RBI is often sold as a way to inspect less. It is better understood as a way to inspect the right things, and the mechanism by which it works is worth understanding properly, because the same arithmetic that extends an interval will extend one it should not.
Two documents get quoted together and they are not the same kind of thing.
API 580 is a recommended practice describing what an RBI programme must contain to be credible: how you define the scope, identify mechanisms, assess probability and consequence, plan the inspection, document it and keep it current. It tells you what a defensible programme looks like. It does not tell you how to compute anything.
API 581 is one accepted way to do the computing. It is a quantitative methodology with equations, tables and damage factors in it. You can satisfy 580 without using 581 — a qualitative matrix built on plant experience can be a legitimate RBI programme — but if somebody says "the RBI number", they usually mean an output of 581 or a commercial implementation of it.
Risk is probability of failure multiplied by consequence of failure. Both halves have to be estimated, and they behave completely differently.
Consequence is mostly fixed by the plant. What is in the pipe, how much of it, how flammable or toxic, what is nearby, what production stops. You can change it by changing the process or the layout, and short of that it is what it is. Inspection does not touch it.
Probability is where inspection lives. In the 581 framework it is built from a generic failure frequency for the equipment type, multiplied by a damage factor that reflects the condition of that specific item, adjusted by a management systems factor covering the quality of the plant's processes.
It is calculated from the damage mechanisms that are credible for that item, the rate at which they run, how much life has been consumed — and the number and effectiveness of the inspections already performed. That last term is the lever. A good inspection does not slow the corrosion down. It reduces your uncertainty about the corrosion, and reduced uncertainty is what lowers the calculated probability.
Inspections in this framework are graded — conventionally from highly effective down to ineffective — and the grade depends on whether the method actually detects the mechanism you are worried about, over enough of the component to matter.
This is the honest justification for spending more on an inspection: not that it is more thorough in the abstract, but that a highly effective inspection lowers the damage factor enough to earn an interval that a poor one does not.
A risk matrix with every item plotted on it, and — this is the part that matters — a plan. For each item: which mechanisms are credible, which method addresses each one, how much coverage, and when. An RBI study that produces a coloured matrix and no method-level plan has done the arithmetic and skipped the deliverable.
Every number downstream depends on having identified the credible damage mechanisms correctly. Miss one — miss that a line runs below dew point at the cold end, that a dead leg exists, that the amine is carrying heat-stable salts, that one elbow is low-silicon — and the damage factor is computed for mechanisms that are not the ones actually running.
The output is not obviously wrong. It is a low risk score with a long interval and a documented basis, which is considerably more dangerous than having no study at all, because it withstands challenge.
The corollary is where the effort belongs. The quality of an RBI programme is decided in the corrosion-loop and mechanism-identification workshop, by people who know the process and the plant, long before any software is opened. Everything after that is bookkeeping on top of that judgement.
Used properly, RBI concentrates effort where the risk is and stops spending it where it is not. Used as a way to justify inspecting less, with a mechanism list nobody argued about, it is a well-documented way to be surprised.
Every risk calculation rests on naming the credible damage mechanisms correctly. The library covers sixty-four of them, with what each one needs in order to run.
Open the mechanism libraryWho writes this. A mechanical engineer with twelve years in oil and gas — in-line inspection, fired heater and furnace inspection, and pipeline integrity. What is here comes from the published standards and from what those years in the field actually looked like. It is not written by an API-certified inspector.
This is not an assessment. Nothing on this site may be used to justify a decision about real equipment. Assessing plant requires the current editions of the applicable codes, data from a licensed source, and a competent engineer who signs for the answer. · Integrity Field Guide