The worst incidents I’ve heard about happen during non-routine work — maintenance windows, shutdowns, contractor jobs. It’s where hazards shift by the hour and the usual controls don’t fully apply.
A permit-to-work system is how you keep that work under control.
This guide is for operations and compliance managers and the field EHS leads who own safety on the ground. I’ll walk you through what a permit to work is, the main permit types, everything that goes on the permit, and the full process from request to closure. Then I’ll cover the roles involved, why permit to work systems fail in real conditions, and how a digital permit to work process fixes the gaps that paper leaves behind.
Contents:
What Is a Permit to Work?
A permit to work is a formal document that authorizes a specific high-risk or non-routine task. It records what work is approved, where, by whom, under what controls, and for how long.
The UK Health and Safety Executive (HSE) describes it as “a more formal system stating exactly what work is to be done and when, and which parts are safe.” A permit doesn’t make the work safe on its own. As HSE puts it, a permit to work helps bring the risk assessment “to life at the sharp end” — it’s a communication and control tool in the first place.
But a permit to work on its own is just a form. What makes it effective is the system built around it.
A permit-to-work system is the governance structure that makes that document functional: the defined roles and responsibilities, the permit types for different hazard categories, the training requirements for authorized personnel, the audit process, and the communication protocols between shifts and crews.
There is one more definition worth knowing in this regard, and it’s a permit-to-work process.
A permit-to-work process is how that system runs in practice: the sequence that moves a task from initial request through hazard assessment, authorization, active monitoring, suspension if conditions change, hand-back, and formal closure.
Without a disciplined process, the system produces approvals that don’t reflect what’s actually happening on site. Without a system, the process has no governance behind it. And without the permit, neither has a record.
But when used together, all three ensure that every high-risk task has a named owner, verified controls, and a complete audit trail.
| What it is | What it does | Who owns it | When it’s created | |
|---|---|---|---|---|
| Permit to work | The document | Authorizes a specific task — records scope, controls, named parties, and duration | Permit authoriser / issuing authority | For each individual task, before work starts |
| Permit-to-work system | The governance structure | Defines who can issue, authorize, and close permits; sets permit types, training requirements, and audit protocols | Safety or operations management | Established at programme level, reviewed periodically |
| Permit-to-work process | The operational sequence | Moves a task from request through assessment, authorization, monitoring, hand-back, and closure | All parties — originator through issuer to performing authority | Defined at system level, followed at task level |
Types of Permit to Work
Different jobs carry different hazards, so the types of permit to work map to the specific risk being controlled. Here are the ones you’ll issue most often.
Hot Work Permit — Covers welding, cutting, grinding, and brazing. Controls spark, fire, and explosion risk, and usually triggers a fire watch and a check for nearby combustibles.
Download a ready-made hot work permit template →
Confined Space Permit — Covers entry into tanks, pits, vessels, and reactors. Controls atmosphere, oxygen levels, toxic gas, and restricted access. Typically requires gas testing before and during entry.
Electrical / Energized-Work Permit — Covers work on panels and live circuits. Controls shock and arc-flash risk, and is tightly linked to isolation and lockout tagout.
Work-at-Height Permit — Covers scaffolds, roofs, towers, and elevated platforms. Controls fall risk and confirms access equipment, anchor points, and rescue plans.
Energy Isolation / Lockout Tagout (LOTO) — Covers de-energizing machinery before servicing. Controls stored and live energy so equipment can’t start while someone’s hands are inside it.
Download a ready lockout tagout procedure checklist →
Excavation and Cold Work — Excavation permits control collapse and buried-service strikes during trenching. Cold work permits cover lower-energy tasks near hazardous areas where a formal check is still warranted.
The point of having different permit types is to match the permit to the actual hazard so the controls on the form reflect the job in front of you.
What Goes on a Permit to Work
Every permit to work should capture these elements:
- Scope and Duration — What work is authorized, exactly where, and for how long. A permit with a vague scope is a permit that will drift into work nobody approved.
- Hazard Identification and Risk Assessment — The specific hazards for this task and the assessment behind them. This ties the permit to the risk assessment rather than restating a generic list.
- Control Measures — The controls that make the work safe: required PPE, isolation and lockout tagout points, gas testing, and a fire watch for hot work.
- Competency and Role Assignments — Who authorizes, who performs, and who supervises. Named people, not job titles alone.
- Communication and Coordination — Which nearby teams need to know, so simultaneous operations don’t create a hazard the crew can’t see.
- Authorization Signatures — Sign-off from the competent, authorized issuer confirming the controls are verified and in place.
- Closure and Hand-Back — Confirmation that the work is done, the area is safe, and normal operations can resume, with the record retained for audit.
Get these elements right and the permit does its job: it puts the same facts in front of everyone who touches the work.
The Permit-to-Work Process, Step by Step
The permit to work process is a lifecycle. Each step exists to catch a specific failure, and skipping one is usually where things go wrong. Here’s the full permit to work procedure, synthesized from HSE HSG250 guidance.
Step 1: Request and initiation. The performing party describes the task, location, and duration and submits it for review. State the work plainly, including equipment and access needs. Flag any obvious hazards up front so the reviewer isn’t surprised.
Step 2: Risk assessment and method. The team identifies hazards and defines the controls, usually alongside a job safety analysis (JSA). Assess the actual task, not a template from a similar job. Confirm the method statement matches the controls on the permit.
Step 3: Authorization and issue. A competent, authorized issuer verifies the controls and signs off. This is where separation of duties matters most — the issuer should be independent from the person doing the work. No sign-off until every listed control is confirmed in place.
There is a principle behind this step that is easy to state and easy to forget under time pressure. Liam Hook, HSE Manager at Marubeni Middle East and Africa Power Limited, describes a moment from his early career that makes it concrete:
“We went around and locked everything off and were doing the checklist. Then my manager asked: ‘Okay, Liam, are you happy? Everything is safe?’ I said yes. He goes: ‘Before you sign the permit — touch it.’ Half an hour later I’m still standing there. He explained it to me: you said it’s safe for somebody else to touch it. If you won’t touch it, you don’t believe you’ve done your job.”
— Liam Hook, HSE Manager, Marubeni Middle East and Africa Power Limited (17 years across power, oil and gas, construction, and aviation)
The authorization signature should reflect that level of certainty. Signing a permit because the form is complete is different from signing because the controls are verified.
Step 4: Display and communication. The permit becomes visible at the worksite and affected teams are informed. Post the permit where the crew is working, not in an office folder. Communicate to nearby crews so simultaneous operations don’t collide.
Step 5: Execution and monitoring. Work proceeds under the stated controls, with supervision continuing throughout. Supervisors check that the controls on paper match the conditions on site. Any deviation stops the work until it’s reviewed.
Step 6: Suspension. If conditions change — weather, a shifting atmosphere, an emergency — the permit is paused or revoked. Suspend immediately when a control is no longer valid. Don’t let a crew “finish quickly” against a permit that no longer holds.
Step 7: Hand-back. The performing party confirms the work is complete and the area is safe. Return the equipment and workspace to a known, safe state. Report anything left incomplete before handing back.
Step 8: Closure and record retention. The issuer inspects the area, restores normal operations, and formally closes the permit. The record is retained for audit. Close the permit deliberately — an open permit is an open liability.
The step that trips up most teams is isolation communication between issue and hand-back. When an isolation or hazard doesn’t get communicated at handover, the permit that looked complete becomes the incident report.
Success Story Success Story How Dyna Crane Services reduced audit time and saved costs with automated safety workflows
Who’s Involved: Roles and Responsibilities
A permit to work system depends on people knowing their role and staying in it. HSG250 defines a clear set of roles built around separation of duties.
Originator — Requests the permit and describes the work to be done.
Permit Authoriser / Issuing Authority — Verifies the controls and authorizes the permit. This is the accountable sign-off.
Area Authority — Owns the location where the work happens and confirms it’s safe to release for the task.
Performing Authority / Permit User — Carries out the work under the permit’s controls.
Supervisory Personnel — Oversee execution and confirm controls hold throughout the job.
In practice these map to real titles: a maintenance planner as originator, a shift supervisor or authorized person as issuer, an area or asset owner as area authority, and the technician or contractor crew as the performing authority.
The core principle is simple. The person who authorizes the permit should be independent from the person doing the work. When one person fills multiple roles without checks, you lose the second set of eyes that the whole system is built to provide.
RELATED RELATED Effective inspection scheduling: what to do, when to do it, and how to stay on track
Why Permit-to-Work Systems Fail (and How to Fix It)
Here are the failure modes, drawn from HSE human-factors guidance and industry practice, with the fix for each.
Weak ownership and accountability
The problem: it’s unclear who reviews, authorizes, supervises, and closes the permit. The consequence: permits sit half-approved, or get closed by whoever’s nearest the clipboard. The fix: assign named roles per HSG250 and make separation of duties non-negotiable.
Poor communication around isolations
The problem: an isolation or hazard isn’t clearly communicated at handover between shifts or crews. The consequence: this is the classic cause of permit-related incidents — someone works against an isolation they didn’t know had changed. The fix: make isolation status explicit on the permit and confirm it verbally at every handover.
User resistance and “bureaucracy”
This is the failure mode that rarely gets named directly, because it sounds like a people problem when it’s a process design problem.
“In oil and gas or the power industry, you need your permit, you need your method statement, you need your risk assessment, your toolbox talk record, your start card. When you’ve got a pack this thick before you do any work, that takes a very long time. I personally believe it’s very bureaucratic. I’ve been on site where they were changing the internals of a valve — a very simple job — but the bureaucracy of the project required a method statement submitted to the project company for review. That took two weeks. They found the damage in the morning and needed to change out the valves that evening.”
— Liam Hook, HSE Manager, Marubeni Middle East and Africa Power Limited
When the permit process feels disproportionate to the task, crews work around it. The fix is to use task-specific permits tied to real hazards, so the form earns its place on every job it’s required for.
Generic, copy-paste permits
The problem: permits get reused across jobs and stop reflecting the actual task. The consequence: the controls on the form don’t match the hazards on site. The fix: build permits from the current risk assessment and JSA, and audit a sample regularly to catch drift.
Paper limitations
The problem: forms get lost, go illegible, or fall out of date, with no real-time visibility. The consequence: slow approvals, no live status, and painful audit prep. The fix: move the process to a digital permit to work system so status, approvals, and records stay current.
Modernizing Permit to Work: From Paper to Digital
Once you see where paper breaks, the case for an electronic permit to work becomes practical.
A digital permit to work process gives you routed approvals, so nobody chases physical signatures across a site. You get real-time visibility of active permits and their locations, which makes conflict and simultaneous-operations detection possible. Records are centralised and auditable, so compliance prep stops being a scramble.
You also get automated reminders for expiry and overdue permits, instant suspension or revocation when conditions change, and built-in checklists that reduce skipped steps. Because the process connects to your JSA and lockout tagout controls, the whole flow stays in one place.
On the adoption question — the one that determines whether a digital system actually gets used or becomes shelfware — John Dunne, Group Head of HSSE at Red Sea Global, is direct:
“With technology, it’s about testing and proving the concept is good for the business and will make it safer. If you rush it, it fails completely. I’m really surprised when technology companies don’t carry out a proof of concept as a matter of course. I always start off really small — a small group of people, a small task, not too complicated, well controlled, with a start time and an end point. Then you review and see if it’s a fit for the business.”
— John Dunne, Group Head of HSSE, Red Sea Global
That principle applies directly to digital permit to work rollouts. Start with one permit type, one crew, one site. Prove it works there before you push it to 90 sites.
Here is how Fluix, inspection management software for field teams, approaches permit to work software:
- Mobile-first and offline-capable — crews request, complete, and close permits from any location, even with no signal
- Rich field capture — photos, GPS geotags, and files attach directly to the permit
- Multi-step approvals — routed sign-offs with real-time notifications
- Corrective actions — assign and track follow-ups, even to people outside Fluix
- Full audit trail — timestamps, photos, notes, e-signatures, user permissions, and version control
- Connected to inspections — the same platform runs your safety inspection software and ties permits into your broader safety management programme
Watch this hot work permit demo to see how it work on practice.
Bringing It Together
A well-run permit to work system comes down to three things: clarity about what work is authorized, accountability for who owns each step, and a clean record you can produce on demand.
Paper can deliver the first two on a good day. It rarely delivers the third, and it tends to break exactly where the field meets the paperwork — at handover, at isolation communication, and when a slow process pushes crews to work around it.
A mobile-first, offline-capable digital process closes those gaps. It captures the request, the hazards and photos, the approvals, the live status, and the closure in one place, connected to your inspections and corrective actions.
Frequently Asked Questions
- What is a permit to work in safety?
A permit to work is a formal, documented authorization that controls high-risk or non-routine work. It confirms hazards are identified, controls are approved and in place, and the right people know about the work before it starts.
2. What are the main types of permit to work?
The most common types are hot work, confined space, electrical or energized work, work at height, energy isolation / lockout tagout, and excavation. Each controls a specific hazard tied to the task.
3. Does OSHA require a permit to work?
OSHA doesn’t use one blanket “permit to work” rule, but permit principles run through its standards for permit-required confined spaces, control of hazardous energy (lockout/tagout), and hot work. Check the relevant OSHA topic pages for the work you’re doing.
4. Permit to work vs. risk assessment vs. JSA — what’s the difference?
The risk assessment and job safety analysis identify and evaluate hazards. The permit to work authorizes the job and confirms those controls are in place before work starts. The permit relies on the assessment; it doesn’t replace it.
5. Paper vs. digital permit to work — which is better?
Paper works until it doesn’t: forms get lost, approvals stall, and audits are painful. A digital permit to work adds routed approvals, real-time status, reminders, and an audit-ready record — which is why most teams with distributed operations move to electronic permits.
6. Who issues and authorizes a permit to work?
A competent, authorized issuer — the permit authoriser or issuing authority under HSG250 — verifies the controls and signs off. Under separation of duties, that person should be independent from the crew performing the work.
