ALARP Principle: Meaning, ALARP Triangle and How to Demonstrate It
ALARP stands for As Low As Reasonably Practicable. A risk is ALARP when every measure that could reduce it further has been implemented, unless its cost in money, time or trouble would be grossly disproportionate to the safety benefit. The principle comes from UK law (the Health and Safety at Work etc. Act 1974 and Edwards v National Coal Board, 1949). It is used worldwide to decide whether a risk has been reduced enough, and the duty holder, not the regulator, must show that it has.
The Question Every Risk Assessment Must Answer
Every risk assessment, at its core, must answer one question: is this risk acceptable?
But "acceptable" is not a fixed standard. A risk that is acceptable in one context — sky diving, motor racing — is not acceptable in another. And in the workplace, "acceptable" has a legal definition that many safety professionals do not fully understand.
In the UK and many other jurisdictions, the legal standard for risk control in the workplace is not "eliminate all risk" — it is ALARP: As Low As Reasonably Practicable.
ALARP acknowledges that zero risk is not achievable, and that society does not benefit from demanding an infinite level of resource expenditure to eliminate residual risk. What it requires is that every reasonably practicable measure that can reduce the risk has been implemented — and that the cost of any further reduction is grossly disproportionate to the safety benefit it would achieve.
This is a precisely specified legal standard. Misunderstanding it creates real risk — of over-investment in low-priority controls, and more dangerously, of under-investment in areas where ALARP has not been demonstrated.
What Does ALARP Mean?
ALARP stands for As Low As Reasonably Practicable. It is the criterion used to determine whether a risk has been reduced sufficiently — or whether additional control measures are required.
The definition of "reasonably practicable" comes from UK case law. In Edwards v National Coal Board [1949], the Court of Appeal established that "reasonably practicable" means:
"Reasonably practicable" is a narrower term than "physically possible" and seems to me to imply that a computation must be made by the owner, in which the quantum of risk is placed on one scale and the sacrifice involved in the measures necessary for averting the risk (whether in money, time or trouble) is placed in the other; and that if it be shown that there is a gross disproportion between them — the risk being insignificant in relation to the sacrifice — the defendants discharge the onus on them.
— Asquith LJ, Edwards v National Coal Board [1949] 1 KB 704
In plain English: a risk is ALARP when the cost (money, time, effort, disruption) of reducing it further is grossly disproportionate to the safety benefit of doing so. If a further risk reduction can be made at a cost that is proportionate to the benefit, it must be made.
Key principle: The duty holder (employer) must implement a control measure unless they can demonstrate that the cost is grossly disproportionate to the benefit. The burden of proof is on the duty holder to demonstrate ALARP — not on the regulator to prove it has not been achieved.
ALARP vs ALARA vs SFAIRP
These three acronyms are frequently confused:
| Acronym | Meaning | Primary Use |
|---|---|---|
| ALARP | As Low As Reasonably Practicable | UK OHS law (HSWA 1974), EU, Australia, and many other jurisdictions |
| ALARA | As Low As Reasonably Achievable | Radiation protection (ICRP standards, nuclear industry) |
| SFAIRP | So Far As Is Reasonably Practicable | Alternative expression of ALARP; used in UK legislation and some OHS management systems |
ALARP and SFAIRP mean the same thing — "so far as is reasonably practicable" and "as low as reasonably practicable" are equivalent expressions. ALARA is used specifically in radiation protection and has a slightly different nuance: it emphasizes continuous improvement toward the lowest achievable level, without requiring the gross disproportion test.
In general industry safety, ALARP is the relevant standard.
The ALARP Triangle: Three Risk Regions
The ALARP principle is typically visualized as a triangle divided into three risk regions:
Region 1: Unacceptable (Upper Zone) Risk is so high that it cannot be justified — work must stop until the risk is reduced to below the upper boundary. This region is defined by regulatory standards or established tolerability limits.
Region 2: ALARP (Middle Zone — "Tolerable Region") Risk is in the range where it might be acceptable, provided it has been reduced as far as reasonably practicable. This is where the cost-benefit analysis is applied. If there are cost-proportionate measures that would further reduce the risk, they must be implemented. If the only remaining reductions would be grossly disproportionate to the benefit, the risk has been demonstrated ALARP.
Region 3: Broadly Acceptable (Lower Zone) Risk is so low that no further control is required — it is broadly comparable to everyday life risks (driving to work, crossing the road). Controls may still exist but are not required by the ALARP standard.
How to Demonstrate ALARP
Demonstrating ALARP requires documented evidence that all reasonably practicable control measures have been identified and implemented. The process:
Step 1: Identify the Hazard and Estimate the Risk Level
Use a recognized risk assessment methodology:
- For quantitative ALARP demonstration: Quantitative Risk Assessment (QRA) to establish a numerical risk level (e.g., individual fatality risk per year)
- For semi-quantitative ALARP: Fine-Kinney method or risk matrix to establish a risk band
- For qualitative ALARP: structured qualitative assessment with documented reasoning
Step 2: Apply the Hierarchy of Controls
For each identified hazard, systematically work through the hierarchy:
| Control Level | Question | ALARP Application |
|---|---|---|
| Elimination | Can the hazard be removed entirely? | If elimination is technically feasible and not grossly costly, it must be implemented |
| Substitution | Can a less hazardous alternative be used? | If substitution is feasible and proportionate, it must be implemented |
| Engineering controls | Can physical safeguards reduce the risk? | Each credible engineering control must be evaluated for cost-benefit |
| Administrative controls | Can procedures, training, or supervision reduce the risk? | Generally low-cost; normally required unless they provide no meaningful risk reduction |
| PPE | Can personal protection reduce the residual risk? | Required where technically feasible; not a substitute for higher controls |
Step 3: Apply the Cost-Benefit Test
For each potential control measure that has NOT been implemented, document:
- The risk reduction the measure would achieve (in quantitative or qualitative terms)
- The cost of implementing the measure (financial cost, operational disruption, time)
- The assessment of whether the cost is grossly disproportionate to the benefit
How gross is "grossly disproportionate"? HSE has not published a fixed ratio. A paper by an HSE specialist inspector, Goose (2006), Gross disproportion, step by step, summarizes the regulator's position:
- HSE has not formulated an algorithm that gives the disproportion factor for a given level of risk.
- Even at the low end of risk, the disproportion must always be gross, reflecting an intended bias in favor of safety.
- The higher the risk, the higher the factor. Factors greater than 10 are unlikely, but not ruled out.
In practice, this means a measure whose cost is roughly equal to its safety benefit must be implemented. A measure can be rejected only when its cost is many times its benefit, and the closer the risk sits to the intolerable limit, the more times that has to be. For most workplace hazards the comparison is qualitative rather than a monetary calculation, but the reasoning must still be written down.
Step 4: Document the ALARP Case
The ALARP case is the documented record of the risk assessment, the controls considered, the controls implemented, and the rationale for any controls not implemented. It should be:
- Specific to the hazard and the location
- Based on best available information (industry data, manufacturer specifications, incident history)
- Proportionate to the severity of the risk (a complex written ALARP case for a high-consequence scenario; a brief recorded assessment for a low-consequence scenario)
- Dated and reviewed when conditions change
Step 5: Review and Update
ALARP demonstration is not a once-and-done exercise. The assessment must be reviewed:
- When the nature of the hazard changes
- When better control technology becomes available
- When an incident or near miss occurs
- When regulatory guidance changes
- At defined periodic intervals (typically annually for high-risk scenarios)
ALARP Demonstration: A Worked Example
The easiest way to understand an ALARP demonstration is to see one. The example below is illustrative: it uses a common warehouse hazard and the Fine-Kinney method to show the shape of the reasoning, not figures from a real site.
Hazard: Forklifts and pedestrians share the aisle between the loading dock and the dispatch office. Pickers cross it dozens of times per shift.
Starting risk (existing controls: high-visibility vests, a painted walkway): Probability 3 (unusual but possible) × Exposure 6 (daily) × Consequence 15 (a single fatality is credible) = 270. This falls in the "high" band and needs urgent corrective action.
Options considered (the options register):
| # | Option | Effect on risk | Sacrifice | Decision | Reason |
|---|---|---|---|---|---|
| 1 | Physical barriers separating the walkway from the forklift route, with gated crossing points | Large: removes most contact points | Moderate: materials, one weekend of installation | Implement | Benefit clearly outweighs cost |
| 2 | Blue-spot / red-zone warning lights on all forklifts | Moderate: warns pedestrians at crossings and blind corners | Low | Implement | Low cost, proportionate |
| 3 | Speed limiters set to walking pace in shared zones | Moderate: reduces both likelihood and severity | Low | Implement | Low cost, proportionate |
| 4 | Pedestrian exclusion from the dock during loading windows | Moderate | Low: schedule change | Implement | Administrative, but cheap and effective alongside 1–3 |
| 5 | Replace forklifts with automated guided vehicles (AGVs) | Small once 1–4 are in place | Very high: fleet replacement and floor rework | Reject for now | Cost grossly disproportionate to the small remaining benefit; revisit at next fleet renewal |
| 6 | Rely on high-visibility vests alone | Minimal | Already in place | Keep, not relied on | PPE is the last line of defense |
Residual risk after options 1–4: Probability 0.2 (practically impossible) × Exposure 6 × Consequence 15 = 18, in the "acceptable" band.
The ALARP statement records the conclusion in plain words, for example: "Risk from forklift and pedestrian interaction at the dispatch aisle has been reduced from 270 to 18 by segregation, warning lights, speed limiting and exclusion windows. The remaining option identified, replacing the fleet with AGVs, has been rejected because its cost is grossly disproportionate to the small additional reduction. This will be reviewed at the next fleet renewal, after any incident or near miss, or when the layout changes."
Three things make this a demonstration rather than a claim:
- Every option is listed, including the rejected ones.
- Each rejection has a reason tied to gross disproportion.
- The review triggers are defined in advance.
A risk score alone never proves ALARP. The options register does.
How to Run an ALARP Workshop
For significant risks, ALARP decisions are best made in a short, structured workshop rather than by one person at a desk. Industry guidance on ALARP stresses that optioneering, the structured search for alternative controls, is the part most often done badly, and that rejected options must be recorded with their justification.
Who attends: the risk owner (line manager), the person who does the task, a maintenance or engineering representative, the safety professional who facilitates, and, for high-consequence risks, someone with budget authority.
Inputs: the current risk assessment, incident and near-miss history, relevant standards and good practice, and photos or a site walk of the area.
Agenda (60–120 minutes per significant risk):
- Confirm the hazard and the current risk. Agree the scenario and score it with the method you use (matrix, Fine-Kinney or QRA).
- Check against good practice. If recognized good practice (standards, HSE guidance, industry codes) is not met, that gap must be closed before any cost argument starts.
- Generate options. Work down the hierarchy of controls: elimination, substitution, engineering, administrative, then PPE. Aim for quantity first; judge later.
- Assess each option. Estimate the risk reduction, the sacrifice, and any new hazards the option introduces.
- Decide and record. Implement, reject (with the gross disproportion reason), or defer (with a date or trigger).
- Assign actions. Give every implemented option an owner and a due date, and track it as a corrective action.
Outputs: an options register like the one above, a written ALARP statement, the action list, and the review triggers.
Common workshop failures:
- Starting from the control you already want and working backwards.
- Rejecting options as "too expensive" without comparing cost against benefit.
- Stopping at administrative controls because engineering options need budget.
- No record of rejected options, which leaves nothing to show an auditor or inspector later.
ALARP in Different Industry Contexts
| Industry | Risk Tolerability Limits | Common ALARP Methods |
|---|---|---|
| Oil and gas | HSE's Tolerability of Risk framework (UK): 10⁻³/year intolerable, 10⁻⁶/year broadly acceptable | QRA with ALARP case; Layer of Protection Analysis (LOPA) |
| Nuclear | Numerical targets set by the regulator (in the UK, the ONR Safety Assessment Principles' Basic Safety Levels and Objectives); ALARA applies to radiation doses | Probabilistic Safety Assessment; ALARA/ALARP combined |
| Construction | Qualitative risk matrix; Fatal Accident Rate (FAR) benchmarks | Qualitative ALARP using hierarchy of controls documentation |
| General industry | ISO 45001 requirements; jurisdiction-specific regulations | Semi-quantitative (Fine-Kinney) or qualitative ALARP case |
| Railways | ALARP within defined risk tolerability limits (10⁻⁴ to 10⁻⁶ for passengers) | Formal Safety Assessment; Cost Per Life Saved analysis |
Common Misunderstandings About ALARP
Misunderstanding 1: "ALARP means we only have to do what is cheap"
ALARP does not set a cost ceiling — it sets a disproportion test. A measure that costs $500,000 must be implemented if the safety benefit warrants it. The "gross disproportion" threshold is not low.
Misunderstanding 2: "Residual risk is automatically ALARP if we've done the risk assessment"
A risk assessment identifies hazards. An ALARP case demonstrates that all proportionate controls have been applied to those hazards. The two are not the same.
Misunderstanding 3: "ALARP doesn't apply in our jurisdiction because we're not UK-regulated"
ALARP as a legal standard originated in UK law, but the principle is embedded in OHS regulation worldwide. ISO 45001 requires that risks be reduced to the extent practicable. OSHA's General Duty Clause requires that recognized hazards be abated using feasible means. The cost-benefit analysis at the heart of ALARP is applied implicitly in every OHS regulatory system.
Misunderstanding 4: "PPE is ALARP because it's the most cost-effective control"
PPE is at the bottom of the hierarchy of controls for a reason — it provides the least reliable protection and must be worn correctly every time to be effective. PPE may be part of an ALARP case, but PPE-only controls are rarely ALARP when engineering or administrative controls are technically feasible.
ALARP and ISO 45001
ISO 45001 does not use the word "ALARP," but the concept is embedded in its requirements:
| ISO 45001 Clause | Requirement | ALARP Connection |
|---|---|---|
| Clause 6.1.2 | Identify hazards and assess risks | ALARP begins with risk assessment |
| Clause 6.1.4 | Determine the actions to address risks | Actions must reduce risk "as far as practicable" |
| Clause 8.1 | Operational planning and control | Controls must be implemented to manage risks |
| Clause 10.3 | Continual improvement | ALARP is not static — better controls must be implemented as they become available |
Frequently Asked Questions
What does ALARP stand for?
ALARP stands for "As Low As Reasonably Practicable". It means a risk has been reduced until the cost, time and effort of any further reduction would be grossly disproportionate to the safety benefit. In UK law the same duty is written as SFAIRP, "so far as is reasonably practicable".
Is there an ALARP certification?
No. Neither HSE nor any other regulator issues an ALARP certificate, and there is no certification scheme that makes a person or a site "ALARP certified". ALARP is demonstrated case by case through a documented risk assessment and options register. Training courses on ALARP and risk tolerability exist, and general OHS qualifications such as NEBOSH and IOSH cover the concept, but none of them certifies that a risk is ALARP.
Can a risk matrix show that a risk is ALARP?
Only partly. A risk matrix or Fine-Kinney score can place a risk in the unacceptable, tolerable or broadly acceptable region. It cannot show that every reasonably practicable control has been applied. That needs the list of options considered, the ones implemented, and the reason each rejected option was grossly disproportionate.
Is there a specific risk level that defines the ALARP boundary?
In quantitative risk assessment contexts (major hazard industries), the UK HSE's Tolerability of Risk framework, published in Reducing risks, protecting people (R2P2), gives indicative limits. An individual fatality risk above 1 in 1,000 per year (10⁻³) is intolerable for workers, and above 1 in 10,000 per year (10⁻⁴) for members of the public. Below 1 in 1,000,000 per year (10⁻⁶) is broadly acceptable (summary of R2P2). Between these levels, the risk is tolerable provided ALARP is demonstrated. These specific limits apply primarily to major hazard facilities. For general industry, qualitative and semi-quantitative methods are used to place risk within one of the three ALARP regions.
Does ALARP require a formal cost-benefit analysis document?
For high-risk scenarios, yes — a documented ALARP case with explicit cost-benefit reasoning is expected by regulators and required for demonstrating compliance with major hazard legislation (COMAH, OSHA PSM). For lower-risk scenarios, the ALARP case may be less formal — but it should still be documented. A risk assessment that notes "risk has been assessed and controls are in place" without documenting what controls were considered and why some were not implemented does not demonstrate ALARP.
How does ALARP relate to the precautionary principle?
The precautionary principle (applied in environmental law and some OHS contexts) suggests that where there is uncertainty about a risk, the burden of proof is on demonstrating safety rather than proving harm. ALARP is more specific — it applies to quantified or assessable risks and requires cost-benefit analysis. In practice, for novel or uncertain risks, the precautionary principle may require more stringent controls than the ALARP cost-benefit test alone would suggest.
Conclusion
ALARP is not a paperwork concept. It is a legal standard, a risk management philosophy, and a practical decision-making tool that answers the fundamental question every safety professional must be able to answer: "Have we done enough?"
The organizations that demonstrate ALARP most effectively are those that approach it not as a compliance exercise but as a genuine inquiry — systematically identifying every credible control measure for each significant hazard, honestly assessing the cost and benefit of each, implementing those that are proportionate, and documenting the reasoning for those that are not.
The alternative — claiming ALARP without the analysis — is not just a compliance failure. It is a failure of the duty of care that the organization owes to the people who work for it.
Download FindRisk to support your ALARP demonstrations — with AI-assisted hazard identification, Fine-Kinney risk scoring, hierarchy-of-controls structured assessments, and documented evidence of control selection rationale.
Try FindRisk
Ready to modernize your safety workflow?
Conduct AI-powered risk assessments, generate reports instantly, and keep your team safe — anywhere, anytime.
