Shiftwork Fatigue: What Mining Research Still Hasn’t Tested

Mining and heavy industry have spent decades measuring shiftwork fatigue. They still have remarkably little controlled evidence showing which countermeasures reduce it on the job.

The uncomfortable number comes from a 2022 systematic review by Dugdale et al.: researchers screened 1,651 papers published between 1980 and 2020, then retained seven after quality appraisal. That is a 0.4% survival rate. None of the qualifying studies took place in mining. Four tested bright light, one tested blue-light-blocking glasses, and two tested sleep hygiene or alertness-management training.

More conspicuously, the review found no controlled industrial studies of napping, caffeine timing, roster design, or fatigue-detection technology. Those tools often appear in Fatigue Risk Management Systems, the policies used to manage schedules, rest, and alertness in high-consequence work. They may help. The point is narrower and more important: controlled trials have not established that they help industrial workers in the settings where the stakes are highest.

A field that describes fatigue better than it tests solutions

The seven-study result does not mean industrial fatigue research is tiny. It means much of it could not answer an intervention question. Observational studies can show that fatigue tracks with shift patterns, sleep loss, or work conditions. They cannot reliably establish whether a change in lighting, schedule, break policy, or wearable system caused a better outcome.

Dugdale and colleagues excluded much of the literature because studies lacked randomization or control conditions, relied on designs that were observational only, or used outcomes that did not match prespecified alertness and performance criteria. That leaves a major evidence gap, not proof that untested practices fail.

It also leaves a generalizability problem. The reviewed industrial participants were predominantly male and came largely from manufacturing, oil and gas, and transportation. The review identified no controlled mining intervention trial and no age- or sex-stratified analyses of intervention response. A worker underground, on a remote fly-in/fly-out roster, or operating heavy equipment is therefore often applying guidance extrapolated from another occupation.

Light has the best case, with no mining-ready recipe

Timed light exposure has the most controlled support among occupational fatigue interventions. A 2023 randomized controlled trial by Cyr et al. tested an evening-light intervention during real night shifts and found significant reductions in subjective fatigue and errors. Its operational setting gives the result more practical weight than a laboratory-only finding, although it did not study mining workers or establish a protocol tailored to extraction sites.

A 2024 randomized trial by Scott et al. adds a complementary result. During simulated night-shift work, circadian-informed overhead lighting accelerated circadian realignment and improved vigilance, sleep, and subjective sleepiness compared with standard lighting. The study supports the biological rationale for timing and spectrum-aware workplace lighting. Simulation is not a mine, though: underground and outdoor operations have different ambient light, task demands, and constraints.

A systematic review of evening-light interventions also found alertness benefits across delivery methods. But the protocols differed in intensity, wavelength, timing, and duration. The evidence supports light as a category; it does not yet identify one validated industrial specification. Anyone selling a universal lux-and-minutes formula is moving faster than the research.

Daytime light may also matter, but the evidence is farther from mining. In a 2026 experiment by de Zeeuw et al., 20 healthy adults were randomized to five mornings of low-intensity incandescent light or higher-intensity fluorescent light. The higher-light condition improved depression markers and corrected dysregulated sleep patterns. This finding raises a reasonable question about prolonged low-light work environments, but it does not measure mine lighting or establish an intervention for miners.

Blue-blocking glasses have mixed results

Blue-light management gets more confidence online than it earns in this literature. A controlled trial by Martin et al. found that in-car dim blue light combined with morning blue-light-blocking glasses did not improve circadian adaptation in fast-rotating shift workers. The null result could reflect a timing mismatch for rapidly rotating schedules rather than a universal failure of blue-light management, but the study does not support treating that protocol as established practice.

Two broader reviews point in the same cautious direction. A 2021 systematic review found evidence for blue-blocking glasses too mixed to support uniform recommendations for sleep and mood disorders. A 2025 meta-analysis of randomized crossover trials found modest improvements in actigraphic sleep outcomes. Actigraphy measures movement-derived sleep patterns; it is not a direct measure of fatigue, alertness, operational errors, or safety. None of that evidence establishes that blue-blocking glasses improve industrial performance.

Sleep timing and melatonin require narrower claims

Behavioral sleep timing has a better controlled result than many industrial programs acknowledge, but in a specific clinical population. A 2022 crossover clinical trial by Cheng et al. enrolled 60 real-world night-shift workers with diagnosed shift work disorder. The sleep-timing intervention improved attention and sleep quality. That is useful evidence for workers with the disorder; it does not prove the same approach produces the same effect in a general mining workforce.

Melatonin also resists one-size-fits-all advice. A controlled trial by Marqueze et al. found that exogenous melatonin reduced circadian misalignment and body weight among shift workers, with effects concentrated in early chronotypes. The result argues for individual assessment, not blanket supplementation. It was a controlled trial rather than a randomized trial, and it did not validate melatonin for mining operations or show uniform benefits across chronotypes.

Roster design and monitoring remain plausible, not proven

Roster design may create tradeoffs that a simple "better schedule" label hides. A 2026 naturalistic study of offshore platform workers found that rotating schedules intended to reduce night-shift health effects also disrupted circadian adaptation. Offshore oil and gas shares remote deployment and extended rotations with some mining operations, but the study was not a controlled optimization trial. It documents a tradeoff; it does not identify the best roster.

Alertness prediction has an interesting, early operational signal. In ICU nurses on rotating shifts, Knock et al. validated a physiologically based model in which schedule data alone predicted alertness about as accurately as models that also used self-reported sleep. That could make prospective schedule modeling practical without invasive monitoring. Yet the validation occurred in healthcare with documented light conditions, not in mines or outdoor extraction settings. Prediction is not proof that changing a roster based on a prediction improves safety.

The same restraint applies to naps, caffeine protocols, fatigue cameras, and wearable alerts. Their absence from qualifying controlled industrial trials does not make them useless. It means a company should avoid presenting them as sector-validated solutions. Track local outcomes, account for individual variation, and treat results as operational learning rather than settled science.

The verdict: For mining and heavy industry, structured shift-timed light management has the closest controlled evidence base, though no validated mining protocol exists. Sleep timing and melatonin have narrower evidence in selected worker groups. Blue-light blocking remains mixed. Naps, caffeine timing, roster optimization, and fatigue-detection technology still need the controlled industrial trials that would show whether they reduce fatigue or errors where they are actually used.