The Science of Workplace Safety

Compliance Training Isn't Safety Training

Workers who passed the OSHA video still get hurt. Crews who memorized the procedure still make the same mistake. The gap between knowing a rule and acting safely under pressure is where incidents happen — and where a 30-minute compliance video has never been able to reach.

Most Workplace Incidents Aren't Rule Violations

They're decision-making failures, situational-awareness lapses, and team-communication breakdowns. None of those skills are taught by a slideshow, and none can be tested by a quiz.

80%
Human Factors

Up to 80% of commercial and military aviation mishaps are tied to human factors — perception, judgment, decision-making, and skill errors — not rule violations.

Department of Defense HFACS analysis
87%
Errors vs. Violations

In an analysis of 487 naval aviation Class A & B mishaps (1999–2009), errors accounted for 87% of unsafe acts. Rule violations accounted for only 13%.

U.S. Naval Safety Center
36%
Human Error in sUAS

RMIT University reviewed 150 civilian small-UAS mishaps (2006–2016). 36% were attributable to human error — even in an environment with strict rules and modern technology.

RMIT University, Australia

HFACS: How Mishaps Actually Happen

The Human Factors Analysis and Classification System was adopted by the U.S. Department of Defense in 2005 to reduce preventable aviation mishaps. It's now used across defense, healthcare, and high-consequence industry. Its core insight: incidents are rarely caused by a single unsafe act — they're produced by a cascade of latent organizational and supervisory failures that line up over time.

Layer 1 Latent · Slow-moving

Organizational Influences

Culture, resourcing, training-program quality, command climate. The slowest-moving and hardest-to-see contributors — yet they shape every decision below them.

Layer 2 Latent

Supervision & Leadership

Inadequate supervision, planned inappropriate operations, failure to correct known problems, supervisory violations. Where organizational culture meets the work itself.

Layer 3 Active & Latent

Preconditions for Unsafe Acts

Environmental conditions, technological limits, mental and physiological state of the operator, team-coordination breakdowns. The conditions present when the work begins.

Layer 4 Active · The visible event

Unsafe Acts

Skill-based errors, decision errors, perceptual errors, and known violations. This is the layer compliance training tries to address — and the only one a post-incident report usually captures.

The implication for training: If your training only addresses Layer 4 — the rules a worker must follow — you're targeting roughly 13% of the problem. The other 87% lives in decision-making under pressure, environmental awareness, team communication, and the cultural permission to stop work when something looks wrong. Those are skills you have to drill, not memorize.

1994 Black Hawk Shootdown: Twenty-Six Failures, One Mishap

On April 14, 1994, two U.S. Army UH-60 Black Hawks were shot down by two U.S. Air Force F-15 fighters over northern Iraq, killing all 26 people aboard. Investigators eventually identified more than two dozen contributing organizational, supervisory, and procedural failures stretching back years. None of them, individually, would have caused the shootdown.

Harvard organizational behavior professor Scott A. Snook spent years dissecting the incident and published Friendly Fire: The Accidental Shootdown of U.S. Black Hawks over Northern Iraq (Princeton University Press, 2000). His causal map traces an unbroken chain from Cold War-era budget cuts through everyday operational drift to the moment of the shootdown. A representative slice of that chain:
01
Organizational Influences

Shrinking post-Cold War defense budgets reduced joint-training opportunities between the Air Force and Army units that would later operate together over northern Iraq.

02
Organizational Influences

The mission (Operation Provide Comfort) was assembled as an ad hoc task force. Helicopters and fixed-wing units lived apart, with no integrated flight operations procedures.

03
Supervision & Leadership

The governing operations order (OPORD 97-1) was never updated to reflect how missions were actually being flown. Helicopters routinely entered the no-fly zone before the fighter sweep — outside the documented procedure.

04
Supervision & Leadership

The AWACS crew supervising the airspace was new to the mission, undermanned, and the mission commander was not technically "mission ready" — a known supervisory deviation.

05
Preconditions

The F-15 pilots had limited low-altitude experience and weak visual-recognition training on the specific helicopter types they would encounter. The Black Hawks were also carrying external fuel tanks — a non-standard silhouette.

06
Preconditions

The Black Hawks and the F-15s were on different radio frequencies. AWACS controllers did not direct them to a common frequency. IFF (friend-or-foe identification) failed.

07
Preconditions

When the F-15 lead called for radar contact, the AWACS response was "clean there" — incorrect. The F-15s proceeded to a visual identification pass under pilot anxiety, in mountainous terrain, expecting no friendly aircraft in the area.

08
Unsafe Act

The F-15 lead misidentified the Black Hawks as Iraqi Mi-24 "Hind" attack helicopters. Both U.S. helicopters were destroyed.

26

contributing failures identified — across all four HFACS layers. Compliance training, taken alone, would have caught at most one of them.

Mishaps are often attributed to operator error or equipment failure without recognition of the systemic factors that made such errors or failures inevitable. — U.S. Military UAV Mishap Assessment (Tvaryanas, Thompson & Constable)

What VR Training Reaches That Video Can't

A compliance video targets one layer of HFACS: the rule. A Humulo VR simulation places the worker inside a scenario where decision errors, perceptual errors, environmental complications, and team-communication breakdowns can all be drilled in the same five minutes — safely, repeatedly, with telemetry. Below is how our existing OSHA-aligned modules map across the framework.

Forklift Operations

OSHA 1910.178

Worker practices recognizing pedestrian-in-blind-spot scenarios, decides whether to proceed, communicates intent, and stops work when conditions warrant. Drills perceptual, decision, and skill layers — not just the rules.

Preconditions Skill Errors Decision Errors

Lockout/Tagout (LOTO)

OSHA 1910.147

Crews work through energy-source identification, verification, and the social pressure of an impatient supervisor or production deadline — the supervisory and precondition factors that drive most real LOTO incidents.

Supervisory Preconditions Decision Errors

HAZWOPER

OSHA 1910.120

Simulates dynamic chemical-release scenarios where the right response depends on situational awareness, PPE judgment, and decision-making under time pressure — not memorized procedures.

Preconditions Decision Errors Team Comms

Electrical Safety

OSHA 1910.333

Includes arc-flash and downed-powerline scenarios that are physically impossible to recreate in a live training environment. VR allows repeated exposure with full debrief — the only modality where this is true.

Perceptual Errors Decision Errors Skill Errors

Fire Extinguisher Safety

OSHA 1910.157

Trainees discriminate fire class, choose an extinguisher, and decide whether to fight or evacuate — under stress, with smoke, and with consequences. The decision-error layer that classroom video cannot reach.

Decision Errors Perceptual Errors Skill Errors

Slips, Trips & Falls

OSHA 1926.502

Hazard-recognition reps drill the perceptual layer directly: workers learn to spot the unsafe condition before someone gets hurt, not after. Verified at the University of Houston in a peer-reviewed ASEE study.

Perceptual Errors Preconditions Hazard Recognition
Coming Soon

Your Near-Misses Should Be Making Your Training Smarter

Most safety programs collect incident reports, file them, and forget them. The lesson never makes it back to the next shift's training.

The Humulo Safety Insights Portal will let your EHS team log incidents and near-misses, classify them against the HFACS framework, and map them directly to the VR scenarios where each gap can be drilled. We close the loop between what happened on the floor and what your workers practice next quarter.

Anonymized, aggregated insights — the patterns we see across hundreds of customers — feed back to your team so you know which hazards are trending in your industry before they reach you.

1

Log the incident

Near-miss, first-aid event, or OSHA-recordable. Date, facility, severity, narrative.

2

Classify the cause

HFACS layer (organizational, supervisory, precondition, unsafe act) with guided prompts — not a free-text guess.

3

Map to training

The portal surfaces which Humulo VR scenarios target this exact failure mode — and which workers haven't yet completed them.

4

Measure the close

Quarterly Safety Insights report shows incident rate against training completion — and benchmarks against anonymized peer data.

Contributor

[Contributor Name & Bio Pending Approval]

The systematic-safety framework presented on this page draws on the published work of safety-and-learning experts including the contributed piece "Big Sky, Many Small UASs" and the academic analysis of organizational accidents by Scott A. Snook (Harvard Business School).

⚠ REVIEW REQUIRED: This section currently has no named contributor. Per discussion, we want Joe Schweitzer (USNA '89, USMC NFO, aviation mishap survivor, published safety/learning author) named here as a contributing expert — but only after he reviews and approves the page in full and confirms how he'd like to be credited. Until then, this block stays anonymous. Do not publish a named version without his written sign-off on the framing of his role and his bio.

Move Beyond Compliance. Build a Safety Culture.

See how Humulo VR drills the decision-making and situational awareness that compliance video can't reach. 20-minute demo with our safety team.

Schedule DEMO

Landing page form
Humulo Virtual Reality - VR safety training icon