Bow-Tie Explained: 4 Parts That Keep Barriers Visible

Bow-Tie analysis is a visual risk model that maps threats on one side of a top event and consequences on the other, with barriers in the middle. It matters when a team needs to show not only what can go wrong, but also which controls should stop the event and which should limit the damage if it still happens.

Bow-Tie analysis is a way to map a hazardous path from threats to a top event and then from that event to consequences, while placing preventive and mitigative barriers on the diagram. It is useful when leaders need a control model that names ownership, shows barrier weakness, and makes the gap between theory and field reality visible.

Across 25+ years leading EHS at multinationals, Andreza Araújo has seen Bow-Tie work only when the team can name the top event in the language of the shift, not the language of the consultant. In Make The Difference: Be a Leader in Health & Safety, she treats leadership as the work of making risk visible enough to act on. That is also why this article points to the field, not just the diagram.

If you want the procedure side, How to Run a Critical-Control Walkdown in 8 Steps shows how barrier checks move from paper into the work area. For the loss side, Residual Risk Explained: 4 Gaps Leaders Must Separate helps separate what remains after controls from what was never controlled at all.

Definition

Bow-Tie sits between a simple risk register and a deeper process-analysis method. It does not only say that a hazard exists. It explains the route from threat to loss, the point where control fails, and the barriers that should keep the event from spreading into consequences.

That is why Andreza Araújo keeps using it in practical conversations with leaders. A diagram only becomes useful when the owner of each barrier can answer one question without hesitation: is this barrier preventive, or is it meant to limit damage after the top event has already happened.

Threats

Conditions, triggers, or failures that can push work toward loss of control.

Top event

The point where control is lost and the hazard is no longer fully contained.

Preventive barrier

A control that should stop the threat from reaching the top event.

Mitigative barrier

A control that should reduce harm once the top event has already occurred.

Barrier owner

The role that must test, maintain, and challenge the barrier in real work.

4 parts

Threats

The left side of the diagram starts with the threats. In operations, those are the conditions that make the work vulnerable, such as missing isolation, degraded guarding, poor housekeeping, or a permit that no one really reads. A good Bow-Tie makes those threats concrete enough that the shift supervisor can recognize them in the field.

Top event

The top event is the pivot. It is the point where containment is lost, but before the full consequence has played out. That distinction matters because many teams skip over it and jump straight to the injury or the fatality. If the model cannot name the top event clearly, the rest of the diagram becomes decoration.

Preventive barriers

Preventive barriers sit before the top event. They should stop the threat from turning into a loss of control. In practice, that means isolation, interlocks, verified permit conditions, competent supervision, or design choices that make the unsafe path harder to reach. HAZOP vs FMEA vs Bow-Tie: Which Method Fits a Shutdown Team? is useful here because it shows when Bow-Tie is the right lens and when another method should do the upstream work first.

Mitigative barriers

Mitigative barriers sit after the top event. They do not pretend to prevent the loss once it has started. Their job is to shrink the harm, contain the release, and buy time for people to respond. Emergency isolation, rescue readiness, suppression systems, and clear escalation paths belong here. James Reason's latent failure logic fits this part well, because the visible event often sits on top of weaker conditions that were never challenged in time.

How Bow-Tie differs from a risk register

A risk register is broad. It is good for portfolio visibility, ownership, and periodic review. Bow-Tie is narrower and more operational. It asks which barrier stands between the threat and the top event, who owns that barrier, and how the team will know it still works on a Tuesday afternoon, not only during audit week.

That difference is why a register and a Bow-Tie should not be treated as substitutes. Andreza Araújo has seen more than 250 cultural-transformation projects produce the same mistake. A company fills the register, closes the meeting, and assumes the risk has been managed. The diagram is useful only when it changes what supervision checks in the field.

For the field-verification layer, How to Run a Critical-Control Walkdown in 8 Steps shows the next move after the diagram is built. If you want the failure-side view first, HAZOP vs FMEA vs Bow-Tie: Which Method Fits a Shutdown Team? gives the comparison leaders usually need before choosing a method.

When should leaders use Bow-Tie?

Use Bow-Tie when the team needs to make barrier logic explicit. It is especially useful in high-consequence work, where the issue is not only whether a hazard exists, but whether the barrier chain is strong enough to prevent a serious outcome or at least limit it.

Bow-Tie is less useful when the organization only wants a broad list of hazards for compliance filing. In that case, a risk register can do the administrative job faster. The mistake is to ask the register to behave like a barrier model, or to ask the Bow-Tie to replace basic risk governance. Each tool has its own job.

Andreza Araújo's view is straightforward. A model matters only when the supervisor, EHS lead, and plant leader can all read the same picture and agree on what must be checked before the work starts. In Make The Difference: Be a Leader in Health & Safety, that is what leadership looks like in practice, because risk language has to become operating discipline.

What should you do next?

If your site already has a Bow-Tie on paper, test it against the field. Ask whether each barrier is real, whether one person owns it, and whether the team can verify it without hiding behind the form. If the answer is no, the diagram is describing a process that does not yet exist in practice.

If your team is still choosing the right tool, start with the problem you need to solve. Use a register for ownership and review, Bow-Tie for barrier logic, and a walkdown for live verification. That sequence keeps the work honest and prevents the diagram from becoming a wall poster.