Skip to content
fatiguemanagement.co.uk

Fatigue Index vs Risk Index

HSE research report RR446 describes two related outputs from fatigue risk index (FRI) methods: the Fatigue Index (FI) and the Risk Index (RI).

Both are calculated from the same roster inputs — shift times, duration, rest between duties, breaks, and cumulative work history — but they answer different questions. They can diverge, especially around time-of-day effects.

Neither FI nor RI is a legal limit, a fitness-for-duty test, or proof that a shift is safe or unsafe. They are modelled, decision-support outputs that should be reviewed by competent persons alongside wider fatigue risk assessment and operational judgement.

Fatigue Index (FI) — modelled sleepiness likelihood

Section titled “Fatigue Index (FI) — modelled sleepiness likelihood”

The Fatigue Index is typically expressed on a scale from 0 to 100.

In the HSE RR446 framework, FI represents a modelled estimate of the probability that a worker would experience very high sleepiness during a duty period. “Very high sleepiness” is linked to levels 8–9 on the Karolinska Sleepiness Scale (KSS) — a validated subjective sleepiness scale.

Plain-English reading:

  • Higher FI → the model predicts greater likelihood of severe sleepiness during that shift
  • Lower FI → the model predicts lower likelihood of severe sleepiness for the pattern
  • FI describes acute fatigue likelihood for the duty — not whether a specific person feels tired today

FI does not measure real-time alertness. It does not predict whether a particular accident will occur.

Example framing (illustrative)

Section titled “Example framing (illustrative)”

If FI = 50 for a shift, the model represents roughly a 50% probability of very high sleepiness for that pattern — not certainty about any individual worker on any given day.

Risk Index (RI) — relative risk comparison

Section titled “Risk Index (RI) — relative risk comparison”

The Risk Index is typically expressed as a relative multiplier, with a reference schedule often normalised around 1.0.

In the HSE RR446 framework, RI compares modelled accident/incident risk for a duty or pattern against that reference schedule — commonly based on research involving 12-hour shifts in a rotating pattern in the rail sector.

Plain-English reading:

  • RI = 1.0 → broadly comparable to the reference pattern in the model
  • RI = 1.5 → model suggests roughly 50% higher relative risk than the reference
  • RI = 2.0 → model suggests roughly double the relative risk

RI is comparative. It is most useful when comparing roster options or identifying shifts that stand out within a schedule. It is not an absolute probability of an accident on a given day.

RI does not prove a shift is dangerous. It does not replace management judgement about task risk, worker competence, or controls in place.

Why FI and RI can diverge

Section titled “Why FI and RI can diverge”

Fatigue and accident risk do not always peak at the same time of day. HSE’s framework reflects this:

Aspect Fatigue Index (FI) Risk Index (RI)
Primary question How likely is very high sleepiness during the duty? How does modelled relative accident risk compare to a reference?
Scale Typically 0–100 (probability-like) Relative multiplier (reference ≈ 1.0)
Time-of-day sensitivity Strong — circadian sleepiness effects Strong — but risk curve differs from fatigue curve
Best used for Comparing sleepiness exposure across patterns Comparing relative risk across patterns
Common mistake Treating FI as fitness for duty Treating RI as accident prediction

HSE’s calculator guidance recommends reviewing both indices when assessing a pattern — and examining individual shifts, not only schedule averages.

Three components behind both scores

Section titled “Three components behind both scores”

FI and RI draw on the same underlying schedule structure described in RR446:

Component What it captures
Cumulative Fatigue build-up across consecutive shifts and insufficient recovery
Duty timing Start time, shift length, and circadian (time-of-day) effects
Job / breaks Workload, vigilance demand, and break patterns within the shift

A single shift is usually assessed in the context of the whole schedule — the cumulative component needs prior shifts and rest days to be meaningful.

Policy thresholds are not HSE mandates

Section titled “Policy thresholds are not HSE mandates”

Many organisations apply colour bands or trigger levels to FI and RI. These are organisational policy thresholds — not universal legal limits.

HSE states it does not currently specify threshold scores for fatigue. Organisations should aim to reduce scores so far as is reasonably practicable, seek feedback from staff on how tiring patterns feel, and not treat any historical survey benchmark as a ceiling on improvement.

Contractual fatigue standards on specific infrastructure or projects may define response levels — those apply to organisations subject to those requirements, not to all UK employers.

How to use FI and RI responsibly

Section titled “How to use FI and RI responsibly”

Sensible practice:

  1. Compare roster options at design stage — not only inspect one pattern in isolation
  2. Review shift-by-shift — averages can hide high-risk individual duties
  3. Read both FI and RI — they emphasise different aspects of exposure
  4. Document assumptions, scores, and who reviewed them
  5. Combine with worker feedback, supervision, and wider FRMS controls
  6. Reassess when actual hours, travel, or assignments change

See limitations of fatigue models for population averages, input quality, and individual variation.

What this page does not cover

Section titled “What this page does not cover”

This page explains FI and RI concepts from HSE RR446-based methods. It does not reproduce proprietary standard thresholds, endorse specific software, or state organisation-specific trigger levels. It does not address fitness for duty or legal working time limits.

Section titled “Related pages”

References

Section titled “References”