How 250+ Projects Moved Fatigue Signals Into Work-Design Decisions

Across more than 250 cultural transformation projects supported by Andreza Araujo, fatigue often appeared first as a personal report and only later as an operational pattern. A worker said the crew was exhausted. A supervisor saw handover mistakes rising after nights. HR saw absence. EHS saw near misses in the last hours of the shift. Operations saw overtime as the cost of demand, not as a control weakness.

The case lesson is uncomfortable because it does not allow leaders to hide behind awareness campaigns. Fatigue becomes a safety control issue when the organization can name the work pattern that removes recovery and then change the decision that keeps reproducing it.

Andreza Araujo's work in Muito Alem do Zero argues that fragile mental health makes physical safety fragile because attention, judgment, and response quality degrade before a visible incident appears. In fatigue risk, that idea becomes practical: the body may be the place where the signal shows up, but the exposure is often designed into workload, timing, staffing, handover, and production recovery. The same logic applies after a serious event, when a critical incident check-in has to turn care into temporary work controls.

Key Takeaways

• Fatigue is not only a wellness concern when it affects driving, maintenance, chemical transfer, line clearance, confined space entry, monitoring, or other critical tasks.
• The signal should trigger a decision on work design, not only a reminder to rest better.
• Useful fatigue governance joins EHS, HR, and operations because each function sees only part of the exposure.
• The strongest metric is decision conversion: how many fatigue signals changed the roster, workload, staffing, task sequence, or recovery window.

Initial scenario

The starting point in many companies is a fatigue program that looks active. There is a toolbox talk, a support line, a sleep hygiene poster, a self-report form, and a supervisor instruction to watch for tiredness. These tools may help, but they do not prove that the organization controls fatigue.

The failure sits in ownership. EHS treats fatigue as a safety behavior issue. HR treats it as health and attendance. Operations treats it as a scheduling constraint. Finance treats overtime as a cost line. Each view contains a piece of truth, although none of them can protect the worker alone.

In the project pattern, the first weak signal was usually not dramatic. It was a small cluster of handover defects, a near miss after repeated night work, a maintenance error after callouts, a driver who reported micro-sleep, or a supervisor who quietly swapped people away from critical tasks because the planned crew was not fit for the work.

The trap was the same across industries. Leaders asked whether people had been trained to report fatigue, but the harder question was whether a fatigue report could change staffing, task timing, overtime approval, shift rotation, or recovery conditions before the next exposure.

Decision

The decisive move was to treat fatigue as an operating decision, not as a private weakness. A worker can be responsible for reporting reduced readiness, but the company remains responsible for the work conditions that repeatedly make readiness unlikely.

That distinction changed the leadership question. Instead of asking whether the person slept enough, leaders asked where the work system was spending recovery faster than people could rebuild it. Was overtime covering a permanent staffing gap? Were the heaviest tasks placed after the lowest-alertness hours? Were handovers compressed because production wanted a fast restart? Was a commute risk hidden behind the formal end of the shift?

The WHO guidelines on mental health at work, published in 2022, separate organizational interventions from individual support. That matters here because a support route may help someone recover, while the prevention decision may require redesigning work before people reach the point of needing recovery.

James Reason's view of latent failures also applies. The tired operator at the end of the chain is visible, but the conditions that shaped the exposure often sit earlier in planning, staffing, maintenance windows, contractor timing, and leadership tolerance for extended work.

Execution

The execution began by mapping fatigue signals to work-design decisions. A self-report was no longer treated as an isolated health note. It was coded against the operating condition behind it: night work, overtime, callout, understaffing, high cognitive load, poor handover, long commute, compressed recovery, or repeated critical tasks near the end of the shift.

That map forced leaders to see patterns they had previously treated as separate events. Three tiredness reports in one week might look like individual weakness when reviewed separately. The same three reports, placed beside overtime records, late maintenance releases, and near misses after midnight, reveal a design problem whose owner is not the worker.

The second execution move was a threshold for critical tasks. If fatigue touched driving, mobile equipment, energized work, line breaking, confined space entry, process monitoring, lifting, patient care, or security response, the signal had to reach a decision owner before the task continued under the same conditions.

The third move was field verification. A schedule change, staffing decision, or task resequencing could not close the item by email. Supervisors had to verify whether the crew actually received a usable recovery window, whether overtime dependency dropped, and whether the new sequence protected the highest-risk tasks from the lowest-alertness period.

Measured result

The measured result should be stated with discipline. This case does not claim a universal percentage reduction from one fatigue routine. The defensible result across the 250+ project pattern is a change in decision quality: fatigue signals moved from individual reports into visible decisions about work design.

That matters because fatigue metrics can become another passive dashboard. If leaders count reports but do not change conditions, the organization may become better at documenting exposure while leaving the same people inside it. In The Illusion of Compliance, Andreza Araujo warns that a system can look complete while its operated risk remains untouched, and fatigue programs are especially vulnerable to that illusion.

The better metric is conversion. How many fatigue signals changed overtime approval, shift rotation, handover overlap, staffing, critical task timing, supervision, or recovery verification? How many signals returned because the first decision did not change the exposure? How many critical tasks were stopped, delayed, or reassigned because readiness was uncertain?

Generalizable lessons

The first lesson is that fatigue needs shared ownership but not blurred ownership. EHS should define the exposure, HR should protect health support and privacy, and operations should own the roster, staffing, task sequence, and production choices that create or reduce fatigue.

The second lesson is that the program must protect medical privacy while still controlling work. A supervisor does not need a diagnosis to act. The supervisor needs to know whether a safety-critical task is being assigned after repeated nights, forced overtime, reduced overlap, or a recovery window that exists only in the policy.

The third lesson is that fatigue indicators must sit beside operating indicators. Overtime hours, callouts, late restarts, handover defects, near misses by shift hour, absence, and task-criticality reviews tell a stronger story together than any single self-report channel can tell alone.

The fourth lesson is that support tools cannot substitute for prevention. Peer support, counseling routes, and manager training may be valuable, but they become ethically weak when the organization asks people to cope with a work pattern that leadership refuses to redesign.

What to apply in your operation

Start with one map. Take the last ten fatigue-related signals, including reports, near misses, handover defects, overtime spikes, absence patterns, and supervisor observations. Place each one beside the work-design condition that may have created it.

Then define the decision threshold. Any signal connected to a safety-critical task should require a named decision before continuation under the same conditions. The decision may be to stop, delay, reassign, add supervision, change sequence, protect rest, or approve extra staffing.

Finally, create a weekly fatigue decision review with EHS, HR, and operations. Keep it short enough to survive production pressure. Review the signal, the work condition, the requested decision, the owner, and the verification date. If the meeting cannot name a work-design decision, it is probably reviewing fatigue as sentiment rather than exposure.

Connect this routine with the 30-day fatigue risk management plan, the psychosocial review before a shift schedule change, and the comparison of work redesign, manager training, and peer support. Those articles cover the practical build, while this case explains the leadership decision pattern behind it.

FAQ

What is a fatigue signal at work? A fatigue signal is evidence that reduced recovery, long hours, night work, sustained mental load, or poor task timing may be affecting safety-critical performance. It can appear through reports, near misses, handover errors, overtime patterns, absence, or supervisor observations.

Why should fatigue be treated as work design? Fatigue should be treated as work design because many exposures are created by rosters, workload, staffing, overtime, commute pressure, task sequence, and recovery windows. Support tools matter, but they cannot compensate for work that repeatedly removes recovery.

What did Andreza Araujo observe across 250+ projects? Across more than 250 cultural transformation projects, Andreza Araujo has observed that fatigue data becomes useful when leaders connect it to decisions on timing, staffing, supervision, task readiness, and field verification.

Which leaders should own fatigue decisions? EHS can structure the risk view, HR can protect health and support routes, and operations must own the work-design decisions. The issue fails when any one function treats fatigue as someone else's topic.

How do leaders know a fatigue program worked? A fatigue program worked when it changed the work condition that created exposure. Evidence includes adjusted rosters, protected rest windows, redesigned handovers, removed overtime dependency, changed critical task timing, or verified recovery before safety-critical work.

Conclusion

Fatigue signals protect people only when they change the work that is draining recovery. A report without a work-design decision may prove that the organization heard the concern, but it does not prove that the next exposed worker is safer.

If your leadership team wants to turn mental health and fatigue signals into operating decisions, use Headline Podcast as a prompt for the next EHS, HR, and operations review at Headline Podcast.