Probability of failure vs consequence of failure
Probability of failure (PoF) is how likely equipment is to lose containment; consequence of failure (CoF) is how bad the loss would be. Risk-based inspection combines the two into risk, with PoF stated for a set period and CoF covering people, the environment and the business. Inspection mainly sharpens the PoF estimate; isolation, detection and inventory shape CoF.
By Jun Zhong, Editor·Researched and drafted by Beacon's AI newsroom·Checked by Jun Zhong, Editor·Last checked ·Next review
Why does it matter?
On 16 February 2007, at a refinery near Sunray, Texas, water that had settled in a control station idle for about 15 years froze and cracked an elbow in high-pressure propane piping 2. That was the probability side: a hazard missed because no management of change review was done when the station was taken out of service 2. Once propane escaped and ignited, operators could not stop the flow because no remotely operable shut-off valves were installed, and flame on unfireproofed support steel brought down a pipe rack 2. Four people were injured, the refinery was evacuated and shut for two months, and reported direct losses topped $50 million: the consequence side, set by isolation and fireproofing rather than by the pipe wall 2.
PoF vs CoF at a glance
In the American Petroleum Institute (API) method, risk-based inspection (RBI) treats failure as loss of containment: a leak to the atmosphere or a rupture 1. The two halves of risk are estimated separately and then combined 5,6.
| Probability of failure (PoF) | Consequence of failure (CoF) | |
|---|---|---|
| Question it answers | How likely is a loss of containment over a set period? 1,5 | If it happens, what does it do to people, property and the environment? 6 |
| How it is expressed | A frequency rating, or a probability from fault trees, reliability models or Bayesian networks 6 | A severity rating, or modeled outcomes such as fire, explosion and toxic exposure, sometimes turned into money 6 |
| Main inputs | Damage mechanisms, material and coating, protection systems, surroundings and actual operating conditions 5,9 | The fluid, how much of it can escape, how fast the release can be stopped, and who is nearby 7,2,5 |
| How it changes over time | Rises as damage builds up toward a risk target 1 | Changes when inventory, isolation or the population around the equipment changes 7,5 |
| What inspection does to it | Narrows the uncertainty about the damage present, so the estimate improves; the inspection itself does not lower risk 1 | Little: what a release does is set by what escapes and how fast it is stopped 2,3 |
| What lowers it | Removing or controlling the damage source, for example draining, isolating or removing an idle dead leg 2; a more resistant material 10 | Remote isolation, leak detection, water curtains, smaller isolatable inventories 2,8 |
| A typical error | Rating it from design data instead of the conditions the unit really runs at 9 | Leaving out how quickly the flow can be stopped 5 |
| In the Library | Probability of failure | Consequence of failure |
Sources: API 2025 catalog, RP 581 entry 1; US Chemical Safety and Hazard Investigation Board (CSB) reports 2007-05-I-TX 2, 2010-08-I-WA 9 and 2012-03-I-CA 10; NTSB/SS-15/01 5; Collina et al., 2026 6; 40 CFR 68.25 and 68.28 7,8.
When does probability of failure decide the action?
- Timing and method of inspection. API describes risk rising as damage accumulates toward a risk target, with an inspection recommended before it is reached 1. In US pipeline rules the same logic sets the schedule: liquid lines that could affect high consequence areas are assessed at intervals based on the risk each line poses 12.
- When the inputs are wrong. At Anacortes, Washington, in 2010, the probability calculation for high temperature hydrogen attack, run on design conditions, rated as low the exchangers that later ruptured and killed seven workers 9. CSB also judged inspection for that damage not reliable enough to depend on, and looked instead to inherently safer materials and to checking the conditions the equipment really sees 9,11.
- When corrosion varies from part to part. At Richmond, California, in 2012, thickness readings taken mostly on fittings high in silicon missed a low-silicon component that thinned much faster 10. CSB pointed to two ways out: inspect every component, or replace the line with a steel higher in chromium, which it called more effective 10.
When does consequence of failure decide the action?
- Isolation. After Sunray, the CSB found that the lack of remote isolation greatly increased the size and duration of the fire 2. At a Philadelphia, Pennsylvania, refinery in 2019 it concluded that remotely operated emergency isolation valves could have shortened the release and prevented the later explosions 3.
- Inventory. US accident prevention rules for chemical plants start the worst case from the largest amount held in one vessel or one pipe, counting administrative limits on that amount 7. The alternative scenario each plant must also report is a release more likely than the worst case, and it may take credit for active and passive mitigation that would survive the event 8. Smaller isolatable inventories and working mitigation shrink the consequence whatever the probability.
- Where people are. Federal gas pipeline integrity programs exist because a failure in a populated area, a high consequence area, carries far more risk; operators have had to run them since 2004 5. In NTSB's 2015 study, operators' consequence models leaned mainly on population density, and none counted emergency response factors such as how far away the shut-off valves were or whether they could be closed remotely 5.
A common mix-up
Risk is a combination, not an average. A 1992 paper by the American Society of Mechanical Engineers (ASME) risk-based inspection task force expressed risk as rupture frequency multiplied by consequence 4. Probabilistic models can give risk in absolute units, such as deaths per mile per year; relative models have no common unit, so they rely on weights to add different threats and consequences together 5. Averaging can hide the item that matters: NTSB warned that a length-weighted risk figure over a long stretch of pipeline may bury a short segment with high risk 5. Plotting the two on a risk matrix, consequence on one axis and likelihood on the other, keeps them visible side by side 6. PoF is also not the same as probability of detection (POD), the chance that an inspection finds a flaw.
Where integrity problems show up
| Decision | Data it depends on | What goes wrong without it | In the Library |
|---|---|---|---|
| Rating likelihood for idle piping | Damage mechanisms of out-of-service lines, such as water collecting and freezing | Sunray, Texas, 2007: a control station idle for about 15 years was left connected as a dead leg; trapped water froze and cracked an elbow 2 | Dead leg |
| Rating likelihood in hydrogen service | Measured temperatures and hydrogen pressures | Anacortes, 2010: design conditions gave a low result for exchangers that ruptured 9 | What is high temperature hydrogen attack (HTHA)? |
| Rating consequence for large flammable inventories | Isolatable inventory and remote isolation | Sunray, 2007: no remotely operable shut-off valves, so the propane could not be stopped 2; Philadelphia, 2019: no remote isolation of large hydrocarbon sources near the failed elbow 3 | Consequence of failure |
| Weighting threats along a pipeline | Threats specific to each segment | NTSB: weights applied across a whole system, favoring a few threats, can understate a segment exposed to another, such as a pipeline crossing a fault line 5 | What is pipeline integrity management? |
Sources: CSB reports 2007-05-I-TX, 2008 2, 2010-08-I-WA, 2014 9, and 2019-04-I-PA, 2022 3; NTSB/SS-15/01, 2015 5.
Common questions
- Which matters more, PoF or CoF?
- Neither alone. Risk combines them, so a high-risk item can be brought down from either side. At Sunray, Texas, the probability side was an idle line left full of water; the consequence side was the lack of remote shut-off valves, which let the fire grow.
- Does inspection reduce the probability of failure?
- Not by itself. API describes inspection as reducing uncertainty about the damage present, which lets the risk be estimated more accurately.
- How is the consequence of failure estimated?
- By a ranking from experienced judgement, or by modeling the release: hole size, fluid, inventory, ignition and the people and assets exposed. See [Qualitative vs semi-quantitative vs quantitative RBI](/library/qualitative-vs-quantitative-rbi/).
Beacon explains these concepts in its own words for education. It does not reproduce or replace the standards cited, and Winda is not affiliated with or endorsed by the organizations that publish them. Engineering decisions need the current edition of the standard and a qualified engineer.
Sources
- 1.2025 Publications Catalog (entry for RP 581) (RP 581, 3rd edition, April 2016) · American Petroleum Institute (API)
- 2.Investigation Report: LPG Fire at Valero-McKee Refinery, Sunray, Texas, February 16, 2007, Report No. 2007-05-I-TX · US Chemical Safety and Hazard Investigation Board (CSB)
- 3.Fire and Explosions at Philadelphia Energy Solutions Refinery Hydrofluoric Acid Alkylation Unit, Investigation Report No. 2019-04-I-PA · US Chemical Safety and Hazard Investigation Board (CSB) ·
- 4.Risk Based ASME Code Requirements (B. F. Gore, T. V. Vo, K. R. Balkey), PNL-SA-21211, OSTI 7129942 · Pacific Northwest Laboratory, US Department of Energy, via OSTI.gov
- 5.Integrity Management of Gas Transmission Pipelines in High Consequence Areas, Safety Study NTSB/SS-15/01 · US National Transportation Safety Board (NTSB) ·
- 6.Toward the application of the Risk-Based Maintenance approach to a hydrogen-fueled manufacturing plant (G. Collina and others), International Journal of Hydrogen Energy 238, 155318, open access under CC BY 4.0 · Elsevier, via OSTI.gov (co-authored by the US National Laboratory of the Rockies)
- 7.40 CFR 68.25, Worst-case release scenario analysis · eCFR, US Office of the Federal Register
- 8.40 CFR 68.28, Alternative release scenario analysis · eCFR, US Office of the Federal Register
- 9.Investigation Report: Catastrophic Rupture of Heat Exchanger (Seven Fatalities), Tesoro Anacortes Refinery, Anacortes, Washington, April 2, 2010, Report No. 2010-08-I-WA · US Chemical Safety and Hazard Investigation Board (CSB) ·
- 10.Final Investigation Report: Chevron Richmond Refinery Pipe Rupture and Fire, Report No. 2012-03-I-CA · US Chemical Safety and Hazard Investigation Board (CSB) ·
- 11.Recommendations Status Change Summary, Recommendation 2010-8-I-WA-R11 to the American Petroleum Institute (API RP 581) · US Chemical Safety and Hazard Investigation Board (CSB) ·
- 12.49 CFR 195.452, Pipeline integrity management in high consequence areas · eCFR, US Office of the Federal Register