Stop-Work Authority Explained: 4 Boundaries for Supervisors

F7 glossary explainer for frontline supervisors, operations leaders, and EHS managers

Stop-work authority is often presented as a slogan, yet supervisors need a decision boundary that tells them when a pause is necessary and what evidence permits a restart. The issue is not whether someone feels uncomfortable. The issue is whether a critical control can still be trusted.

Stop-work authority is the defined right and responsibility to pause a task when a hazard is uncontrolled, a critical control is uncertain, or the work has changed beyond its approved conditions. It protects the person who raises the concern and gives the supervisor a clear route from interruption to verification, correction, and controlled restart.

Definition

A useful stop-work process separates three actions that are often blurred together. A worker or supervisor identifies a concern, the responsible leader contains the exposure, and the control owner verifies whether the task can resume. That sequence matters because a pause without verification becomes theater, while a restart without containment turns a speaking-up policy into a liability.

James Reason's work on latent conditions helps explain why the first visible concern is not necessarily the whole failure. A missing barrier may reflect weak planning, unclear ownership, or a design condition whose effect only becomes obvious at the workface. Andreza Araujo makes the same leadership point in Make The Difference: Be a Leader in Health & Safety, where accountability is connected to the decisions that shape exposure rather than assigned only to the person closest to the hazard.

4 Boundaries that define a legitimate stop

1. The hazard boundary

The first boundary is physical. Stop the task when people, equipment, energy, or materials can enter an exposure path that the approved plan does not control. Examples include unexpected movement, an unverified isolation, a changed line-up, a missing guard, or weather that changes the stability of the work.

A concern becomes actionable when the supervisor can identify what could harm someone, where the exposure exists, and which control should prevent it. If the answer remains vague, the task still needs clarification before work continues.

2. The control boundary

The second boundary concerns confidence in the barrier itself. A permit, checklist, or training record is not proof that a control is working. The supervisor should pause when the control is absent, bypassed, damaged, incompatible with the task, or supported only by an assumption that nobody has checked.

This is where Andreza Araujo's approach of combining engineering, creativity, and care becomes practical. Engineering identifies the barrier, creativity tests what could defeat it, and care ensures that the person who reports the weakness is not treated as the problem.

3. The change boundary

Work can become unsafe even when the original plan was sound. Stop-work authority applies when the task, crew, equipment, sequence, environment, or interface changes in a way that invalidates the prior risk decision. A replacement component, a late contractor, a new simultaneous operation, or a rushed handover may each create a new control question.

The relevant question is not whether the change looks small. It is whether the change affects the assumptions on which the control depended. A minor deviation can matter when it occurs at the exact point where the plan had no margin.

4. The restart boundary

The fourth boundary protects the restart decision. Work should remain paused until the responsible person can show what changed, which control was restored, and how the team knows the control will hold during the next step. The evidence may be a field verification, a tested isolation, a revised permit, a corrected method, or a briefing whose questions confirm shared understanding.

A restart is not approved because the schedule is tight or because the team has already lost time. It is approved when the exposure is controlled and the remaining uncertainty has an owner, a due date, and an interim measure.

How to differentiate stop-work authority in practice

When should a supervisor use it?

Use stop-work authority whenever the supervisor cannot explain which control protects the next action, who owns that control, and what evidence confirms its condition. That test is simple enough for a pre-job brief, yet demanding enough to expose paper compliance.

Across 25+ years of executive EHS experience, Andreza Araujo has treated safety leadership as a decision discipline, not a motivational message. Her work with multinational operations, including the PepsiCo South America plan that reduced the accident ratio by 50% in six months, reinforces why a pause must lead to a better operating decision rather than a ceremonial restart.

Stop-work authority works when leaders make the boundary visible before the task begins, protect the person who raises the concern, and require evidence before production resumes. Safety is about coming home, and that principle becomes operational when the organization knows exactly when work must stop.

Explore more safety leadership guidance from Andreza Araujo.