns that shared instruments, utilities and management practices can defeat claimed independence.
Applied case
Fictional exercise: An alarm and trip are claimed as independent, but their shared sensor dependency is unexamined.
Layer
Its function and independence need evidence.
SIL
A specialist requirement for a safety function, not a product slogan.
Test your decision
What should be challenged?
A learner-selected SIL for the sensor.
The independence and justified credit of the claimed layers.
Only whether they have different names.
Compare the explanation
The independence and justified credit of the claimed layers.
Layer and integrity claims require competent scenario-specific justification.
A fault tree asks what combinations could produce an unwanted event. An event tree asks what may follow an initiating event as protective functions succeed or fail.
Choose the direction of the question
An event tree starts from an initiating event and follows possible outcomes through success or failure of relevant functions. A fault tree starts with a defined unwanted event and examines combinations that could produce it. Their direction helps answer different questions.
Define scope and logic before adding numerical values. The same label can hide a different function or condition, so use clear event descriptions. Neither diagram is proof by appearance. Review assumptions, completeness and dependencies with suitable competence before using results to support a risk decision.
Choose the direction of the analysis
A fault tree asks what combinations could produce an unwanted event. An event tree asks what may follow an initiating event as protective functions succeed or fail.
Fictional learning example
Separate the evidence from the decision
Part of the picture
What the example establishes
Question A
What could allow a receiving tank to overfill?
Question B
If level rises unexpectedly, which protective responses work and what outcomes remain?
Evidence boundary
The branches do not justify numerical probabilities or independent-layer credit on their own.
Pause and decide: Would drawing more branches prove that the risk is adequately controlled?
Compare the reasoning
No. Define the scenario and the purpose of the method, then check the logic, dependencies and evidence behind each claimed function. Use competent assessment for design or quantitative claims. A diagram can expose assumptions; it cannot make them true.
When should event trees, fault trees or bowties be used?
Use the method that best answers the decision question rather than forcing every issue into one diagram.
A fault tree is strongest when the question is 'what could cause this top event?' An event tree is strongest when the question is 'what could happen after this initiating event?' A bowtie is useful for communicating threats, top event, preventive barriers, consequences and mitigative barriers in one visual model.
Many mature process safety programmes use these methods together. A fault tree may identify causes of loss of containment; an event tree may explore escalation after release; a bowtie may summarise barriers for operations and assurance.
Understand why combinations need more than multiplication.
Photorealistic illustrative UK workplace scene. Review checks: Avoid / redesign: Use a fault tree to ask how a defined top event could occur, then an event tree to ask how an initiating event could develop as barriers succeed or fail. Control the task: For a tank overfill, a fault tree tests how level control, instrument and operator failures combine to allow the release. Verify before release: Use a fault tree to ask how a defined top event could occur, then an event tree to ask how an initiating event could develop as barriers succeed or fail.
Challenge dependency assumptions
Fault-tree logic distinguishes alternative contributing paths from combinations required for an event. The logic must match the process. Adding numbers does not repair an incorrect relationship or missing cause.
Shared utilities, environment, maintenance or information can connect apparently separate functions. If independence is assumed without justification, a numerical result may look precise while understating the scenario. Awareness learners should identify that assumption and obtain competent review, not calculate a credible frequency from invented data. Qualitative reasoning can expose a question even when no numerical conclusion is justified.
Why is common-cause failure so important?
Common-cause failure is important because one weakness can defeat several barriers that look independent on paper.
Major accidents often involve barriers that appear separate but are not truly independent. They may share the same power supply, instrument tapping, maintenance team, calibration procedure, human decision, software logic, emergency access route, drainage system or organisational priority.
This is why a barrier register should not only list controls. It should test dependence. If two controls rely on the same person noticing the same alarm during a high workload period, they may not be two independent barriers.
Separate threats from consequences around loss of control.
Existing SHAREhub photographic illustration. The written fictional case supplies the condition and decision evidence.
Locate the top event
A bowtie places a defined top event between threats that could lead to it and consequences that could follow. The hazard supplies the potential for harm; the top event describes loss of control. Clear wording prevents cause, event and harm from collapsing into one label.
Locate preventive and mitigative functions on the appropriate paths. A diagram helps communicate the scenario, but it does not establish that the barriers are effective, independent or present in the field. Those claims need their own performance and assurance basis.
What are the main parts of a bowtie?
A bowtie contains a hazard, top event, threats, consequences, preventive barriers, mitigative barriers and degradation controls.
A good bowtie is specific enough to support decisions. If the top event is too broad, such as 'fire', the diagram becomes a collection of generic controls. If it is too narrow, it may miss important escalation paths.
The language used in the bowtie should be plain enough for operators and specific enough for engineers. A barrier such as 'training' is usually too vague; a stronger barrier would define the safety critical task, competence requirement, assessment method and refresher interval.
The diagram should be linked to source evidence: hazard and operability study (HAZOP) actions, layer of protection analysis (LOPA) assumptions, maintenance routines, procedures, emergency plans, alarm response, inspection records, management of change (MOC) and incident learning.
Threat. Left of the top event. Threats are written as vague hazards rather than credible causes.
Preventive barrier. Between threat and top event. The barrier is not independent or is not maintained.
Top event. Centre of the bowtie. The top event is too broad to test.
Mitigative barrier. Between top event and consequence. Emergency response is assumed without testing.
Consequence. Right of the top event. Consequences ignore escalation or off-site harm.
Degradation control. Linked to a barrier. The bowtie does not show what weakens controls.
Applied case
Fictional exercise: A bowtie uses “flammable liquid” as both hazard, top event and consequence.
Threat
A cause path towards loss of control.
Consequence
A harm outcome after that loss.
Test your decision
What would improve it?
Treat the material name as a complete scenario.
Define a specific loss-of-control event and distinguish its threats and consequences.
Add more barrier symbols without definitions.
Compare the explanation
Define a specific loss-of-control event and distinguish its threats and consequences.
The parts must perform distinct explanatory roles.
Photorealistic illustrative UK workplace scene. Review checks: Avoid / redesign: Eliminate unnecessary transfer or inventory Control the task: Provide independent level protection, shutdown and containment Verify before release: Barrier owner checks independence and tests; stop transfer on impaired high-high protection See HSE tank overfill safety alert
Manage barriers beyond the diagram
A barrier can become ineffective through a degraded condition or changed dependency. Identify what can defeat its function and what maintains or verifies it. Ownership should connect performance requirements, inspections or tests and the response to an unavailable condition.
Do not use a green symbol as a substitute for current evidence. A degradation-control activity also needs a justified relationship to the barrier; another checklist is not automatically protection. The awareness task is to ask whether the claimed function is supported and what happens when its condition is unconfirmed.
What are degradation factors in bowtie analysis?
Degradation factors are conditions that can weaken or defeat a barrier before it is needed.
A barrier may exist on the drawing but be weak in reality. A detector may be poorly positioned. A relief valve may be isolated. A procedure may be too complex. A person may not be competent. A fire pump may not start. These are degradation factors.
Good bowties identify degradation factors and the controls that manage them. For example, the barrier may be 'gas detection'; degradation factors may be sensor poisoning, poor calibration, wrong location and alarm flooding. Degradation controls include calibration, detector mapping, maintenance and alarm response training.
This is where many bowties become valuable for audits. Instead of auditing a whole system vaguely, the auditor can test specific degradation controls for specific barriers.
Gas detector. Poor positioning or overdue calibration. Detector mapping and calibration schedule.
Relief valve. Isolated, fouled or not tested. Locked-open control and proof testing.
Procedure. Too long, unclear or not field-verified. Procedure walk-through and operator feedback.
Emergency isolation. Valve inaccessible or actuator failed. Remote test and access review.
Training. Attendance recorded but competence not assessed. Practical assessment and refresher.
Firewater. Insufficient flow or impaired pump. Flow test and impairment management.
Applied case
Fictional exercise: A bowtie barrier stays green although its required verification result is unresolved.
Degradation
What can prevent the function working as required?
Assurance
What evidence shows its relevant condition now?
Test your decision
What should the review do?
Examine current evidence and apply the defined response to the unresolved condition.
Leave the colour because the diagram is approved.
Count the colour as another barrier.
Compare the explanation
Examine current evidence and apply the defined response to the unresolved condition.
Barrier status must follow the relevant condition and assurance basis.
Distinguish justified risk reduction from a preferred budget.
Existing SHAREhub photographic illustration. The written fictional case supplies the condition and decision evidence.
Understand ALARP without a shortcut
ALARP concerns reducing risk as low as reasonably practicable within the applicable legal and assessment context. It is not a rule that a favourable matrix colour ends the search for improvements. Relevant good practice, further options and the scenario’s significance need a reasoned review.
Do not treat ordinary cost-benefit equality or a manager’s preferred budget as the whole test. Gross-disproportion reasoning is weighted towards risk reduction and needs competent justification. This course supplies no universal numerical boundary, monetary factor or acceptance authority. Establish applicable GB or NI requirements rather than assuming one example determines them.
What do SFAIRP and ALARP actually require?
The exact legal wording comes first: where a duty is qualified by "so far as is reasonably practicable", risk reduction is expected unless a further measure would involve grossly disproportionate sacrifice.
ALARP is the practical expression of that level of control, not a separate universal legal test. Risk and the money, time and trouble of a further measure are weighed with the balance favouring safety. Do not treat "not zero risk" as permission to leave an obvious practicable control undone. First meet any specific mandatory standard, then apply relevant good practice and test whether further measures are reasonably practicable.
Legal duty and decision vocabulary
Term
Use
What it cannot do
SFAIRP or SFARP
Statutory phrase in many GB and NI general duties.
Replace a different or absolute legal requirement.
ALARP
Describes reducing risk to the reasonably practicable level.
Act as a self-issued certificate or simple score.
Good practice
Applicable regulator guidance, ACOPs, standards and established methods.
Excuse ignoring site-specific or new hazards.
Tolerability
Framework for judging whether risk can be borne under continued control.
Mean harmless, universally acceptable or permanently approved.
Gross disproportion
A strong reason needed to reject further effective control.
Become a fixed numerical cost ratio for all cases.
Applied case
Fictional exercise: A proposal is rejected only because it exceeds the department’s preferred budget.
Question
What further risk reduction is available and why is it not adopted?
Boundary
The course does not perform a real legal or engineering acceptance decision.
Test your decision
What is missing?
Only a cheaper colour on the risk matrix.
A universal cost factor chosen by the learner.
A justified review of risk reduction, relevant good practice and the basis for rejecting the option.
Compare the explanation
A justified review of risk reduction, relevant good practice and the basis for rejecting the option.
Budget preference alone is not an ALARP demonstration.
Photorealistic illustrative UK workplace scene. Review checks: Avoid / redesign: Remove the hazard or activity where reasonably practicable. Control the task: Separate people, ignition sources or vulnerable receptors from the hazard. Verify before release: Remove a hazardous intermediate, avoid confined entry, design out manual transfer or avoid hot work.
Integrate an assessment decision
Methods support different questions; none removes the need for sound information, competent judgement and implemented controls. An assessment should explain its scenario, selected method, assumptions, credited functions, further options and outstanding actions. A method name alone does not demonstrate adequacy.
For this fictional exercise, keep useful findings while identifying limits. If a shared dependency undermines a layer claim, route that question to competent review rather than manufacturing a numerical answer. Preserve relevant restrictions and action ownership. The outcome is a reasoned learning recommendation, not real risk acceptance or permission to operate.
How does barrier thinking strengthen ALARP?
Barrier thinking strengthens ALARP by asking what physically or organisationally prevents, detects, controls or mitigates the event.
Independence, condition, testing and response time must be evidenced before a barrier is credited in a decision.
A barrier is more than a control name. It should have a function, performance standard, owner, inspection or proof-test requirement, and a response when it is unavailable or degraded. This is where many ALARP arguments fail: the risk assessment assumes a barrier is present, but the field evidence shows it is bypassed, overdue, poorly understood or not independent.
Preventive barriers reduce the likelihood of the top event. Detection barriers identify that control is being lost. Control barriers stop or limit the release or escalation. Mitigation barriers reduce consequences after the event has started. Emergency response barriers protect people and prevent escalation once the incident has moved beyond normal control. Assurance evidence then proves the barriers are still present, available and effective.
Barrier-led ALARP is useful because it makes the decision testable. Instead of saying “risk is controlled”, the organisation can ask: what prevents the event, what detects it, what controls it, what mitigates it, what emergency response is needed, who owns each barrier, what standard applies and what evidence proves it is healthy today?