When an inspection finds metal loss on a pipeline, the immediate question is whether the pipe is still safe at its operating pressure or whether it needs a repair. ASME B31G is the long-established method that answers this by estimating how much strength a corroded region has lost. It takes the length and depth of a metal-loss defect and computes a predicted failure pressure and a safe operating pressure, giving integrity engineers a defensible basis for a repair-or-monitor decision. This guide walks through what B31G calculates, why it uses a simplified parabolic area assumption, how modified B31G improves the model, and how inline-inspection anomaly dimensions feed the calculation.
B31G Remaining Strength in one line: The B31G remaining strength calculation is an ASME method that estimates the failure pressure and safe operating pressure of a corroded pipeline from the length and maximum depth of a metal-loss defect. It compares the strength lost to a corroded area against the intact pipe to predict the pressure at which the flaw would fail, then applies a safety factor to set a safe operating pressure. The original B31G is deliberately conservative and uses a simple parabolic area assumption, while modified B31G refines the flow-stress and area terms to give a more realistic and less conservative result.
The core idea of B31G is that a patch of corrosion weakens a pipe in proportion to how much wall it has removed and how far that removal extends along the pipe. Two dimensions drive the result: the maximum depth of the metal loss, expressed as a fraction of the original wall thickness, and the axial length of the defect. A short, shallow flaw barely affects the pipe, while a long, deep one removes enough load-bearing wall to matter. The method combines these with the pipe's diameter, wall thickness, and material strength to estimate the pressure at which the corroded region would burst.
From that predicted failure pressure the method derives a safe operating pressure by applying a design safety factor, so the pipe is not run anywhere near the calculated burst point. The comparison that comes out of the calculation is straightforward in spirit: if the maximum allowable operating pressure sits comfortably below the safe pressure the defect can tolerate, the flaw is acceptable to leave in service, and if it does not, the pipe needs a pressure reduction, a repair, or replacement of that section. This single answer is what makes B31G so widely used for triage.
It is important to be clear about what the method assumes. B31G treats the corrosion as a blunt, longitudinally oriented area of metal loss in a ductile line-pipe steel, and it assumes the failure is driven by the hoop stress from internal pressure. It is not intended for sharp, crack-like defects, for damage such as dents or gouges, or for very deep localized pits that pass through most of the wall, all of which fall outside its scope. Recognizing where B31G applies is as important as running the numbers, because using it on the wrong kind of flaw gives a misleading answer.
The original B31G represents the shape of the metal loss with a simple geometric idealization. Rather than mapping the real, irregular contour of the corroded region, it approximates the missing cross-sectional area as a parabola for shorter defects and as a rectangle for longer ones. This parabolic area assumption is a modelling shortcut that made the method usable with hand calculations and a single depth-and-length measurement, and it deliberately errs on the safe side, so the predicted strength tends to be lower than the pipe's real capacity.
That conservatism is useful for a quick screen but can flag defects as unacceptable that a more careful analysis would clear, which drives unnecessary digs and repairs. Modified B31G was developed to reduce that over-conservatism in two main ways. It uses a more realistic flow-stress value, the effective strength at which the steel is treated as yielding, rather than the coarser assumption in the original. It also replaces the rigid parabola-or-rectangle choice with an area factor that better represents a general corroded shape, giving a metal-loss area estimate closer to reality.
The practical effect is that modified B31G usually predicts a higher, more accurate remaining strength than the original for the same defect, so a flaw that the original method rejects may pass under the modified method. Both remain level-one style assessments that work from a defect's length and a single maximum depth, which keeps them simple and quick. When even the modified method is too conservative, or when a detailed depth profile of the corrosion is available, engineers move to an effective-area analysis that uses the real measured shape instead of an assumed one.
In modern pipeline integrity work the length and depth that B31G needs usually come from inline inspection. An inspection tool run through the line reports each metal-loss anomaly with an estimated depth as a percentage of wall thickness and an axial length, and those two numbers are exactly the inputs the calculation consumes. An integrity management system can take the tool's anomaly list and run B31G or modified B31G across every reported feature, producing a predicted safe pressure for each so the population can be ranked by severity.
That automated screening is what turns thousands of reported anomalies into a manageable action list. Features whose calculated safe pressure comfortably exceeds the operating pressure can be scheduled for monitoring or left for the next inspection, while features that fall below the required margin are prioritized for a verification dig and possible repair. Because inline tools report depth within a sizing tolerance rather than exactly, prudent programs apply the tool's stated uncertainty to the depth before the calculation, so a feature reported near the acceptance line is treated conservatively rather than being cleared on an optimistic reading.
Bringing the results into a monitoring and reporting platform lets operators track how these defects behave over time and tie the engineering answer to real conditions in the field. Where a cloud SCADA platform such as Merobix logs the line's operating pressure history, that record can be set against the B31G safe pressure for a given anomaly to confirm the pipe has actually been operated within the assessed envelope, and pressure trends and any planned deratings can be surfaced alongside the integrity results. Successive inspections then let the depth of a feature be trended run over run, so a slowly growing defect is caught and re-assessed before it approaches its calculated limit.
B31G needs the geometry of the metal-loss defect and the pipe itself. From the defect it needs the maximum depth, expressed as a fraction of the wall thickness, and the axial length. From the pipe it needs the outside diameter, the nominal wall thickness, and the material's specified strength. With those it computes a predicted failure pressure and, after a safety factor, a safe operating pressure for the corroded region.
Modified B31G refines two parts of the original model to reduce its over-conservatism. It uses a more realistic flow-stress value for the steel instead of the coarser original assumption, and it replaces the fixed parabola-or-rectangle area choice with an area factor that better represents a general corroded shape. The result is usually a higher and more accurate remaining strength for the same defect, so a flaw the original method rejects may pass under the modified method.
B31G and modified B31G work from a single maximum depth and a length, which makes them simple but conservative because they assume an idealized defect shape. When that conservatism would drive unnecessary repairs and a detailed measured depth profile of the corrosion is available, an effective-area analysis such as RSTRENG is used instead. It works from the real river-bottom profile of the feature and iterates over subsections to find the worst case, giving a less conservative and more realistic burst-pressure estimate.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.