Do We Really Need Another HAZID?

The question I keep getting

It usually comes from someone in operations, and it usually comes politely.

"We've done this job twice. We've done the HAZID twice. Why are we doing it again?"

I've been asked it more times than I can count, and I understand why. Pre-shutdown, everything is urgent. There is a high volume of work, the schedule is tight, and a HAZID is one more meeting in a week full of them.

My answer has always been the same: the urgency is exactly why we strategise before we do. When the pressure's on, the temptation is to either roll the old one over or run the full workshop on autopilot. Both feel like progress. Neither tells you whether this job is actually the same job.

What I've been wrestling with

The room is right more often than I'd like to admit. A third run of the same workshop, with the same precedent and mostly the same people, can produce a document that looks thorough and changes nothing. The facilitator reads the nodes, everyone nods, the ratings don't move. I've felt a room go flat on a third pass, and I've felt myself go flat with it.

I don't want to be the HSE guy who insists on paperwork for its own sake. The people in that room are experienced. Many of them have done this job more times than I have. If I can't explain why the time is worth it, I'm spending their goodwill — and I'll need that goodwill when something genuinely matters.

But I also can't un-know what I've seen go wrong. The same work order isn't the same job. The crew is different. The contractors are different. The plant is two years older. Something that ran alone last time now runs beside three other jobs. Reissuing last year's HAZID as if none of that changed isn't efficient. It's a guess with a signature on it.

The hardest one: the actions from last time. More often than I'd like, the previous HAZID raised actions and nobody in the room can say for certain they were closed out — or whether the controls actually worked when the job ran. That one keeps me up. If they weren't, then reusing the HAZID doesn't reuse a risk assessment. It reuses a promise nobody kept.

Urgency pushes toward the two worst options. Under pressure, it's tempting to either roll the old one over or run the full workshop on autopilot. I get why "can we just roll it over?" sounds reasonable. Sometimes it is — but you only know that if you've stopped to check.

What the law actually says

When I went back to the legislation, it helped more than I expected — and it took both sides of the argument seriously.

• It doesn't demand a fresh assessment every time. The model Code of Practice How to manage work health and safety risks says that where hazards are well known with established controls, you may simply implement those controls.
• It allows reuse, on conditions. A risk assessment may be reused where all the hazards, tasks, things, workers or circumstances are the same, and only if no one will be exposed to greater, additional or different risks. If anything at the workplace changes, a new assessment should be done.
• It sets clear review triggers. Regulation 38 of the model WHS Regulations requires a review when a control isn't effective, before a change likely to create a new or different risk, when a new hazard is identified, when consultation says so, or when an HSR asks. Incidents count — even ones where nobody was hurt.
• It expects the workers to be involved. Consultation is required at every step.

Reading that helped me reframe the conversation with ops. The law isn't asking us to repeat the work. It's asking us to show we checked whether anything changed. That's a very different meeting.

Strategise before you do

This is the step I'd defend hardest, especially when things are urgent. Before anyone books a room, a short strategy session — thirty minutes to an hour, with the area owner, the planner and HSE — works through five questions:

1. What's changed? Scope, plant condition, crew and contractors, SIMOPS, season and shift pattern, and any incidents since the last run — ours or the industry's.
2. What happened last time? Were the actions closed out? Did the controls work when the job actually ran? Who'd know?
3. What can kill someone on this job? Those risks get fresh eyes regardless of what the precedent says.
4. So what do we actually need? Three honest options. Confirm: nothing material has changed and last time's actions are closed; a short walk-through of the precedent and a sign-off, around thirty minutes. Targeted revalidation: some things have changed; start from the precedent and spend the time on the changes and the fatal risks, an hour or two. Full HAZID: the scope, method or plant has changed significantly, or last time's controls didn't hold.
5. Who needs to be in the room? The people who'll do the work this time — not just the people who planned it last time.

Whatever format comes out of that, the workshop itself starts from the precedent rather than a blank page, and finishes with clear traceability: what we carried forward, what we changed, and why.

Strategising first is what makes the urgency workable. Ops get an answer to why are we doing this before they walk in, the room's time goes where it's needed, and the session is shorter than the first one and more useful than the second. More importantly, it respects the experience in the room. Ops aren't being asked to redo their work. They're being asked to tell us what's different — which is exactly what they know better than anyone.

What I'm still not sure about

I don't have all of this solved.

• I'm still working out how to make the "were last time's actions actually closed?" question land without it feeling like an audit of the people in the room. It's a structural gap, not a personal one — but it can feel personal when you're the one asked.
• I'm still working out where the line sits for a fully unchanged, low-risk job. There probably are jobs where a short confirmation is genuinely enough, and I don't want to drag people through a process just to protect myself.
• I'm still learning how to read a room that's gone quiet on a familiar job — whether that's confidence, or fatigue, or someone who's noticed something and doesn't want to be the one to slow things down.

Closing

The question was never really whether to do a third HAZID. It was whether we'd stop long enough to find out what kind we needed.

A precedent is a starting point, not a substitute. The experience of the people who built it is worth respecting, and so is the experience of the people who'll do the work this time. Strategising first is how you honour both: you reuse what's genuinely unchanged, you put your time into what isn't, and you never sign off on a promise from last time that nobody can show was kept.

When the pressure's on, that half hour before the workshop is the most valuable time in the whole process. It's the difference between a room that's going through the motions and a room that's looking for what's changed.

The job may be the same job on paper. It's our responsibility to prove it still is.

First published in The Field Note on LinkedIn.

Read and comment on LinkedIn