Decision Fatigue: 7 Supervisor Checks Before Risk Rises

OSHA's worker-fatigue guidance identifies impaired decision making, reduced alertness, and lower concentration as direct safety effects of long hours and irregular shifts. This article gives supervisors seven checks to redesign the moment when a tired team is most likely to accept the easiest answer instead of the safest one.

Why decision fatigue turns safe behavior into default behavior

Decision fatigue in safety is the gradual loss of judgment quality after a supervisor, operator, or EHS professional has made too many choices under pressure. It matters because the dangerous decision is rarely dramatic; more often, it is the quiet acceptance of a weak pre-task assessment, an incomplete isolation check, or a rushed handover because the shift is nearly over.

The common response is to tell people to pay more attention, yet that answer misses the operational design problem. Across 25+ years leading EHS at multinationals, Andreza Araújo has observed that exhausted teams do not usually abandon safety values; they simplify decisions until the path of least resistance looks acceptable.

For a supervisor, the practical goal is not to become mentally invincible. The goal is to reduce unnecessary choices, move critical decisions earlier in the work cycle, and create triggers that force a pause when fatigue, production pressure, or routine work begins to distort risk perception. The same control logic applies one level up in leader mental health, where decision load can quietly become a safety exposure.

1. Map the decisions that repeat every shift

Decision fatigue starts when the same person must make dozens of low-value choices before the high-risk task even begins. In a maintenance shift, the supervisor may approve manpower changes, sign permits, answer schedule conflicts, review a pre-task risk assessment, solve a tool shortage, and still be expected to notice a subtle change in the job environment. For deeper context, see behavioral Observation.

What most programs miss is that repetition does not always create mastery. Repetition can also create automatic acceptance, especially when every form looks familiar and every team says the task is routine. As Andreza Araújo argues in Safety Culture: From Theory to Practice, culture is visible in repeated habits, which means a tired approval habit is already a cultural signal.

Supervisors should list the ten decisions they make most often during a normal shift and separate them into three groups: decisions that can be standardized, decisions that require judgment, and decisions that must be escalated. This simple map removes noise before the critical work starts because it prevents every request from arriving as if it deserved the same mental energy.

2. Move critical approvals away from the end of the shift

Critical approvals should not be concentrated in the last hour, because cognitive load rises precisely when the team is trying to close work, clean the area, and protect production continuity. OSHA names long hours and irregular shifts as fatigue hazards, and its guidance links fatigue with reduced alertness, memory problems, and impaired decision making.

The trap is administrative convenience. A supervisor who signs everything at the end of the day may believe the process is efficient, although the timing makes the review weaker. In more than 250 cultural-transformation projects supported by Andreza Araújo's team, the pattern appears often: the company has a rule, but the calendar teaches people to apply it when attention is lowest.

Move high-consequence approvals to a fixed window before the task begins, not after the work has gathered momentum. If the task involves energy isolation, work at height, confined space, hot work, or simultaneous operations, the approval window should happen while the supervisor can still walk the area and challenge assumptions without fighting the closing pressure of the shift.

3. Treat shortcuts as workload signals before treating them as attitude problems

A shortcut is often a visible behavior with an invisible workload history behind it. When a team bypasses a checklist, accepts a vague answer, or signs a permit without walking the area, the easy explanation is attitude, but the better first question is whether the system made the correct decision too hard to repeat. For deeper context, see conformity Pressure.

This is where decision fatigue connects directly with normalization of deviance. The first shortcut may be noticed, the second may be debated, and the seventh may become the way the job is done because nothing bad happened yesterday. James Reason's work on latent conditions helps explain why the visible act is only one layer of the event.

The supervisor should ask three workload questions before concluding that the worker does not care: how many approvals happened in the last two hours, how many changes were introduced after the briefing, and which barrier required the most effort to maintain. Those answers do not excuse the shortcut, but they identify which part of the workflow must be redesigned.

4. Use a stop rule for late changes

A late change is any modification introduced after the team has already accepted the risk plan, including a different worker, different tool, altered sequence, missing spare part, weather change, delayed isolation, or added simultaneous task. Late changes are dangerous because the original risk assessment no longer describes the work as executed. For deeper context, see line of Fire Safety.

The weak safety culture response is to ask the supervisor to judge every late change from scratch, even when the supervisor is tired and production is waiting. The stronger response is a stop rule. If the change affects energy, height, confined space, lifting, hot work, or line of fire, work pauses until the risk assessment is reopened.

During her PepsiCo South America tenure, where the accident ratio fell 50% in six months, Andreza Araújo learned that a rule only works when it changes the work rhythm. A stop rule protects the supervisor from negotiating safety one tired exception at a time.

5. Convert behavioral observation into dialogue, not surveillance

Behavioral observation becomes useful when it reveals why a person made a decision, not merely whether the person complied. A tired operator may choose a familiar path, skip a verification step, or avoid raising a doubt because the observation system rewards visible compliance more than honest conversation.

The Vamos a Hablar? methodology proposes observation through dialogue, and that matters for decision fatigue because fatigue hides inside explanations. A checklist may record that the worker used the guard, while a conversation may reveal that the worker nearly removed it because the setup takes too long during changeover.

Use observation questions that expose decision pressure: what made this step harder today, what changed after the briefing, what would you stop if you had full authority, and what part of the task feels routine but still deserves attention. These questions connect behavioral observation with BBS failures that supervisors can fix, without turning the process into blame.

6. Protect the first unsafe doubt

The first unsafe doubt is the earliest moment when someone senses that the plan no longer fits the task. It may sound small, such as a mechanic saying the isolation point does not look like the drawing, or a forklift driver saying the aisle feels tighter than usual.

Decision fatigue makes supervisors vulnerable to dismissing weak signals because weak signals require mental work. The supervisor must investigate, compare, decide, communicate, and possibly delay the job, which is exactly why the tired brain prefers reassurance. Amy Edmondson's work on psychological safety helps explain why teams need permission to voice uncertainty before certainty is available.

The practical response is to make the first doubt operationally cheap. A supervisor can say, "If someone raises a mismatch, we pause for five minutes before debating whether it is real." That rule protects risk perception in routine work because it treats uncertainty as useful data, not as resistance.

7. Audit the decisions that incidents did not expose

Decision fatigue cannot be managed only through incident investigation because many weak decisions do not end in injury. The absence of harm may reflect luck, barrier redundancy, or someone else's last-minute correction, which means the organization must review decision quality before the lagging indicator moves. For deeper context, see normalization of Deviance.

In her Portuguese title Muito Além do Zero, "Far Beyond Zero", Andreza Araújo critiques the illusion that zero accidents proves a healthy safety system. That critique is essential here because decision fatigue may be growing while TRIR remains quiet.

Audit five samples each week: one permit approved late in the shift, one task with a late change, one observation with a weak conversation, one near miss that almost stayed hidden, and one job where production pressure changed the sequence. The question is not whether the supervisor was right or wrong; the question is whether the workflow made the better decision easier to choose.

Each week without a decision-fatigue review allows the operation to convert tired judgment into routine practice, while the safety dashboard may still report stable lagging indicators.

Comparison: decision fatigue controls vs awareness reminders

Conclusion

Decision fatigue is not solved by asking supervisors to care more, because the real safety gain comes from designing the shift so the better decision is easier to make when fatigue, routine, and pressure are already present.

If your operation needs to diagnose whether safe behavior is being supported by the workflow or undermined by decision load, ACS Global Ventures can help connect cultural diagnosis, supervisor routines, and practical controls through Andreza Araújo's safety culture work.