Full public teaching and readers are free without an account. A free learner account is required for saved LMS progress and recorded assessment. Completion and any course-completion certificate depend on the configured LMS conditions. Full course outline and access routes.
Lesson 1
Recognise the operating boundary before the leak grows
Distinguish normal operation, abnormal response and maintenance release.
Recognise the operating boundary before the leak grows
A pump can appear productive while moving towards loss of containment. In this fictional terminal, a rising vibration trend, unusual seal temperature and an unexplained flow change deserve a connected explanation. The learner’s first task is to compare the evidence with the approved operating envelope and abnormal-operation instructions. A loading target cannot define acceptable equipment condition. Neither the model animation nor a colleague’s confidence measures the real pump.
Consider the decision in three parts: what is known, what might follow, and what authority is available now. Fresh liquid near a seal could expose a cleaner, feed a pool or vapour release, and reach drains. Avoid approaching to identify a substance by smell or touching the equipment. Report the observation and use the agreed alarm, exclusion and safe-state arrangements through trained personnel from a safe location. Do not improvise valve movements from this lesson.
Good procedures connect recognisable conditions to clear actions, boundaries and escalation. A checklist that names an obsolete valve cannot safely support the next step. Stop at the uncertain instruction, preserve the required safe condition, and ask the responsible operating team to reconcile the document with actual equipment. Correcting an identifier through controlled change also prevents the next shift repeating the uncertainty.
For the learning case, write a short operating log: observed trend, time, affected route, immediate safe action, person informed and unresolved question. Separate observation from diagnosis. “Liquid observed near seal” is evidence; “only condensation” is an unproved conclusion. This discipline makes later maintenance planning more useful and helps the receiving shift understand why the pump remains unavailable.
What should happen when the envelope is challenged?
Excursions should trigger stabilisation, escalation, recording and learning, not informal tolerance.
An excursion is useful safety information. Even if no harm occurs, it may show that the process, procedure, alarm, instrument, training or staffing assumption is weak. See HSE operating procedures for the relevant source and scope.
Repeated small excursions are especially important. They often reveal normalisation before a serious loss of control.
Excursion stage
Expected response
Evidence
Approaching limit
Recognise trend and take planned corrective action.
Hold or safely stop an operation with an impaired credited safety function unless the designated competent authority demonstrates effective protection for that specific operating state. Record the permitted limits, concurrent impairments, owner, expiry and review triggers. A signature and time limit alone do not make a bypass safe.HSE control systems
Applied case
An operator sees fresh liquid beneath P-204 and a worsening vibration trend. The dispatch board requests another loading run; an old checklist refers to a valve that no longer exists.
Decision and explanation
Apply the abnormal-operation response and resolve the procedure discrepancy through the authorised team.
The trend, liquid and mismatched procedure make continued operation uncertain. Reach the agreed safe state and obtain competent diagnosis; production demand does not establish readiness.
Look closer
Explore the numbered equipment and control points. Existing SHAREhub teaching schematic; illustrative and not to scale.Selected route
Trace the selected product to its receiving tank. Reconcile current drawings and field identity; colour or memory is insufficient.
Proved work boundary
Routing and intrusive work require different evidence. Isolation must address connected inventory, energy, leakage and possible reaccumulation.
Drain and handback
Plan containment, draining, verification and coordinated reinstatement. This conceptual drawing does not prescribe an isolation method.
01 · Read the situation
An operator sees fresh liquid beneath P-204 and a worsening vibration trend. The dispatch board requests another loading run; an old checklist refers to a valve that no longer exists.
02 · Explain the principle
A pump can appear productive while moving towards loss of containment. In this fictional terminal, a rising vibration trend, unusual seal temperature and an unexplained flow change deserve a connected explanation. The learner’s first task is to compare the evidence with the approved operating envelope and abnormal-operation instructions. A loading target cannot define acceptable equipment condition. Neither the model animation nor a colleague’s confidence measures the real pump.
03 · Carry it into work
The trend, liquid and mismatched procedure make continued operation uncertain. Reach the agreed safe state and obtain competent diagnosis; production demand does not establish readiness.
Lesson 2
Isolation controls sources, not just the start button
Explain the evidence boundary for pump maintenance and a line break.
Isolation controls sources, not just the start button
Stopping a motor changes its operating state; it does not by itself remove electrical, pressure, chemical, gravity or mechanical energy. The fictional P-204 can receive product from its suction line, retain pressure in the discharge and trap liquid between closed boundaries. A remote command might also restart equipment. Before planning work, identify what the task exposes and every credible source that can reach that exposure.
An isolation plan translates this understanding into a specific, authorised boundary. Competent people select suitable isolation arrangements, secure them against unintended change, safely address trapped material and prove the required condition by approved methods. The exact arrangement depends on the substance, inventory, pressure, work and site standard. A closed control valve and a gauge displaying zero are insufficient evidence for the proposed manifold flange opening in this course.
Trace the boundary on current information and reconcile it with the field before release. Ask who owns upstream connections, how shared branches are controlled, how a testing point is known to function, and what could reintroduce energy. Draining or venting must lead to an assessed safe system; discharging fuel or vapour into the work area creates another hazard. This is a planning exercise, not an instruction to loosen a connection to test it.
Record the plant identity, isolation references, safe-condition verification, residual hazards and named responsibilities so the performing team can understand the arrangement. If the work scope changes, suspend and reassess rather than extending an existing signature. Controlled reinstatement matters equally: account for people, tools, temporary equipment and isolations, then follow the authorised sequence. An apparently successful repair does not cancel other teams’ protections.
How should isolation be verified?
Verification proves the isolation is effective at the point of work, not just applied somewhere upstream.
Verification is the difference between a hopeful isolation and a controlled job. Electrical work needs proving dead; process work may need depressurisation, drain/vent checks, gas testing, sampling, pressure indication, bleed verification, flushing, purging or visual confirmation of blinds. See HSG253 - The safe isolation of plant and equipment.
Verification should be independent enough for the risk. A person should not simply trust the same label or drawing that could be wrong. Where consequences are high, independent verification and authoriser challenge are essential.
Hazard
Verification method
Common failure
Electrical energy
Test for dead with approved voltage indicator.
Tester not proved before/after or wrong circuit tested.
Pressure
Check gauges, vents, drains and bleed points.
Blocked gauge line gives false reading.
Flammable atmosphere
Gas test after purge and before line break.
Single test misses later vapour return.
Toxic/corrosive contents
Sample, drain, flush, isolate and PPE.
Dead leg or trapped pocket remains.
Mechanical movement
Try-start where safe, block and physical check.
Stored gravity/hydraulic energy not controlled.
Blind/spade
Visual check and blind list verification.
Wrong size/rating or wrong location.
Learning check: two apparent protections
A locked but wrongly identified isolator does not protect the work point. A correctly identified boundary still requires proof of every relevant energy/feed and each specified DBB barrier, with accepted evidence and a safe discharge destination. If identity or proof is missing, hold the intrusive work and use the roles/competence route. Do not turn this check into a generic valve-operating sequence or invent acceptance pressures.
Applied case
A fitter proposes opening a manifold flange after closing the control valve. The local gauge reads zero, but its function and a connected branch have not been checked.
Decision and explanation
There is no suitable authorised, secured and proved isolation for all relevant sources.
A zero indication and closed control valve do not prove an effective isolation boundary. The competent team must establish and verify the task-specific safe condition before a line break.
Look closer
Photorealistic illustrative UK workplace scene. Review checks: Source / exposure: Common failures are usually predictable and should be designed out or caught by assurance. Possible consequence: Wrong line identity, passing valves, back-feed, blocked drains and incomplete handback can each defeat a signed plan. Control failure: Treat these as testable failure modes with named checks, not as reminders to be careful.01 · Read the situation
A fitter proposes opening a manifold flange after closing the control valve. The local gauge reads zero, but its function and a connected branch have not been checked.
02 · Explain the principle
Stopping a motor changes its operating state; it does not by itself remove electrical, pressure, chemical, gravity or mechanical energy. The fictional P-204 can receive product from its suction line, retain pressure in the discharge and trap liquid between closed boundaries. A remote command might also restart equipment. Before planning work, identify what the task exposes and every credible source that can reach that exposure.
03 · Carry it into work
A zero indication and closed control valve do not prove an effective isolation boundary. The competent team must establish and verify the task-specific safe condition before a line break.
Lesson 3
Make permits describe the real shared worksite
Coordinate hot work and transfers without treating documents as physical protection.
Make permits describe the real shared worksite
A permit-to-work system communicates a controlled task and its conditions; a signature does not physically isolate product or remove ignition sources. In the fictional manifold case, one team holds a hot-work permit while another schedules a neighbouring transfer. Each team can describe its own job correctly and still miss the risk created where the jobs meet. The controlling team must understand both activities together.
Before release, reconcile the task, equipment identity, boundaries, duration, precautions, competent roles and stop triggers at the actual worksite. The performing authority should be able to explain what would make the permit invalid. A changed wind, unexpected product, loss of gas-monitoring arrangements or a new transfer request may change the assessed conditions. The correct response is the specified suspension and communication process, not a handwritten exception made to keep the job moving.
For simultaneous operations, draw a simple interface map with people, routes, inventories, ignition sources and emergency access. Ask how one team’s isolation, test or discharge could affect another team’s assumptions. Resolve conflicting work through the authorised controllers before either affected task proceeds. Merely cross-referencing two permit numbers leaves the physical conflict unresolved. Brief both teams on the agreed boundary and how each can stop the work.
Shift changes and breaks need explicit status: what is suspended, which protections remain, who controls the area and what must be rechecked. When the task is complete, inspect and hand back the plant through the defined process, including post-work precautions where required. In the exercise, your useful output is a specific unresolved-interface statement and a named action owner. “Permits checked” says little unless the connected conditions were actually reconciled.
What does a good permit lifecycle look like?
A good permit follows the job from planning through risk control, issue, live monitoring, suspension, handback and audit.
The lifecycle is the backbone of permit quality: plan -> risk assess -> isolate -> test -> issue -> brief -> perform -> monitor -> suspend if needed -> hand back -> close out -> audit. Each stage has a different purpose and a different owner. Skipping one stage often creates the conditions for a serious incident.
For process safety, the lifecycle also protects major accident barriers. A line break, hot work task or confined-space entry may defeat containment, introduce ignition, remove safeguards or put people into an atmosphere that can change. The permit system must connect to isolation standards, gas testing, SIMOPS, emergency response and management of change.
Stage
Purpose
Evidence
Plan
Decide whether a permit is required and define work scope.
Work pack, risk assessment and permit type.
Risk assess
Identify hazards, affected people, adjacent work and emergency needs.
Task risk assessment and field check.
Isolate and make safe
Remove, lock, tag, drain, vent or control energy and substances.
Isolation certificate and verification.
Test and verify
Confirm atmosphere, dead state, zero energy or safe access where needed.
Gas test, prove-dead record or pressure/containment check.
Issue and brief
Authorise only when conditions and roles are understood.
Signed permit and toolbox talk.
Monitor
Check that work stays within permit conditions.
Area checks, fire watch, gas retest or supervisor log.
Suspend
Pause work when conditions, scope, weather, SIMOPS or shift change affect safety.
Suspension note and revalidation.
Hand back and close
Confirm work complete, area safe and safeguards reinstated.
Handback signature and restart checks.
Audit
Check permit quality and field reality.
Permit audit findings and corrective actions.
Applied case
Welding is scheduled beside a manifold at 10:00. A product-transfer team expects to use a connected branch at 10:15, and neither work pack mentions the other.
Decision and explanation
Hold the conflicting activities and resolve their combined conditions with both authorised teams.
The common physical boundary creates the risk. Resolve the transfer, isolation, atmosphere, ignition and emergency-access interfaces before releasing the affected work.
Look closer
Photorealistic illustrative UK workplace scene. Review checks: Eliminate: Eliminate or redesign: the maintenance planner checks whether the repair can be completed without opening the line, or whether the pump can be taken fully out of service and worked on in a controlled workshop. Reduce: Substitute or reduce: engineering considers a less hazardous cleaning medium or reduced inventory and pressure before opening, only if process compatibility and the approved procedure allow it. Isolate: Isolate and engineer: the isolation authority selects a risk-based physical isolation scheme under HSE HSG253, identifies the exact line, secures energy sources, drains and vents to a safe system, and proves the boundary before work. Control: Control, PPE and Discipline: the area authority coordinates simultaneous operations, gas-test timing, permitted scope, time limit, fire/containment arrangements and shift handover under HSG250. Control: Read the full task example, checks and stop points on this page.01 · Read the situation
Welding is scheduled beside a manifold at 10:00. A product-transfer team expects to use a connected branch at 10:15, and neither work pack mentions the other.
02 · Explain the principle
A permit-to-work system communicates a controlled task and its conditions; a signature does not physically isolate product or remove ignition sources. In the fictional manifold case, one team holds a hot-work permit while another schedules a neighbouring transfer. Each team can describe its own job correctly and still miss the risk created where the jobs meet. The controlling team must understand both activities together.
03 · Carry it into work
The common physical boundary creates the risk. Resolve the transfer, isolation, atmosphere, ignition and emergency-access interfaces before releasing the affected work.
Lesson 4
Avoid tank entry, then plan a workable rescue if entry is unavoidable
Identify the distinction between a cleaned tank and an authorised safe entry.
Avoid tank entry, then plan a workable rescue if entry is unavoidable
An empty-looking tank can contain hazardous vapour, oxygen deficiency, residues or connected sources of liquid and gas. Confined-space assessment depends on enclosure and reasonably foreseeable specified risks, not whether a person can stand upright. In this fictional case, a request to inspect internal fittings begins with a question: can the required information be obtained safely from outside? Avoiding entry removes exposure that monitoring alone cannot eliminate.
Where entry cannot reasonably be avoided, competent people must establish a suitable safe system for the actual task. This includes the isolation boundary, preparation, atmosphere assessment and controls, communication, supervision and emergency arrangements. Cleaning does not prove that the atmosphere will stay safe. Work, residues, ventilation changes and connected services can alter conditions. A respiratory mask selected for dust does not supply breathable air in an oxygen-deficient environment.
Rescue must be planned before entry and workable with the available people and equipment. “Call the emergency services” is not, by itself, an effective site rescue arrangement. Consider access geometry, retrieval limitations, responder protection, communication and how a casualty can be reached without creating more casualties. Only competent, equipped people act within the established rescue plan; an unplanned attempt to follow a collapsed colleague into a tank can kill the rescuer.
The learning decision finishes at readiness, not at an entry instruction. In the exercise, ask the responsible team to demonstrate the proposed arrangements safely, confirm current evidence and identify the conditions that stop entry. Record who can authorise the task and who controls restoration. The later handback must account for entrants, tools, temporary services and barriers so no restart can begin while people remain exposed.
How should rescue capability be governed?
Rescue capability should be treated as a controlled emergency function with defined people, equipment, maintenance, drills and review.
Rescue arrangements need the same discipline as entry controls. The organisation should know which types of entry require non-entry retrieval, internal rescue, breathing apparatus, specialist contractors, external emergency-service liaison or a combination. The plan should match the geometry of the space and the likely condition of the casualty.
Drills should test credible scenarios, not comfortable ones. A good drill asks whether communications work, whether equipment reaches the casualty, whether the casualty can be extracted through the access point, whether rescuers are protected and whether the response time is realistic.
Rescue capability governance
Capability
Management requirement
Evidence
Rescue plan library
Plans matched to space classes and tasks.
Task-specific rescue sheets.
People
Competent rescue personnel available during entry.
Training, authorisation and rota.
Equipment
Suitable and maintained retrieval, BA/RPE, communication and first-response kit.
Inspection and maintenance records.
Drills
Scenario-based practice and debrief.
Drill report and action close-out.
External liaison
Emergency services know site access and hazards where relevant.
Emergency plan and liaison notes.
Review
Weaknesses escalated and resourced.
Management review and improvement plan.
Applied case
A cleaned tank needs an internal inspection. The pack says “gas test before entry; rescue by emergency services” but does not establish isolation, changing atmosphere conditions or a workable retrieval method.
Decision and explanation
First examine a safe external inspection method; if entry is unavoidable, hold it until the complete safe system and effective rescue are ready.
Avoid entry wherever reasonably practicable. If entry is necessary, current isolation, atmosphere, work-control and effective emergency arrangements must be established before exposure.
Look closer
Related SHAREhub topic guide. Apply its prompts to the specific case below.01 · Read the situation
A cleaned tank needs an internal inspection. The pack says “gas test before entry; rescue by emergency services” but does not establish isolation, changing atmosphere conditions or a workable retrieval method.
02 · Explain the principle
An empty-looking tank can contain hazardous vapour, oxygen deficiency, residues or connected sources of liquid and gas. Confined-space assessment depends on enclosure and reasonably foreseeable specified risks, not whether a person can stand upright. In this fictional case, a request to inspect internal fittings begins with a question: can the required information be obtained safely from outside? Avoiding entry removes exposure that monitoring alone cannot eliminate.
03 · Carry it into work
Avoid entry wherever reasonably practicable. If entry is necessary, current isolation, atmosphere, work-control and effective emergency arrangements must be established before exposure.
Lesson 5
Pump seals: contain the release before treating the floor
A clean floor does not demonstrate intact primary containment. Investigate the source before calling the defect closed.
Understand this fuel-industry task
The pump adds energy to move fuel from storage towards the loading lanes. Its casing, seals and connections maintain primary containment while the shaft rotates. Fresh liquid and odour beside a seal can therefore be evidence of an active process release, not simply a housekeeping defect. Depending on the product and conditions, a leak can produce a spray, spreading pool or vapour, with an ignition point some distance from the visible wet patch. Approaching to wipe the seal can put someone directly into the release path.
The task here is to recognise deteriorating containment and use the planned safe response. Keep people clear, report the condition and activate the required alarm or shutdown arrangements from a safe location through trained personnel. Absorbents may help a controlled clean-up later, but cannot repair a live seal or make the atmosphere safe. Repeated small leaks should trigger cause investigation: equipment selection, seal support systems, alignment, operating conditions or maintenance quality may be contributing. Restart needs a controlled repair and evidence that the cause and affected safeguards have been addressed, not just a newly cleaned floor.
Trace the initiating event and loss of control
Step
Case pathway
initiatingEvent
A seal or supporting operating condition deteriorates during fuel transfer.
lossOfControl
Fuel escapes primary containment while the pump or connected inventory can continue feeding the release.
consequences
Exposure, slipping, a pool or vapour-cloud fire and escalation to adjacent pumps or tankers become credible.
Choose effective safeguards
Control priority
Practical application
Eliminate
Remove unused connections and avoid opening live process systems.
Substitute
Choose a less intrusive maintenance or sampling method where suitable.
Engineer
Provide an assessed isolation boundary, contained drains and means to prove its condition.
Manage
Walk down equipment identity, record isolation, coordinate users and control reinstatement.
Protect
Use PPE selected for the substances and residual exposure; it cannot make a live line safe.
Keep people and ignition sources out of the suspected release area; use the site’s alarm, safe shutdown and emergency-isolation response rather than close investigation by an unprotected worker.
Prevent recurrence through suitable seal and material selection, approved operating conditions, condition monitoring and planned maintenance with clear defect limits.
Use secured and proved mechanical and electrical isolation for repair, addressing connected inventories and trapped pressure before opening the equipment.
Control clean-up, waste, drains and return to service under the incident and maintenance process; record repeated leaks as barrier deterioration requiring investigation.
Verify effectiveness before release
Compare leak history, operating trends and maintenance findings to establish the mechanism, rather than accepting “seal changed” as the complete cause analysis.
Check the repair scope, isolation handback and appropriate integrity or functional test evidence for this pump and its affected connections.
Confirm the authorised monitored restart meets the approved conditions, no ongoing release is evident through the safe inspection method and unresolved actions have explicit ownership.
Transfer the decision to the workplace
At a loading arm or tank flange, apply the same source-first reasoning: control the release and inventory before asking someone to manage the visible residue.
What should happen after a leak, rupture, failed safeguard or imminent-danger finding?
Protect people first, place the plant in the defined safe state from a safe location, raise the alarm, isolate the area and let the emergency plan and competent technical authority control recovery.
Integrity failure emergency actions
Stage
Action
Do not
Record or handover
Immediate
Stop the task, warn others, withdraw by the safe route and raise the site alarm.
Approach a leak, vapour cloud, unstable equipment or fire to investigate informally.
Time, location, asset tag, observed conditions and people affected.
Control
Use remote shutdown or isolation only if trained, authorised and safe; account for people.
Create ignition, operate unassessed valves or rely on a failed safeguard.
Alarm, trip, ESD and isolation status.
Response
Give responders substance, pressure, inventory, drainage and barrier information.
Assume the contents, wind direction or containment route.
Emergency command log and environmental actions.
Quarantine
Preserve failed components, data and settings; identify similar equipment that may share the cause.
Repair, reset or discard evidence before authorised examination.
Quarantine tag, photographs, samples and control-system history.
Recovery
Use competent investigation, MOC, repair, test, hand-back and restart approval.
Restart because the visible leak has stopped or production pressure is high.
Technical acceptance, actions, learning brief and restart decision.
An operator finds fresh fuel near a running pump. The cleaner has been told to wipe it up before the next tanker reaches the bay.
Decision and explanation
Keep people clear and use the site’s release response, with safe shutdown/isolation by authorised personnel.
Absorbent can become additional contaminated material while the live release continues. The containment failure and ignition risk remain unresolved. This stops treating the visible pool as the whole problem and addresses the source and connected inventory before clean-up. Close inspection of live leaking equipment can expose the person to spray, vapour and moving parts. Establish a safe condition before intrusive diagnosis.
Look closer
Photorealistic illustrative UK workplace scene. Review checks: General corrosion: Pipework, vessels, tanks and supports. Corrosion under insulation: Insulated pipework, vessels and equipment exposed to moisture. Erosion: High velocity flow, bends, reducers, slurry or particle-laden streams.01 · What to notice
The pump-skid case reports fresh liquid and odour near a seal; the person asked to clean it has not established the source or safe conditions.
02 · Trace the harm
A pressurised leak can expose people, form a pool or vapour cloud and find ignition; cleaning can bring a worker into the release area.
03 · Test the safeguard
Establish the cause, repair and test through the controlled work process, then confirm no ongoing release and acceptable barrier status before authorised return to service.
Lesson 6
A stopped pump still contains hazardous energy
Stopping is an operating action; isolation establishes and maintains the safe condition needed for the work.
Understand this fuel-industry task
Pressing stop removes an operating demand from the motor; it does not necessarily secure the electrical supply, prevent a remote start or isolate the fuel system. Suction head, discharge pressure, reverse flow, bypasses and trapped liquid can remain. A pump that is silent can still release fuel when a cover or connection is opened, and another operation elsewhere on the site can change the pressure at this maintenance location.
The proposed guard removal and maintenance must therefore start with an authorised isolation plan covering the actual work boundary and every relevant energy or material source. Suitable isolation, securing, controlled draining or venting and proving the required safe condition are linked steps. Selection depends on the substance, task and consequences; neither a stop button nor an assumed universal valve arrangement is adequate. Coordinate the permit and connected operations, identify who controls the locks and boundary, and preserve the isolation across a shift change. Handback then accounts for people, tools, guards and temporary items before controlled reinstatement.
Trace the initiating event and loss of control
Step
Case pathway
initiatingEvent
Maintenance is proposed after stopping the motor but before controlling the complete work boundary.
lossOfControl
Remote restart, reverse flow or trapped pressure reaches the equipment while a person is exposed.
consequences
Entanglement, crushing, fuel injection or chemical exposure, fire and a wider loss of containment can follow.
Choose effective safeguards
Control priority
Practical application
Eliminate
Avoid intrusive work on hazardous systems where possible.
Substitute
Use safer inspection and maintenance methods.
Engineer
Provide suitable isolation, contained drains and verified protective arrangements.
Manage
Coordinate permits, contractor competence, gas testing, rescue and handback as applicable.
Protect
Select suitable task PPE and rescue protection; never improvise a confined-space rescue.
Plan the isolation using current plant information and a field identification of electrical, process and other stored-energy sources, including bypass and reverse-flow paths.
Apply the authorised isolation standard for the task, secure it against interference and establish control of locks, tags and any shared boundary.
Prove the required safe condition through an approved method before work begins; release pressure or liquid only through suitable controlled arrangements.
Keep permits, shift handover and connected operations aligned, then reinstate through an explicit handback and readiness decision rather than simply removing a tag.
Verify effectiveness before release
Ask the authorised team to explain each source and how the selected isolation prevents it reaching the work; compare the explanation with the actual equipment identity.
Review the recorded proving evidence and the arrangements for maintaining isolation while the equipment is open or the work spans shifts.
Check final accounting for workers, guards, drains, vents and temporary isolations, and verify operations have accepted the correct configuration before restart.
Transfer the decision to the workplace
Apply the same distinction to tanker disconnection, instrument-air maintenance and opening a steam trap: no motion or sound does not prove zero hazardous energy.
Which energy sources must be identified?
A competent isolation plan looks beyond the obvious power switch.
Energy can be electrical, mechanical, hydraulic, pneumatic, chemical, thermal, gravitational, pressure, radiation or stored in springs, capacitors, accumulators, batteries, elevated loads and process inventory. Multi-energy systems are common in modern plant: a conveyor may have electrical drives, stored mechanical energy, pneumatic actuators, gravity hazards and automated controls. See HSG253 - The safe isolation of plant and equipment.
The isolation plan should be based on physical walkdown, drawings, operating knowledge, previous changes, site experience and challenge from the workers doing the task.
Energy source
How it can harm
Control example
Electrical
Shock, electrocution, arc flash or unexpected start.
Isolate, lock, prove dead and control back-feed.
Mechanical movement
Crushing, shearing, entanglement or amputation.
Block, pin, disengage and verify no movement.
Pressure
Injection, blast, projectile, release or burns.
Depressurise, drain, vent and verify.
Hydraulic/pneumatic
Unexpected actuator movement or stored pressure.
Isolate supply, bleed accumulators and secure parts.
Thermal
Burns from steam, hot product or cold cryogenic release.
Cool, drain, vent and wear suitable PPE.
Chemical/process inventory
Toxic, corrosive, flammable or asphyxiant release.
Clean, purge, gas test and use positive isolation where needed.
Gravity
Falling loads or drifting equipment.
Chock, prop, block or lower to safe position.
Applied case
The motor is stationary and its local stop button is pressed. A fitter is ready to remove the coupling guard, but the pump remains connected to the fuel manifold.
Decision and explanation
Control, secure and prove all relevant energy and process sources under the authorised isolation plan.
The maintenance boundary includes the connected fuel and possible restart sources. A quiet motor provides no assurance about those hazards. A reminder is vulnerable to misunderstanding and handover failure. It does not physically secure or prove the required isolation. A zero reading cannot establish electrical isolation, prevent remote energisation or control process pressure and reverse flow.
Look closer
Photorealistic illustrative UK workplace scene. Review checks: Elimination: Use a suitable external diagnostic or redesigned sampling arrangement if it removes the intrusive valve task. Engineering controls: Use an approved closed sampling system or safely removable module to reduce open access. Engineering controls: Isolate connected feeds and energy sources, remove hazardous contents and verify the defined safe state before breaking containment. Engineering controls: Provide suitable drainage, containment, local extraction and support for the valve and connected pipework. Administrative controls: Agree the permit, contamination checks, line identification and hand-back with operations and contractors.01 · What to notice
The maintenance request proposes pump work after pressing stop; suction, discharge, electrical and trapped-energy sources have not been positively controlled.
02 · Trace the harm
Remote restart, reverse flow or stored product can injure the maintainer even when the motor is not turning.
03 · Test the safeguard
Have the authorised team identify each source and prove the required safe condition, then control reinstatement so people, tools and temporary isolations are accounted for.
Lesson 7
Abnormal vibration: protect the operating envelope
An excursion is information requiring a defined response; surviving previous excursions does not validate them.
Understand this fuel-industry task
A pump needs more than an acceptable delivery total. Its approved operating envelope includes the conditions under which its mechanical parts, seals and protective systems can perform safely. Flow, suction conditions, temperature, vibration and product properties can be relevant. A change in these indicators may reveal a restriction, unsuitable inlet condition, mechanical deterioration or another fault. The scene’s low-flow and rising-vibration evidence needs diagnosis; it does not identify a single cause by itself.
Increasing speed to recover the loading target can worsen an unresolved condition or move another parameter beyond its limit. Use the plant’s abnormal-operation instructions to reach the required safe state and involve competent operations and engineering support. Normal targets, operator intervention points, protective trips and physical design limits have different purposes and must not be treated as interchangeable. A trip is a last protective action, not a desirable operating target. After diagnosis and repair, the restart decision should show that the pump and connected route can meet the actual duty without suppressing alarms or relying on repeated excursions.
Trace the initiating event and loss of control
Step
Case pathway
initiatingEvent
A developing restriction, unsuitable suction condition or mechanical defect produces abnormal vibration, temperature or flow.
lossOfControl
The team compensates by unreviewed speed changes and continues outside demonstrated safe operating conditions.
consequences
Mechanical damage and seal failure can release fuel, creating exposure, fire and interruption of connected safety functions.
Choose effective safeguards
Control priority
Practical application
Eliminate
Remove unused connections and avoid opening live process systems.
Substitute
Choose a less intrusive maintenance or sampling method where suitable.
Engineer
Provide an assessed isolation boundary, contained drains and means to prove its condition.
Manage
Walk down equipment identity, record isolation, coordinate users and control reinstatement.
Protect
Use PPE selected for the substances and residual exposure; it cannot make a live line safe.
Use approved parameter limits, trend information and explicit abnormal-response actions; ensure operators can identify when reduction, shutdown or escalation is required.
Pause the loading demand and reach the defined safe state when conditions cannot be demonstrated; do not disable protection to recover production.
Investigate the cause across the pump, suction source, transfer route and product conditions rather than replacing a component without a system review.
Approve changes to speed, duty or settings through the appropriate technical and change-control process, with clear monitoring for the authorised return.
Verify effectiveness before release
Compare actual trends and operating mode with the controlled envelope, including the basis and required response for each relevant abnormal indication.
Review diagnosis and corrective evidence against the whole flow path; confirm that an apparently normal single reading does not conceal unresolved deterioration.
Observe or review the approved monitored restart and demonstrate stable duty with protection available, then retain a clear response if the abnormal pattern returns.
Transfer the decision to the workplace
Use this logic for tank levels, VRU liquid level and utility pressure: a production target must remain inside the approved safety conditions and available protections.
What should happen when the envelope is challenged?
Excursions should trigger stabilisation, escalation, recording and learning, not informal tolerance.
An excursion is useful safety information. Even if no harm occurs, it may show that the process, procedure, alarm, instrument, training or staffing assumption is weak. See HSE operating procedures for the relevant source and scope.
Repeated small excursions are especially important. They often reveal normalisation before a serious loss of control.
Excursion stage
Expected response
Evidence
Approaching limit
Recognise trend and take planned corrective action.
Hold or safely stop an operation with an impaired credited safety function unless the designated competent authority demonstrates effective protection for that specific operating state. Record the permitted limits, concurrent impairments, owner, expiry and review triggers. A signature and time limit alone do not make a bypass safe.HSE control systems
Applied case
The loading rate drops while vibration and temperature depart from the normal trend. Dispatch asks for higher speed to finish a waiting tanker.
Decision and explanation
Follow the abnormal-response method, establish a safe condition and diagnose before an authorised monitored return.
The target is not a safety criterion. Compensating for an unknown fault may increase stress or worsen the condition causing the deviation. Operator intervention is intended to prevent escalation before the protective trip. Waiting for it can consume the available safety margin. This uses the technical operating basis and preserves protective margins rather than making an unsupported production adjustment.
Look closer
Photorealistic illustrative UK workplace scene. Review checks: Before transfer: Confirm tank level, ullage, route line-up, alarm status and transfer limits. (HSE) During transfer: Monitor trend, flow, level and alarm status against defined response time. (HSE) Alarm issue: Alarm impairment reviewed, temporary restriction set and senior approval required. (HSE)01 · What to notice
The exercise trend reports abnormal vibration, temperature and flow compared with the approved operating envelope; operators propose increasing speed.
02 · Trace the harm
An unsuitable suction condition, blockage or mechanical defect can damage seals or equipment and lead to loss of containment.
03 · Test the safeguard
Confirm the cause has been established, required defects corrected and an authorised monitored return demonstrates acceptable operation without defeating protection.
Lesson 8
When the procedure and plant disagree, stop the uncertain step
A signature approves a document; field validation shows whether a person can safely use it.
Understand this fuel-industry task
The checklist at this pump refers to a valve identifier that no longer exists and an indicator removed by an earlier change. This makes the next action ambiguous at exactly the point where the procedure should control a safety-critical decision. An operator choosing the most plausible nearby valve may create the wrong route, start against an unsuitable condition or leave protection unavailable. The worker’s experience can help identify the discrepancy, but it cannot authorise an undocumented replacement step.
A useful operating procedure reflects the actual task and field configuration. It establishes prerequisites, clear identifiers, checks, limits, hold points and what to do when a condition is absent. Operators should help design and validate it using a safe walkthrough that tests how the work is really done. Here the immediate decision is to hold the uncertain action and preserve the safe plant state while authorised people resolve the mismatch. Investigate the change-control gap too: if drawings, labels, training and procedures were not updated together, other tasks may have inherited the same uncertainty.
Trace the initiating event and loss of control
Step
Case pathway
initiatingEvent
A plant modification or document error leaves incorrect equipment references in the start instruction.
lossOfControl
The operator must guess during line-up or invent a substitute procedure without a reviewed technical basis.
consequences
Wrong routing, unsafe start, equipment damage or loss of containment can expose people and create a major-accident demand.
Choose effective safeguards
Control priority
Practical application
Eliminate
Remove unnecessary safety-critical manual tasks and avoid excessive workload.
Substitute
Use clearer interfaces and simpler tasks where suitable.
Engineer
Design usable displays, protected workstations and accessible clean welfare facilities.
Manage
Plan staffing, rest, handovers, competence and a culture that acts on concerns.
Protect
Provide appropriate personal protection and hygiene for remaining task exposure.
Hold the ambiguous step and use the authorised discrepancy and escalation route, maintaining the defined safe plant condition.
Write and review instructions with competent operators and technical owners, matching each critical step to current equipment identities and protection requirements.
Validate the complete task safely in the field, including unavailable equipment, abnormal indications and the decision to stop.
Control plant and procedural change together; remove obsolete versions, communicate revisions and establish task competence before release for use.
Verify effectiveness before release
Compare the current instruction, labels and drawings at each safety-critical point, checking that two different users identify the same equipment and required action.
Ask a user to explain the response to the missing indicator without improvisation; confirm the approved alternative and its limits are explicit if one exists.
Trace the discrepancy to its change history and review related procedures and training so the correction addresses the cause rather than only one printed sheet.
Transfer the decision to the workplace
Use the same procedure-to-plant comparison for tanker connection, roof landing and firewater impairment: wrong identifiers and missing prerequisites are stop points, not literacy tests.
How should SOP validation and walkdown be done?
Validation should test the procedure against the actual workplace before it is issued or relied on.
A desk review can find grammar mistakes. A walkdown finds operational traps: missing valve labels, unreachable gauges, poor lighting, steps in the wrong order, alarms with unclear response, PPE that makes a task impossible, or instructions that conflict with the permit or control screen.
The best validation involves operators, supervisors, engineering, HSE and technical authority where needed. It should check normal, abnormal and foreseeable upset conditions, then update training and records.
Validation check
What to test
Common finding
Physical sequence
Can the steps be done in the order written?
Valve or sample point is not accessible.
Labels and nomenclature
Do procedure names match plant labels?
Different names used in field and control room.
Critical parameters
Are limits and readings visible?
Gauge not readable from task position.
Human factors
Can one person realistically do the step safely?
Task needs two people or better communication.
Abnormal states
Does the procedure say what to do if a reading is wrong?
Stop condition missing.
Interface
Does it align with permits, isolations and handover?
Procedure conflicts with permit requirement.
Records
Can the required evidence be captured?
Checklist misses critical control verification.
A walkdown is not permission to operate unsafe plant or perform a hazardous step just to test the document. Use non-operational checks and safe simulation where appropriate; live trials need their own assessed, authorised commissioning arrangements and readiness checks. HSE COMAH operating procedures.
Applied case
The start sheet names an old valve. A senior operator knows which new valve they believe it means and offers to explain it verbally each shift.
Decision and explanation
Hold the step and reconcile the controlled instruction with the current plant through authorised review and field validation.
Experience is useful input, but a recurring verbal workaround leaves new staff, other shifts and abnormal tasks exposed to inconsistent interpretation. This produces a usable and sustainable work method, and checks whether the original modification left wider document or competence gaps. Uncontrolled copies can diverge and hide a technical disagreement. A common authorised method must resolve the actual safety basis.
Look closer
Related SHAREhub topic guide. Apply its prompts to the specific case below.01 · What to notice
The pump-start checklist refers to an old valve identifier and a deleted local indicator; the operator must guess which control is intended.
02 · Trace the harm
An unusable instruction encourages improvisation during a safety-critical task and can produce a wrong line-up or unprotected start.
03 · Test the safeguard
Use a safe task walk-through with a competent user to reconcile every critical identifier and action, then confirm the current revision is available at use.
Lesson 9
Alarm flooding: acknowledgement is not corrective action
Judge an alarm by the timely, correct action it enables, not by whether it sounded or was acknowledged.
Understand this fuel-industry task
An alarm should tell the operator about a condition needing a defined action within an available opportunity. Repeated pump and tank alarms can overwhelm attention, particularly when nuisance warnings, standing faults and genuine developing events share the same display. Acknowledging a message records that it has been seen; it does not restore a leaking seal, incorrect level or failing utility. The learner should therefore follow the abnormal process condition behind the sound, not treat the sound itself as the hazard.
This operator cannot identify priority or the expected response. That points to alarm design, procedure and workload as well as training. Rationalisation establishes why each alarm exists, its consequence, priority and actionable response. The whole chain includes detection, display, diagnosis, communication and the action that restores or protects the process. If an operator action is credited as a safeguard, the available time, information and competing demands must make that action credible. Repeated unofficial shelving or suppression can conceal deterioration; authorised temporary suppression still needs ownership, limits, review and restoration.
Trace the initiating event and loss of control
Step
Case pathway
initiatingEvent
A developing disturbance produces multiple alarms against a background of nuisance and standing messages.
lossOfControl
The operator repeatedly acknowledges or suppresses warnings without identifying and completing the required protective action.
consequences
An important excursion is missed or acted on too late, allowing loss of containment, fire or escalation while the display appears managed.
Choose effective safeguards
Control priority
Practical application
Eliminate
Remove unnecessary hazardous material, energy and transfer operations.
Substitute
Choose a less hazardous product or simpler process where practicable.
Engineer
Prefer effective engineered preventive and mitigative barriers.
Manage
Use suitable hazard studies, verify assumptions and act on weak-signal indicators.
Protect
Keep PPE as an additional exposure control within its actual capabilities.
Use the site’s abnormal-situation response and competent assistance to reach the required safe state; do not suppress warnings merely to maintain throughput.
Rationalise alarms by scenario, purpose, consequence, priority, response and available time, removing nuisance at its cause through controlled engineering work.
Design displays, instructions, staffing and communication for realistic disturbance workload, including simultaneous pump, tank and utility conditions.
Manage standing alarms and temporary shelving through defined authority, records, operating restrictions and timely restoration; retain learning from recurring demands.
Verify effectiveness before release
Review actual event sequences and standing-alarm records with operators, asking which condition required action and whether the useful signal was distinguishable.
Exercise a representative abnormal scenario safely and confirm the full response can be completed with the expected competing duties and communications.
Check that nuisance-alarm corrections and suppression removal have been verified against the technical basis, rather than measuring success only by a quieter panel.
Transfer the decision to the workplace
At the VRU detector and loading ESD, trace the same signal-to-safe-state chain; a healthy display or recorded command cannot prove the final protective result.
How should control rooms prevent an alarm flood from hiding a critical signal?
Every alarm should have a useful purpose and defined action; prioritisation and workload must remain usable during an abnormal event.
Standing, repeating and nuisance alarms can mask a new fire, gas or shutdown signal. A flood during an upset can exceed human diagnosis capacity. Record alarm rate, standing duration, suppressed state, operator response and real outcomes; investigate the initiating fault rather than teaching staff to acknowledge messages faster.
Rationalise alarms by consequence and required response, make safety-critical alerts distinct, keep response procedures current and resource the abnormal scenario. Shelving, suppression or mode changes need defined authority, visibility, expiry and return to service. Never silence or bypass a credited safety alarm for convenience.
Official context: HSE alarm management; HSE control systems. Check the stated jurisdiction and premises; practical examples require a competent site assessment.
Applied case
During transfer, several pump warnings repeat and a tank alarm appears among them. The operator proposes silencing the familiar group permanently.
Decision and explanation
Apply the abnormal-response method, then review priorities, causes and real operator response before authorising alarm changes.
This addresses the current plant condition and the underlying alarm system. A lower alarm count alone is not evidence of adequate risk control. More noise does not establish priority, a usable response or sufficient attention. It can make workload and diagnosis worse. Acknowledgement and process recovery are different events. The required response must actually change or protect the hazardous condition.
Look closer
Photorealistic illustrative UK workplace scene. Review checks: Elimination: Defer the detector work to a planned shutdown where practicable. Substitution: No separate approved substitution is specified; assess a genuinely safer material or process before relying on lower controls. Engineering controls: Use named, time-limited impairment or bypass authority: identify disabled functions, remaining protection, restrictions, control-room coordination and abort actions. Administrative controls: Use named, time-limited impairment or bypass authority: identify disabled functions, remaining protection, restrictions, control-room coordination and abort actions. PPE: Select compatible task PPE for residual fire, chemical, eye and hand hazards.01 · What to notice
The control exercise shows repeated pump and tank alarms during transfer; the operator reports frequent acknowledgements but cannot identify the priority or required response.
02 · Trace the harm
Alarm flooding, nuisance alarms and unclear response can obscure an escalating event; an acknowledgement does not correct its cause.
03 · Test the safeguard
Practise a safe representative scenario and examine alarm records: the operator detects the important condition and completes the justified response within the required opportunity.
Lesson 10
After a trip: a reset is not a readiness review
Safe restoration is a controlled task with evidence, ownership and a defined sequence.
Understand this fuel-industry task
A protective trip means a defined condition has demanded a safety action. The immediate plant state may have prevented escalation, but resetting the indication does not identify why it happened or establish that the process is ready. A fault could remain in the pump, transfer route, tank protection or supporting utility. Maintenance may have started during the shutdown, creating a new exposure if energy or flow returns unexpectedly.
The scene proposes an immediate reset before checking the cause and connected work. Preserve the agreed safe state while the authorised team reviews the trip, permits, isolations, bypasses, equipment condition and dependent protection. Restoration should separate diagnosis, repair, safe handback, protective-function verification and restart authority. Some recoveries need a formal pre-start review proportionate to the changes and consequences. Returning power or clearing a trip should not silently reintroduce an unsafe operation. A monitored restart follows the approved sequence, with clear communications and stop conditions if the abnormal condition recurs.
Trace the initiating event and loss of control
Step
Case pathway
initiatingEvent
A pump or connected protective system trips during transfer.
lossOfControl
The team resets and restarts before resolving the initiating condition, open work or unavailable protection.
consequences
The hazardous condition returns, equipment energises around maintainers or a protective barrier is defeated, leading to release or injury.
Choose effective safeguards
Control priority
Practical application
Eliminate
Avoid intrusive work on hazardous systems where possible.
Substitute
Use safer inspection and maintenance methods.
Engineer
Provide suitable isolation, contained drains and verified protective arrangements.
Manage
Coordinate permits, contractor competence, gas testing, rescue and handback as applicable.
Protect
Select suitable task PPE and rescue protection; never improvise a confined-space rescue.
Keep the affected operation in its defined safe state until the trip’s cause and required recovery are understood by authorised people.
Reconcile permits, isolations and simultaneous work across the pump, tanks, gantry and utilities before any restoration of energy or flow.
Confirm protection, alarms and any temporary changes are correctly restored and verified, with explicit limits for unresolved conditions.
Authorise a controlled monitored restart using current procedures, competent personnel and communication with everyone affected by the restored operation.
Verify effectiveness before release
Trace the first relevant event in the trip history and supporting field evidence; distinguish the initiating cause from consequential alarms.
Check actual work and isolation status against the permit board and shift handover, including tasks started after the trip.
Review cause closure, relevant safety-function tests and the readiness decision, then check restart observations demonstrate stable operation without repeated resets.
Transfer the decision to the workplace
Use this approach after utility return, VRU shutdown or firewater maintenance; availability of energy is only one prerequisite for safe operation.
What should a start-up readiness gate include?
A readiness gate proves the plant, people, procedures and safeguards are ready before hazardous operation begins.
The readiness gate should be stronger when the consequence of wrong start-up is serious. It should be a decision point, not a paperwork ritual. HSE operating procedures.
The accountable person should be able to say which items must be closed, which may remain open under restriction, and who has authority to approve exceptions.
The punch-list categories and named release decision below are illustrative site controls. HSE describes commissioning checks, reservation lists and responsible supervision; the competent site owner must set the actual start-up criteria.
Readiness item
What to verify
Cannot start if
Design and plant complete
Plant matches approved design or authorised change.
Alarms, trips, interlocks, relief and emergency systems available.
Safety-critical barrier unavailable without competent technical review and an approved safe alternative. (HSE control systems)
Training and staffing
Competent people available for start-up phase.
New or understaffed team without support. (HSE procedures)
Punch list
Open actions classified by safety significance.
Safety-critical action remains open or its effect on the start-up cannot be resolved. (HSE commissioning)
Authorised start-up
Named person approves start-up conditions.
Pressure to start without decision owner. (HSE procedures)
Illustrative purge evidence: the approved procedure names the medium, target atmosphere, sampling locations, selected and checked instruments, safe discharge route and acceptance record. Unknown evidence means no introduction of feed or ignition. Inerting can also create oxygen-deficient atmospheres at connected spaces and discharge points; retain the separate atmosphere and entry controls.HSE safe isolation, HSG253HSE operating procedures
Applied case
A pump trips twice. The display now permits reset, while a contractor is investigating an associated instrument and dispatch is waiting.
Decision and explanation
Confirm the cause, open work, configuration and protection, then obtain authorisation for the approved monitored restart.
A reset permissive does not establish cause removal or work clearance. Repeated attempts can challenge the same protection while creating an exposure for the contractor. An improvised operating change does not replace isolation handback, instrument restoration and the authorised technical recovery decision. Readiness integrates both the process fault and the work performed during the shutdown. It is more than clearing a latched indication.
Look closer
Photorealistic illustrative UK workplace scene. Review checks: Permit close: Close only after physical handback and field inspection. (HSE source) Line-up: Use marked line-up sheet and independent check for critical valves. (HSE source) Safeguards: Do not start with an unavailable essential alarm or trip; competent technical authority must prove any effective alternative within defined safe limits before a specific restart decision. (HSE source)01 · What to notice
After a protective trip, the proposed response is to reset and restart before the cause, connected permits and protection status are checked.
02 · Trace the harm
Restart can reintroduce a hazardous condition, energise equipment under work or defeat the protection that prevented escalation.
03 · Test the safeguard
Check the cause is resolved, permits handed back, plant configuration and safety functions proved, and the actual restart follows the approved monitored sequence.
Lesson 11
Prove the isolation before opening containment
Isolation is a verified physical condition with controlled ownership, not a valve command or a reassuring gauge alone.
Understand this fuel-industry task
The proposed line break relies on a closed control valve and an unverified zero pressure reading. A control valve can pass product, the gauge connection can be blocked or isolated, and a low point or branch can retain liquid after the main line appears drained. The work therefore remains connected to possible fuel, pressure and other energy sources until the complete boundary is identified, secured and proved through the approved method. Pressing stop or seeing no movement does not establish a safe intrusive-work condition.
Select the isolation standard through competent assessment of the actual hazardous inventory, system and task, using relevant safe-isolation guidance. Different work can require different arrangements; do not convert one model example into a universal valve recipe. The plan includes all feeds and backfeeds, draining and venting to a suitable system, stored contents, locks or other secured devices, verification and controlled reinstatement. A permit records authority and conditions, while the isolation evidence demonstrates the physical boundary. Slowly loosening the flange to 'see if anything comes out' is not an acceptable substitute for proving that boundary safely.
Trace the initiating event and loss of control
Step
Case pathway
initiatingEvent
A flange is opened while isolation depends on one unproved valve and a potentially misleading instrument reading.
lossOfControl
Live product or trapped energy reaches the open joint when the assumed boundary fails or proves incomplete.
consequences
Workers can be struck or sprayed, inhale fuel vapour or face ignition; the connected inventory can sustain a substantial release.
Choose effective safeguards
Control priority
Practical application
Eliminate
Avoid intrusive work on hazardous systems where possible.
Substitute
Use safer inspection and maintenance methods.
Engineer
Provide suitable isolation, contained drains and verified protective arrangements.
Manage
Coordinate permits, contractor competence, gas testing, rescue and handback as applicable.
Protect
Select suitable task PPE and rescue protection; never improvise a confined-space rescue.
Hold the line break and have competent authorised people plan the required isolation for the real service, work scope and potential consequences.
Identify all material and energy sources, including bypasses, backfeeds, elevated inventories and trapped pockets; use current plant information verified at the field location.
Apply and secure the selected boundary, remove pressure and contents through assessed safe routes, and prove the required condition using approved techniques and independent checks where specified.
Link the isolation to the permit and ownership arrangements, manage any change or loss of integrity, and control handback and reinstatement after the intrusive work is complete.
Verify effectiveness before release
Compare the isolation plan with the complete physical source-to-worksite system and current drawings, including small branches and bypasses rather than only the nearest main valves.
Review the actual proving evidence and instrument suitability: establish what demonstrates the required safe condition and what might make a zero indication misleading.
Ask the work team who controls locks, how a passing boundary is detected and what stops work; verify that reinstatement cannot bypass people or unfinished work.
Transfer the decision to the workplace
Use the same boundary-and-proof reasoning for pump removal, instrument maintenance, filter opening and tank entry, tailoring the isolation standard to the actual task.
What makes line breaking especially hazardous?
Line breaking is the point where containment is opened and hidden assumptions are tested.
The first crack of a flange can release pressure, vapour, liquid, sludge, solids, steam, corrosive material, toxic gas or flammable atmosphere. Even a small residual quantity can injure a worker if it is hot, under pressure, toxic or reactive. See HSG253 - The safe isolation of plant and equipment.
Line breaking should therefore use a deliberate break plan: verify isolation, control drainage, position workers out of the line of fire, loosen bolts cautiously, use drip trays or containment, wear task-specific PPE, control ignition sources and stop if unexpected pressure or material appears.
Control point
Why it matters
Good evidence
Line contents known
Controls depend on substance.
SDS/process data and sampling where needed.
Pressure removed
Prevents violent release.
Vent/drain verification.
Worker position
Avoids line of fire.
Pre-job briefing and observation.
Containment
Controls spills and exposure.
Drip tray, bunding or drain isolation.
PPE/RPE
Protects against residual risk.
Task-specific PPE assessment.
Stop criteria
Controls unexpected release.
Briefed hold points.
Applied case
A flange must be removed. The control valve is closed and a gauge reads zero, but its connection has not been checked and the line has an unreviewed bypass.
Decision and explanation
A complete suitable isolation is planned, secured, drained or vented safely and proved through the authorised system.
A slower line break still uses loss of containment as the test and can expose the worker to pressure or hazardous product. This addresses the real sources and verifies the safe condition before containment is broken, rather than relying on a single indication. A stopped motor does not prevent gravity flow, backfeed, passing valves or release of trapped pressure and contents.
Look closer
Photorealistic illustrative UK workplace scene. Review checks: Source / exposure: Common failures are usually predictable and should be designed out or caught by assurance. Possible consequence: Wrong line identity, passing valves, back-feed, blocked drains and incomplete handback can each defeat a signed plan. Control failure: Treat these as testable failure modes with named checks, not as reminders to be careful.01 · What to notice
The maintenance request proposes opening a manifold flange with only a closed control valve and a zero reading on an unverified gauge.
02 · Trace the harm
A passing valve, trapped pocket or wrong isolation can release hazardous product or stored energy when containment is broken.
03 · Test the safeguard
At a suitable safe stage, prove the isolation and zero-energy condition using approved methods, independently confirm the boundary where required and retain controlled handback.
Lesson 12
Resolve interacting jobs before either permit is released
Control the interaction between activities, including changes over time; separate paperwork does not create physical independence.
Understand this fuel-industry task
The welding team and transfer team each have a reasonable description of their own task, but the permits have not been checked together. A fuel release, vent or opened system on one branch can affect hot work on the next; barriers and a welding screen do not guarantee that vapour stays within a drawn work boundary. One team may also change a valve, isolation or operating state relied on by the other. The hazard lies in the interaction, so two individually approved documents do not automatically control it.
Consider whether one activity can be eliminated, performed elsewhere or rescheduled to remove the interface. If concurrent work remains justified, a competent coordinated plan must address actual plant status, release pathways, ignition sources, isolation, atmosphere checks, nearby operations and response. Gas testing represents conditions at the place and time tested; it cannot guarantee that a later transfer creates no vapour. Define what changes suspend the work and who communicates and authorises any resumption. The model's spacing and marker position are teaching aids rather than a real hot-work separation distance.
Trace the initiating event and loss of control
Step
Case pathway
initiatingEvent
Hot work and fuel transfer are released independently without considering their shared process and atmosphere interfaces.
lossOfControl
Flammable material or a changed plant condition reaches an ignition source or invalidates a work isolation.
consequences
A local release can ignite and escalate through adjacent pipework, with danger to both teams and responders.
Choose effective safeguards
Control priority
Practical application
Eliminate
Avoid intrusive work on hazardous systems where possible.
Substitute
Use safer inspection and maintenance methods.
Engineer
Provide suitable isolation, contained drains and verified protective arrangements.
Manage
Coordinate permits, contractor competence, gas testing, rescue and handback as applicable.
Protect
Select suitable task PPE and rescue protection; never improvise a confined-space rescue.
Hold conflicting activities and first consider avoiding the overlap through rescheduling, relocation or a safer work method rather than adding paperwork to simultaneous exposure.
Assign a coordinating authority to reconcile permits, connected plant, isolation boundaries and operational changes with both work groups and the area controller.
Where concurrent work is justified, establish assessed release and ignition controls, suitable atmosphere checks, monitoring where required and usable emergency and escape arrangements.
Define specific suspension triggers, including changed transfer status, alarms, weather or work scope where relevant; require communication and controlled reassessment before restart.
Verify effectiveness before release
Walk the shared boundary safely with both teams and compare actual live lines, vents, drains and escape routes with the coordinated permit conditions.
Ask each group what the other is doing, which plant states the work depends on and what event requires immediate suspension and notification.
Review atmosphere-testing scope and timing, simultaneous-work records and change control; confirm that a prior satisfactory reading was not treated as permanent permission.
Transfer the decision to the workplace
Look for the same simultaneous-operation risk when lifting over live plant, disabling firewater during transfer, opening drains or conducting tank work beside neighbouring operations.
How should permits and SIMOPS be connected?
Permits should communicate live work, plant state and interactions, not just authorise isolated tasks.
HSE permit guidance calls for linked permits when separate jobs affect one another.
HSE emphasises that a permit does not make a job safe by itself. It is a communication and coordination tool. For SIMOPS, that means permit issuers and area authorities must see related permits and consider simultaneous tasks and interdependent activities.
A permit board, electronic permit view or control-room display should show live, suspended and closed work clearly. It should also show isolations, bypasses, alarms, emergency restrictions and handover notes that could affect other work.
Welding is permitted beside the manifold. Another controller schedules transfer through a neighbouring branch and argues that it has a separate permit and no shared valve.
Decision and explanation
Hold the interaction and have the authorised controller reconcile the shared release, ignition and plant-status risks with both teams.
The relevant interface extends beyond a shared valve. Vapour pathways, changing operations and emergency arrangements must be considered together. Separate signatures can preserve conflicting assumptions. The permit system must coordinate the relationship between the tasks. A past reading does not predict conditions during the new transfer, and a screen is not a general barrier to flammable vapour.
Look closer
Photorealistic illustrative UK workplace scene. Review checks: Eliminate: Reschedule work to remove the simultaneous exposure. Reduce: Assess a suitable cold method instead of hot work. Isolate: Prove isolation, contain material and establish effective separation. Control: Coordinate linked permits and changing conditions with both crews. PPE: PPE cannot compensate for an unresolved release and ignition interaction.01 · What to notice
An exercise schedules hot work beside a manifold while another team plans product transfer through a neighbouring branch; the permits have not been cross-checked.
02 · Trace the harm
A release or vented vapour can reach an ignition source, and one team’s isolation or restart may defeat the other team’s assumptions.
03 · Test the safeguard
Walk the shared boundary with both work groups and verify the actual plant status, permit conditions and stop triggers, not merely that two permits exist.
Lesson 13
Hot work: a permit must describe proven conditions
Permit validity depends on the real job and current conditions, including work happening around it.
Understand this fuel-industry task
Stand at the maintenance board and trace the proposed welding job back to the pipework, tanks and nearby drains. The signature authorises a defined job under defined conditions; it cannot remove fuel, secure an isolation or stop vapour travelling from another activity. Cutting and grinding can provide ignition as well as heat. A small task on an apparently empty item can therefore initiate a much larger event when residue, connected equipment or surrounding operations have been overlooked.
The learning decision begins before choosing a permit form: can the repair use a suitable cold method, be moved to a safer workshop or be scheduled with the equipment safely removed from service? Where hot work remains necessary, the issuer and performing team need a common understanding of the exact boundary, preparation, isolation, atmosphere assessment and changing conditions. Link concurrent permits and transfer status. A gas test is evidence about the tested place and time, not a guarantee for the whole shift. Suspension, shift handover and controlled handback are part of the same system.
Trace the initiating event and loss of control
Step
Case pathway
initiatingEvent
A repair is released on paperwork while residue or a connected drain remains able to supply flammable vapour.
lossOfControl
Welding or grinding ignites an atmosphere outside the assumed work boundary.
consequences
Workers suffer burns or blast injury, and fire can damage adjacent fuel systems and interrupt emergency access.
Choose effective safeguards
Control priority
Practical application
Eliminate
Avoid intrusive work on hazardous systems where possible.
Substitute
Use safer inspection and maintenance methods.
Engineer
Provide suitable isolation, contained drains and verified protective arrangements.
Manage
Coordinate permits, contractor competence, gas testing, rescue and handback as applicable.
Protect
Select suitable task PPE and rescue protection; never improvise a confined-space rescue.
Challenge the need for ignition-producing work and choose an assessed safer method or location wherever reasonably practicable.
Define and verify the actual isolation and preparation against current plant identification, including connections and nearby drain or vent paths.
Use competent atmosphere assessment and appropriate testing or monitoring, coordinated with changing operations and explicit stop conditions.
Provide the assessed fire precautions, supervision and post-work checks; suspend on change and formally hand back before reinstatement.
Verify effectiveness before release
Walk the job with the issuer and contractor: can both identify the equipment, boundary, precautions and activities that invalidate the permit?
Reconcile the isolation record, actual plant state and atmosphere evidence; challenge an unexplained difference before work is released.
Use a safe tabletop change of conditions to test who stops work, suspends linked permits and confirms completion and reinstatement.
Transfer the decision to the workplace
Apply the same reasoning to line breaking, electrical work and excavation: authorisation must be supported by verified task-specific controls.
What does a good permit lifecycle look like?
A good permit follows the job from planning through risk control, issue, live monitoring, suspension, handback and audit.
The lifecycle is the backbone of permit quality: plan -> risk assess -> isolate -> test -> issue -> brief -> perform -> monitor -> suspend if needed -> hand back -> close out -> audit. Each stage has a different purpose and a different owner. Skipping one stage often creates the conditions for a serious incident.
For process safety, the lifecycle also protects major accident barriers. A line break, hot work task or confined-space entry may defeat containment, introduce ignition, remove safeguards or put people into an atmosphere that can change. The permit system must connect to isolation standards, gas testing, SIMOPS, emergency response and management of change.
Stage
Purpose
Evidence
Plan
Decide whether a permit is required and define work scope.
Work pack, risk assessment and permit type.
Risk assess
Identify hazards, affected people, adjacent work and emergency needs.
Task risk assessment and field check.
Isolate and make safe
Remove, lock, tag, drain, vent or control energy and substances.
Isolation certificate and verification.
Test and verify
Confirm atmosphere, dead state, zero energy or safe access where needed.
Gas test, prove-dead record or pressure/containment check.
Issue and brief
Authorise only when conditions and roles are understood.
Signed permit and toolbox talk.
Monitor
Check that work stays within permit conditions.
Area checks, fire watch, gas retest or supervisor log.
Suspend
Pause work when conditions, scope, weather, SIMOPS or shift change affect safety.
Suspension note and revalidation.
Hand back and close
Confirm work complete, area safe and safeguards reinstated.
Handback signature and restart checks.
Audit
Check permit quality and field reality.
Permit audit findings and corrective actions.
Applied case
A welding permit was issued early. Before the team starts, the control room schedules a nearby product transfer connected to the drainage catchment.
Decision and explanation
Hold the work and have the issuer reassess the changed interface and physical precautions with operations.
Expiry is only one condition. A new operation can introduce vapour or change the risk while a document remains in date. Correct. The changed conditions need coordinated review and field confirmation before the job can be reauthorised. An extinguisher is a mitigation resource; it cannot establish isolation or prevent a dangerous atmosphere reaching ignition.
Look closer
Photorealistic illustrative UK workplace scene. Review checks: Setting: Replace a gasket on hydrocarbon pump P-204 at a fuel terminal during a planned shutdown. Act: Identify the exact flange and boundary, plan the isolation, drain and vent to a safe route, and check concurrent work. Hold / review: A passing valve, blocked drain, residual pressure, odour, gas alarm or conflicting operation suspends the job for reassessment.01 · What to notice
The permit board has a signed example. The proposed hot work lacks confirmed isolation, atmosphere assessment, adjacent-operation coordination and suitable fire precautions.
02 · Trace the harm
A permit records a controlled work decision; signatures cannot prevent ignition when the physical conditions and dependent tasks are unresolved.
03 · Test the safeguard
Inspect the real worksite with issuer and performing authority, reconcile every critical condition and stop trigger, and check post-work and handback requirements.
Lesson 14
Tank entry: avoid entry, then prove the complete safe system
Avoid entry first; necessary entry requires a complete verified safe system and effective rescue arrangements in advance.
Understand this fuel-industry task
The tank-entry marker represents inspection inside a largely enclosed tank where dangerous vapour, oxygen deficiency, unexpected inflow and the work itself may create specified risks. A tank can remain hazardous after cleaning: residue may release vapour, connected lines may admit material and an inert atmosphere may not support life. A clean-looking shell and last week's atmosphere result cannot establish today's entry conditions. The preferred learning question is whether inspection, cleaning or measurement can be completed safely from outside.
If entry is necessary, competent planning must cover the entire job and emergency arrangements before anyone enters. Identify and secure all relevant energy and material sources, prepare the tank, assess the atmosphere and maintain the necessary ventilation and monitoring. Match communication, access, equipment and people to the actual tank geometry and task. A retrieval arrangement that cannot bring an incapacitated person through the opening is not a workable rescue plan. Calling emergency services is essential when needed, but cannot substitute for adequate entry-specific rescue arrangements. An unprotected colleague must never become an improvised second entrant.
Trace the initiating event and loss of control
Step
Case pathway
initiatingEvent
An entrant relies on cleaning history while atmosphere controls or an isolation boundary are incomplete.
lossOfControl
The person is exposed to an unsafe atmosphere or unexpected inflow and cannot escape independently.
consequences
Fatal asphyxiation, poisoning, fire or drowning can follow; unplanned rescuers may be overcome in turn.
Choose effective safeguards
Control priority
Practical application
Eliminate
Avoid intrusive work on hazardous systems where possible.
Substitute
Use safer inspection and maintenance methods.
Engineer
Provide suitable isolation, contained drains and verified protective arrangements.
Manage
Coordinate permits, contractor competence, gas testing, rescue and handback as applicable.
Protect
Select suitable task PPE and rescue protection; never improvise a confined-space rescue.
Avoid the entry through an assessed external inspection or other suitable method wherever reasonably practicable.
If unavoidable, define the tank and job, verify effective isolation and preparation, and specify competent atmosphere assessment and ongoing controls.
Provide suitable trained people, communication, access, ventilation and equipment; stop and withdraw through the planned method when conditions change.
Establish an effective rescue system before entry, with competent rescuers and suitable equipment available for the actual opening and internal obstacles.
Verify effectiveness before release
Compare the entry pack with every actual connection and energy source, including temporary supplies and work on adjacent equipment.
Ask the responsible team to explain the atmosphere-test locations, equipment suitability, monitoring triggers and what happens after an interruption.
Use a safe rescue rehearsal or validated demonstration to show an incapacitated casualty can be recovered without exposing unprotected people.
Transfer the decision to the workplace
Interceptors, pits and enclosed plant compartments may raise similar questions; assess their actual enclosure, specified risks and task rather than relying on the asset name.
How should rescue capability be governed?
Rescue capability should be treated as a controlled emergency function with defined people, equipment, maintenance, drills and review.
Rescue arrangements need the same discipline as entry controls. The organisation should know which types of entry require non-entry retrieval, internal rescue, breathing apparatus, specialist contractors, external emergency-service liaison or a combination. The plan should match the geometry of the space and the likely condition of the casualty.
Drills should test credible scenarios, not comfortable ones. A good drill asks whether communications work, whether equipment reaches the casualty, whether the casualty can be extracted through the access point, whether rescuers are protected and whether the response time is realistic.
Rescue capability governance
Capability
Management requirement
Evidence
Rescue plan library
Plans matched to space classes and tasks.
Task-specific rescue sheets.
People
Competent rescue personnel available during entry.
Training, authorisation and rota.
Equipment
Suitable and maintained retrieval, BA/RPE, communication and first-response kit.
Inspection and maintenance records.
Drills
Scenario-based practice and debrief.
Drill report and action close-out.
External liaison
Emergency services know site access and hazards where relevant.
Emergency plan and liaison notes.
Review
Weaknesses escalated and resourced.
Management review and improvement plan.
Applied case
The inspection can be performed using a suitable external camera. A contractor nevertheless offers to enter because the tank was cleaned yesterday and the rescue team is an hour away.
Decision and explanation
Use the suitable external inspection method and avoid the unnecessary entry.
Correct. Avoiding the entry removes the exposure; a convenient entry is not justified by yesterday’s cleaning. A person at the opening does not establish isolation, safe atmosphere or a competent rescue capability. Symptoms may arrive too late for self-rescue. Emergency arrangements must exist before a necessary entry begins.
Look closer
Related SHAREhub topic guide. Apply its prompts to the specific case below.01 · What to notice
The tank-entry learning case proposes inspection after cleaning, but effective isolation, atmosphere evidence and a practised rescue method are not established.
02 · Trace the harm
Residual product, oxygen deficiency, toxic exposure or unexpected inflow can incapacitate an entrant; an improvised rescue can create more casualties.
03 · Test the safeguard
Check the complete current entry boundary and atmosphere controls and safely demonstrate that the planned rescue can work with the people and equipment available.
Lesson 15
The repair is finished: is the pump ready?
Integrate defect, permit, isolation and protective-system evidence into one restart decision.
Bring the evidence together
Follow a fictional pump from an abnormal trend through isolation, permit coordination and tank-entry planning to handback. Learn to challenge incomplete evidence, identify connected work and state when an operation must remain unavailable. Model missions let you practise the decisions; the course does not authorise live plant work.
Consider the whole sequence of interfaces and degraded conditions. A local repair or completed permit cannot by itself show that every required barrier is available. Separate immediate protection, the competent technical resolution, the operating decision and independent verification.
Use the worked cases and the linked primary guidance to identify the missing information. State who owns the decision and what would keep the operation restricted. The fictional scene illustrates choices; it cannot establish actual process limits, emergency capability or permission to operate.
How should handover and handback be controlled?
Isolation risk increases during shift change, job pause, recommissioning and return to service.
Handover must explain what is isolated, what is open, what is drained, what is blinded, which locks are applied, what work is incomplete and what must not be operated. Incoming teams should not inherit assumptions without walking down critical isolations where risk justifies it. See HSG253 - The safe isolation of plant and equipment.
Handback is not just removing locks. It includes checking tools and people are clear, blinds and spades are removed or correctly positioned, drains and vents are closed, guards are replaced, leak testing is complete, alarms/interlocks are restored, overrides are removed and the system is safe for restart.
Shift handover — Status misunderstood. Written and verbal isolation handover.
Job suspended — Plant condition changes while work paused. Suspend permit and secure open systems.
Reinstatement — Blind, lock or drain left in wrong state. Reinstatement checklist and second check.
Leak test — Containment not proven. Controlled pressurisation and inspection.
Restart — People or tools remain in danger zone. Final sweep and authorisation.
Applied case
P-204 has a new seal. The maintenance sheet is signed, but the trip cause is still unresolved, a manifold work party has not handed back its permit and the operating procedure references an obsolete valve. The loading supervisor asks for a single decision before the next tanker arrives. State the required safe plant condition, who must resolve each gap and what evidence would justify an authorised monitored restart. Do not use a reset as a diagnostic experiment.
Decision and explanation
Keep the pump unavailable until the cause, work-party handback, configuration and protection evidence are resolved through the authorised restart process.
Repair completion is one input. Restart also depends on the trip cause, connected work, correct configuration, usable procedure and effective protections.
Look closer
Photorealistic illustrative UK workplace scene. Review checks: Elimination: Use a suitable external diagnostic or redesigned sampling arrangement if it removes the intrusive valve task. Engineering controls: Use an approved closed sampling system or safely removable module to reduce open access. Engineering controls: Isolate connected feeds and energy sources, remove hazardous contents and verify the defined safe state before breaking containment. Engineering controls: Provide suitable drainage, containment, local extraction and support for the valve and connected pipework. Administrative controls: Agree the permit, contamination checks, line identification and hand-back with operations and contractors.01 · Read the situation
P-204 has a new seal. The maintenance sheet is signed, but the trip cause is still unresolved, a manifold work party has not handed back its permit and the operating procedure references an obsolete valve. The loading supervisor asks for a single decision before the next tanker arrives. State the required safe plant condition, who must resolve each gap and what evidence would justify an authorised monitored restart. Do not use a reset as a diagnostic experiment.
02 · Explain the principle
Follow a fictional pump from an abnormal trend through isolation, permit coordination and tank-entry planning to handback. Learn to challenge incomplete evidence, identify connected work and state when an operation must remain unavailable. Model missions let you practise the decisions; the course does not authorise live plant work.
03 · Carry it into work
Repair completion is one input. Restart also depends on the trip cause, connected work, correct configuration, usable procedure and effective protections.
Lesson 16
Turn the learning into a verified workplace action
Choose a bounded action, owner, task-specific evidence and escalation trigger within your role.
Make the course useful at work
Fuel-terminal operators, maintenance and permit teams; no instruction to perform live isolation, gas testing or entry without competence and site authority.
Choose one relevant task from this course. Describe the material, equipment, people and interfaces, then explain the credible harm pathway before choosing the control. Follow the actual site process for obtaining missing technical evidence and deciding whether work may proceed.
Use the linked resource with a specific question. A briefing should test understanding; a checklist should identify missing facts; an audit should compare claims with observations. Keep this exercise fictional and separate from live operating records.
Record the planned action, responsible role, verification evidence and a trigger for escalation or reassessment. Completion of this course is a learning record, not an operating instruction, engineering design, specialist licence or authority to enter, isolate or fight a fire.
What should the readiness gate prove before start-up?
The readiness gate should define the minimum evidence required before hazardous material, pressure, energy or operation is introduced.
A PSSR gate is strongest when it is binary on safety-critical requirements. If design is incomplete, isolations are unverified, alarms or interlocks are untested, operators are not trained, emergency arrangements are not ready or critical punch-list items remain open, start-up should not proceed. This is the point where production pressure must meet independent evidence.
For complex sites, each gate item should have an owner, acceptance standard, evidence reference and escalation rule. The gate should not ask whether a department is comfortable; it should ask whether the specific barrier, procedure, competence or document is ready for the start-up state.
Readiness gate
Pass standard
Evidence required
Design complete
Design intent, operating envelope, piping and instrumentation diagrams (P&IDs), relief basis, control philosophy and hazardous-area assumptions are current.
Approved design pack, current drawings and MOC/process safety information (PSI) update.
Procedures approved
Start-up, shutdown, abnormal, emergency, sampling, cleaning and handback procedures are in place and field-usable.
Controlled procedures with operations acceptance.
Isolations verified
Construction, maintenance and commissioning isolations are removed, reinstated or controlled; line-up is correct.
Isolation certificates, valve line-up, blinds/spades status and walkdown record.
Alarms/interlocks tested
Cause-and-effect, set points, trips, bypasses and operator response are demonstrated effective in the intended start-up configuration; any engineered equivalent is approved through change control, tested and available before hazardous introduction.
Functional test sheets, bypass register and alarm response briefing.
Training complete
Operators, maintainers, supervisors and emergency responders understand changes and safety-critical tasks.
Training records with understanding checks, not attendance only.
Punch list closed
Safety-critical punch items are closed; non-critical open items are risk-ranked and owned.
Punch-list register with risk rank, owner, due date and verifier.
Temporary controls approved
Temporary utilities, bypasses, scaffolds, hoses, drains or procedures are risk assessed, time-limited and visible.
Temporary control register, approvals and expiry/review date.
Authorised start-up
Competent authority accepts readiness conditions and any exceptions before hazardous introduction.
Signed start-up certificate with hold points and exception register.
What can be released, and on what evidence?
A signature, later test or extra operator vigilance cannot replace a credited function. An engineered equivalent needs approved change control, testing, documentation and availability before hazardous charge. Illustrative hold record: barrier HH-204; required independent shutdown proved in the intended configuration; test reference T-204 failed; operations hold solvent introduction; engineering owns repair/retest; operations releases only after competent acceptance and verified restoration. Staged energisation or inert commissioning may proceed only inside its separately assessed, authorised boundary; it does not release hazardous introduction or prove chemical compatibility.
Applied case
Your handover action says improve fuel safety but names no operation, control owner, evidence or stopping condition.
Decision and explanation
Name the actual task, competent control owner, evidence required and the hold/escalation trigger.
An effective plan connects a real task and control with a responsible role, objective evidence and a clear decision boundary. A course badge or signature does not establish the plant condition.
Look closer
Photorealistic illustrative UK workplace scene. Review checks: Setting: First solvent transfer into a modified vessel on a COMAH site after a shutdown. Act: Walk the line against the current drawing; prove isolation, relief route, level alarm and trip; check training and emergency response; classify every open item. Hold / review: No hazardous charge if a credited trip is untested, a bypass is open without approved risk control, the drain route is uncertain or a critical action remains unresolved.01 · Read the situation
Your handover action says improve fuel safety but names no operation, control owner, evidence or stopping condition.
02 · Explain the principle
Fuel-terminal operators, maintenance and permit teams; no instruction to perform live isolation, gas testing or entry without competence and site authority.
03 · Carry it into work
An effective plan connects a real task and control with a responsible role, objective evidence and a clear decision boundary. A course badge or signature does not establish the plant condition.