6.1 POF Methodologies: Qualitative, Semi-Quantitative, and Quantitative Modeling

Foundational Principles of Probability of Failure (POF) Analysis

Under API Recommended Practice 580 (4th Edition, August 2023 with Addenda 1, March 2025), Probability of Failure (POF) represents the numerical or categorical likelihood that a pressure-retaining component will experience a loss of containment (LOPC) event within a defined operating timeframe (typically expressed as failures per year). In a Risk-Based Inspection (RBI) program, POF evaluation operates in tandem with Consequence of Failure (COF) analysis to quantify total equipment risk ($Risk = POF \times COF$).

Section 10 of API RP 580 dictates that POF is not a static property; rather, it reflects the progressive physical degradation of materials under process conditions, modified by operational management controls and verified through targeted Non-Destructive Examination (NDE). API RP 580 establishes three standardized methodologies for modeling POF: Qualitative, Semi-Quantitative, and Quantitative.

---

1. Qualitative POF Assessment Methodology

Concept and Architecture

Qualitative POF assessment relies on multidisciplinary engineering analysis, historical operating experience, and structured ordinal scales (such as 1 to 5, or Category A through E) to rank failure likelihood. Rather than calculating exact numerical failure rates, qualitative models group equipment into broad relative likelihood categories based on discrete risk criteria.

Data Inputs and Evaluation Workflow

• Process and Environmental Severity: Categorizes operating fluids based on general aggressiveness (e.g., highly corrosive wet $\text{H}_2\text{S}$ vs non-corrosive dry natural gas).
• Material Susceptibility: Evaluates alloy compatibility without detailed kinetic rate modeling.
• Inspection History Rating: Assigns discrete credit scores based on overall inspection frequency and qualitative quality impressions.
• Damage Mechanism Screening: Relies on expert judgment from corrosion specialists to identify active and potential degradation mechanisms (API RP 571).

Advantages and Operational Constraints

---

2. Semi-Quantitative POF Assessment Methodology

Concept and Architecture

Semi-Quantitative POF modeling functions as a hybrid framework, combining the speed and structured logic of qualitative assessments with numerical scoring algorithms and bounded lookup tables. It translates qualitative engineering inputs (e.g., temperature bands, fluid severity tiers) into discrete numerical points, which are subsequently aggregated to compute a relative POF score.

Calculation Structure

In a typical semi-quantitative model:

1. Damage Susceptibility Score ($S_{damage}$): Assigned from 1 to 100 based on process parameters (pH, velocity, temperature) and material properties.
2. Inspection Credit Factor ($C_{insp}$): Subtracts points based on the number and quality of past inspections.
3. Management and Maintenance Score ($M_{maint}$): Modifies the score based on facility operational discipline and IOW compliance.

$\text{Semi-Quantitative POF Score} = (S_{damage} \times M_{maint}) - C_{insp}$

The resulting composite score maps directly into discrete risk matrix categories (e.g., Score 0–20 = Category 1 Low; Score 81–100 = Category 5 High). This approach provides greater repeatability and analytical rigor than purely qualitative methods without requiring the massive data infrastructure of fully quantitative modeling.

---

3. Quantitative POF Assessment Methodology (API RP 581 Framework)

The Foundational Mathematical Equation

Quantitative POF modeling, as standardized in API RP 581 (Risk-Based Inspection Methodology), calculates an absolute, time-dependent annual failure frequency ($P_{of}(t)$, expressed in failures per year) for each pressure-containing component item. The core mathematical formulation is:

$P_{of}(t) = gff \times DF(t) \times F_{MS}$

Where:

• $P_{of}(t)$: The total time-dependent Probability of Failure at operating time $t$ (failures/year).
• $gff$: The baseline Generic Failure Frequency for the specific equipment component type and release hole size.
• $DF(t)$: The component structural Damage Factor at operating time $t$, quantifying physical degradation from active damage mechanisms.
• $F_{MS}$: The Management System Factor, reflecting facility-wide mechanical integrity management system quality.

---

Generic Failure Frequency ($gff$) Derivation and Release Hole Sizes

In API RP 581 (Part 2), baseline generic failure frequencies ($gff$) are established from statistical analysis of extensive historical industry failure databases across refining, petrochemical, and chemical process facilities.

Release Hole Size Classification

To account for varying failure severity, API RP 581 defines four discrete release hole sizes for each component category:

1. Small Hole ($0.25\text{ in} / 6.4\text{ mm}$): Represents pinhole leaks, packing failures, small fitting cracks, or localized pitting penetrations.
2. Medium Hole ($1\text{ in} / 25\text{ mm}$): Represents moderate line cracking, small nozzle separations, or localized wall thinning failures.
3. Large Hole ($4\text{ in} / 102\text{ mm}$): Represents major pipe splits, large nozzle ruptures, or significant structural breaches.
4. Rupture ($>4\text{ in}$ or full bore): Represents complete catastrophic structural failure, full-bore pipe guillotine rupture, or vessel shell burst.

Mathematical Summation of $gff$

The total baseline generic failure frequency ($gff_{total}$) for a component is the sum of the generic failure frequencies across all four release hole sizes:

$gff_{total} = \sum_{i=1}^{4} gff_i = gff_{small} + gff_{medium} + gff_{large} + gff_{rupture}$

For example, a standard carbon steel process pipe circuit has individual hole size baseline frequencies defined in API RP 581:

• $gff_{small} = 8.0 \times 10^{-6}$ failures/yr
• $gff_{medium} = 2.0 \times 10^{-5}$ failures/yr
• $gff_{large} = 2.0 \times 10^{-6}$ failures/yr
• $gff_{rupture} = 6.0 \times 10^{-7}$ failures/yr
• $gff_{total} = 3.06 \times 10^{-5}$ failures/yr

---

Comparison Matrix of POF Methodologies (API RP 580 Section 10)

---

Selection Criteria for POF Methodologies

API RP 580 Section 10 emphasizes that no single methodology is universally superior; the choice depends on:

1. Objective of the RBI Study: Plant-wide prioritization screening favors qualitative methods, whereas optimizing turnaround NDE plans for high-consequence units mandates quantitative modeling.
2. Data Availability and Quality: Complete material test reports, continuous operating temperature/pressure trends, and extensive thickness monitoring location (TML) data enable quantitative POF; incomplete data sets necessitate qualitative or semi-quantitative approaches.
3. Resource and Cost Constraints: Quantitative modeling requires specialized RBI software and detailed corrosion engineering modeling, carrying higher implementation costs.