Seveso Inspection Checklist: What DCMR & NLA Actually Check

Why a generic safety checklist doesn't work at a Seveso site

Search "workplace safety checklist" and you'll find fire extinguisher tags, PPE sign-off sheets, and slip-and-trip walkthroughs. Useful for a warehouse. Not what gets checked when the Nederlandse Arbeidsinspectie (NLA) or DCMR shows up at a Seveso-classified site.

Seveso III (in the Netherlands, implemented through the BRZO 2015) doesn't regulate general workplace safety. It regulates the prevention of major accidents involving hazardous substances - fires, explosions, toxic releases - and the mitigation of their consequences if prevention fails. The inspection scope reflects that: regulators are checking whether your veiligheidsbeheersysteem (VBS) actually functions as a live management system, not whether someone's wearing safety glasses.

Below is a breakdown of the categories a Seveso VBS audit typically covers, and what tends to trip sites up when their readiness lives in binders and spreadsheets instead of a system that's current by default.

What regulators actually look at during a Seveso inspection

1. Major-hazard identification and risk assessment

• Is the list of hazardous substances and their classified quantities current and matched against Seveso thresholds?
• Do risk scenarios (fire, explosion, toxic release, domino effects) reflect the actual installations on site - not a study from three permit cycles ago?
• Are risk assessments linked to the specific safety measures meant to control them, with a clear owner for each?

2. Safety management system (VBS) documentation and function

• Does the VBS cover all required elements: organization and personnel, hazard identification, operational control, management of change, emergency planning, performance monitoring, and audit/review?
• Is there evidence the VBS is used, not just written - meeting records, action tracking, closed-loop follow-up?
• Can you show how a change on site (new equipment, altered process, modified quantity) triggered a management-of-change review?

3. Technical integrity of safety-critical equipment

• Are safety-critical systems (relief valves, gas detection, emergency shutdown, containment/bunding, fire and gas suppression) on a documented inspection and maintenance schedule?
• Is there a record trail showing inspections actually happened on schedule, not just a plan saying they should?
• Are deviations or overdue maintenance items visible and tracked to closure, rather than only surfacing during an incident review?

4. Emergency preparedness and response

• Is the internal emergency plan current, and does it match the actual hazard inventory and site layout?
• Have emergency drills been run and evaluated, with findings fed back into the plan?
• Is coordination with external emergency services (fire brigade, regional safety authority) documented and up to date?

5. Operational procedures and permit-to-work

• Are procedures for hazardous activities (hot work, confined space entry, work on pressurized systems) documented, followed, and verifiable?
• Does the permit-to-work system show sign-off, not just a template?
• Is there a clear line between what the procedure says and what inspectors observe happening on the floor?

6. Management of change

• Is there a formal process for reviewing safety impact before technical, organizational, or procedural changes go live?
• Can you produce the review trail for a specific change if asked - who assessed it, what was approved, what conditions were attached?

7. Training and competence

• Can you demonstrate that personnel operating or maintaining safety-critical systems are trained and currently competent, not just that a training existed at some point?
• Is refresher training tracked against a schedule, with gaps visible before they become findings?

8. Incident and near-miss follow-up

• Are incidents and near-misses logged, investigated, and linked to corrective actions?
• Are corrective actions tracked to closure with dates and owners, or do they stall after the initial report?
• Does the pattern of past findings show genuine follow-through, or the same issue recurring across inspection cycles?

9. Internal audit and management review

• Has the VBS itself been audited on the required cycle?
• Do audit findings feed into a documented management review, with evidence that leadership acted on them?

The real problem: most of this exists, just not where anyone can find it fast

Almost every Seveso site has answers to the questions above. The gap isn't usually that the safety measures don't exist - it's that the evidence is scattered across maintenance logs, shared drives, paper permits, and someone's inbox. When an inspection is announced, the week before becomes a scramble: pulling records together, chasing signatures, hoping nothing's overdue.

That scramble is itself a signal to a regulator. A VBS that only looks complete right before a visit isn't functioning as a management system - it's functioning as a filing exercise.

What changes with continuous digital tracking

A static checklist, whether on paper or in a PDF, captures a single point in time. It tells you what was true on the day someone filled it in. A Seveso VBS needs to be true every day, because the obligation isn't "pass the inspection" - it's "control the major-hazard risk continuously."

Running inspections, permits, maintenance checks, and corrective actions through a digital system changes the posture in a few concrete ways:

• Overdue items surface automatically, instead of being discovered when an inspector asks for the maintenance log.
• Corrective actions stay assigned and visible until closed, rather than living in an email thread that goes quiet.
• Management-of-change reviews leave a timestamped trail, so you can answer "who approved this and when" without searching.
• The audit trail is always current, because it's built from the inspections and checks people are doing anyway - not reconstructed afterward.

This is the core idea behind Capptions' Seveso Control: running the VBS as a live digital system so the site is always inspection-ready, rather than doing a fire drill every time DCMR or the Nederlandse Arbeidsinspectie is due to visit.

Using this as a working checklist

Treat the nine categories above as a starting structure for a self-assessment, not a substitute for your formal risk assessment or the specific conditions in your safety report. For each category, the useful question is the same: if an inspector asked for evidence right now, could you produce it in minutes, or would someone need to go looking for it?

If the honest answer is "go looking," that's the gap worth closing first - and it's usually a process gap, not a safety gap. The measures are there. The system for proving it, continuously, is what's missing.