Custom app development for emergency plans and safety rounds at a Seveso site
The internal emergency plan of a Seveso establishment is usually in order. What is rarely there is proof that it has been practised and that the measures beneath it work. That proof is created on the installation: during a round, during an exercise, at the moment someone notices something. And there is no keyboard there.
What changed in 2024
The Major Accidents Risks Decree 2015 (Besluit risico's zware ongevallen 2015) lapsed on 1 January 2024 and was absorbed into the Environment and Planning Act (Omgevingswet). What used to be called a Brzo company is now a Seveso establishment, and the substantive rules are in section 4.2 of the Environmental Activities Decree (Besluit activiteiten leefomgeving). The name changed, the weight did not: a major accident prevention policy, a safety management system, an internal emergency plan and, for upper-tier establishments, a safety report.
Supervision in this category is unusually heavy and unusually shared. The competent authority, the safety region and the Netherlands Labour Authority (Nederlandse Arbeidsinspectie) look on together, each from its own angle, and they visit the site. What they ask is not whether the plan exists, but whether it is alive.
That is where it goes wrong. A safety management system that exists only in documents is hard to defend during an inspection. Rounds are walked and ticked off on paper, exercises are held and summarised in a report, deviations are reported and followed up by email. Everything happens; it just cannot be found when someone asks about it three months later.
Two related topics have their own page. For the substance inventory, the threshold test and the management system, there is software for Seveso obligations; for releasing work on an installation, there is the app for work permits and securing equipment.
What happens to the findings after an inspection is a separate process: multiple supervisory authorities, their own deadlines and a public summary. See software for Seveso inspections.
How we build this
The observation on the installation is the unit here. If it is captured at the moment itself, the exercise reports and the inspection answers follow from it.
We look at how things currently work, what gets written down and where coverage drops out. In a process installation, that happens in more places than you would expect.
You have a safety management system with established procedures. We build its field side rather than a new framework alongside it, because two frameworks are worse than no app.
The app works without a connection and synchronises later. In tank pits, cellars and behind steel, that is not an edge case but the normal situation.
We ask a question an inspector would ask and check whether the answer is on the table within a minute. What is missing then will be missing later too.
What the app actually does
The app does little, and that little must always work, with gloves on and in an environment where a phone is sometimes not even allowed. Which components you need depends on your installation and your threshold category.
Rounds along installation sections
For each round, which points are due and what to look out for. The app guides the round and records what was seen, so it is clear afterwards which section was inspected and when.
Provable presence at each point
A tag or code at the installation proves that someone was actually there. Without it, a ticked-off round is just a claim, and that is exactly what a regulator tests.
Report deviations immediately, with follow-up
A leak, a blocked escape route, a missing shut-off valve. The report is assigned an owner and a deadline on the spot, so it doesn't sit unread in a mailbox.
Drills recorded with lessons learned
When was the drill held, who took part, what went wrong and what was done about it. In an inspection, your internal emergency plan is worth only as much as the drill record behind it.
Photo with time and location
Images of the actual situation during a deviation or a drill. In a subsequent investigation, this is the most convincing piece of evidence and the quickest to capture.
Working fully offline
Rounds, reports and photos carry on without a connection and sync later, with each round showing whether the sync succeeded.
Who we build for
The nature of the installation determines what the app looks like. Four situations.
Chemicals and process industry
Sprawling installations with zones where equipment must meet strict requirements. Here the round is the main means of observation and the place where an app delivers the most.
Storage of hazardous substances
Tank storage and transfer terminals deal with changing parties and many visitors. Besides the round, what matters here is who was on site and for what reason.
Maintenance and contractors on site
Work permits, final checks and handover after a job. These moments fall between parties and are therefore the worst documented. For the inspection of the work equipment itself, there is the inspection log.
Production with large cooling installations
Ammonia brings companies into this category that don't see themselves as chemical businesses. The obligations are the same, while the organisation isn't set up for them. We cover the risk assessment underneath in Risk assessment tool with action plan.
Test your idea first: a working prototype in 1 day
With OneDayBuild, we make your idea tangible in a single day for €1,150, so you know whether further development is worth the investment. Decide to go ahead with the full build? Then we deduct the cost in full.
View OneDayBuild →Technology and integrations
The app is the field side of your safety management system, not a new framework. What it records ends up in the file where your procedures and emergency plan already sit.
Why Appfront
Regulators test whether the plan is alive
Three regulators are watching, and they do visit. That is why we build the trail of rounds and drills: it is what they ask for and what paper cannot provide.
There is no signal at the installation
Tank pits, basements and steel structures. Offline is the starting point; an app that stalls there won't be used, and then the observation simply doesn't exist.
We don't build a second framework
You already have a safety management system with established procedures. The app is its field side, not a methodology of its own.
Connecting to what you already manage
Installation data and maintenance planning already exist. We connect to them through integrations rather than retyping them.
Security and privacy
This app contains the weak points of an installation where major accidents can happen: which substance is where, which shut-off valve doesn't work, which escape route was blocked. That is more sensitive than an ordinary inspection file. We limit what is stored offline on the device to the current round, tie permissions to the user and to the installation component, and make sure a lost device can be disconnected remotely.
There is also something here that does not apply to other apps: in some zones, an ordinary device is not allowed. That is not a software question, but it does shape the design, because the app also has to work on the equipment that is permitted there and has to cope with part of a round being entered later. On the evidence side, the photo, timestamp and observer must be tamper-proof; a drill report that has been edited afterwards proves nothing. How we handle security ourselves is set out in our information security policy; reports from outside go through our coordinated vulnerability disclosure policy.
Frequently asked questions about Seveso and the safety app
Seveso. The Brzo 2015 lapsed on 1 January 2024 and was absorbed into the Environment and Planning Act (Omgevingswet); since then the regulations refer to a Seveso establishment, and the substantive rules are in section 4.2 of the Activities (Living Environment) Decree (Besluit activiteiten leefomgeving). The term Brzo still appears in older documents and in everyday speech, but in documents sent to the regulator you should use the new term.
The core obligations remain: a major accident prevention policy, a safety management system, an internal emergency plan and, for upper-tier establishments, a safety report. What did shift is its place in the legal framework, and with it the link to permits and the competent authority. Have your own adviser check what that means for your site.
The system describes what needs to happen; the app records that it has happened. That difference is what a regulator tests. Rounds ticked off on paper do not show that anyone actually visited the plant section, and a drill without learning points and follow-up is, at inspection, a report with no follow-through.
That depends on the zone. In some areas an ordinary device is not allowed and you work with equipment that is suitable for it. We design the app so that it runs on that equipment and so that part of a round can be entered later without the record losing its value. Which zone requirements apply to you is for your safety specialist to determine.
By shifting the question from searching to showing. A regulator asks when a plant section was last checked, or what was done about a deviation from last quarter. If that is in a single overview with the time, the person and the follow-up, the conversation is short. Otherwise a search through mailboxes begins.
Lower-tier establishments do not need a safety report, but the prevention policy, the management system and the internal emergency plan still apply. In practice the app is lighter here, because there are fewer plant sections and fewer rounds. What stays the same is that the evidence is created on site.
Often yes, and that is usually sensible because it involves the same people and the same plant sections. Maintenance planning and plant data come from your existing software via integrations; the app adds the recording instead of a second administration.
That depends on the number of plant sections and rounds, whether drills are included and which software it needs to integrate with. The rounds with recording are usually quick to get working; drills and integrations cost more. We give a substantiated estimate after the discovery phase.
Make rounds and drills demonstrable?
Ask when a random plant section was last checked and see how long it takes to get the answer. That is the question a regulator asks. We build this as a standalone app and as part of a broader project, custom app or custom software.