Pipeline Integrity Assessment Methods: ILI, Pressure Testing, and Direct Assessment

Pipeline integrity assessment is the process by which operators evaluate the actual physical condition of a covered pipeline segment. The three primary regulatory pathways are in-line inspection (ILI), hydrostatic pressure testing, and direct assessment. Each method has defined applicability criteria, threat coverage characteristics, and technical limitations — and selecting the right method for a given segment is not a vendor decision or a scheduling convenience. It flows directly from the threat identification that precedes it. Using an assessment method that cannot detect the primary identified threat on a segment does not satisfy the regulatory requirement for that segment, regardless of when the assessment was performed.

What Are Pipeline Integrity Assessment Methods?

Under 49 CFR Part 192 Subpart O and 49 CFR 195.452, operators must perform integrity assessments of all covered pipeline segments — those that could affect high consequence areas. The regulation does not mandate a single approach. Instead, it defines acceptable methods and requires that the method selected be appropriate for the threats present on the segment being assessed.

The acceptable methods are:

• In-line inspection (ILI) — deploying instrumented tools through the pipeline to detect, locate, and size anomalies
• Hydrostatic pressure testing — pressurizing the pipeline to demonstrate its ability to sustain a target pressure
• Direct assessment — structured field programs that evaluate specific threats through surface surveys and targeted excavations
• Other technology — alternative methods an operator demonstrates provide equivalent understanding of pipe condition

Each method produces different information at different confidence levels. ILI produces a quantitative picture of anomaly populations across the entire assessed length. Pressure testing produces a pass/fail result at a specific test pressure. Direct assessment produces condition data at discrete excavation sites selected by a structured protocol. Understanding what each method can and cannot tell you is the foundation of defensible assessment program design.

Why Assessment Method Selection Matters

The most important decision in integrity assessment planning is upstream of method selection: it is the threat identification. When threats are correctly and specifically identified for each covered segment, the appropriate assessment method or combination of methods is largely prescribed by the technical standards — ASME B31.4, B31.8S, and the applicable API matrices. The vulnerability is not in the method selection process itself; it is in the threat identification that feeds it.

When threat identification is generic or incomplete, method selection inherits those gaps. A program that identifies corrosion as the only threat on a segment and selects standard axial MFL ILI — without evaluating whether cracking is also present — has selected a method that is appropriate for its stated threat but may be blind to an unevaluated one. An MFL tool has a known physical limitation in detecting crack features with low volumetric cross-section loss. Drawing conclusions about the absence of cracking from an MFL data set alone is not technically supportable.

Every technology has a detection bias. Sound assessment program management requires knowing what each tool is designed to find, what it is not designed to find, and where the boundaries of its performance specification apply in practice.

In-Line Inspection (ILI)

ILI is the most widely used and generally preferred integrity assessment method for piggable transmission pipelines. Instrumented tools — commonly called smart pigs — traverse the pipeline interior, collecting continuous sensor data that is processed to detect, locate, and size anomalies. The goal is to identify, characterize, and size anomalies across the entire inspected length.

ILI Tool Technologies

Magnetic Flux Leakage (MFL) is the most commonly deployed ILI technology. MFL tools detect volumetric metal loss — corrosion, pitting, and general wall thinning — by measuring flux leakage from magnetized pipe wall. MFL is the standard method for corrosion assessment and produces reliable detection and sizing for metal loss features above its performance specification thresholds. It is not a crack detection technology. Low-volume crack features do not produce sufficient flux leakage to be reliably detected by standard axial MFL.

Ultrasonic Testing (UT) tools measure wall thickness directly using acoustic transducers. UT provides high-resolution wall thickness data and can detect cracking anomalies. However, UT tools require a liquid couplant between the transducer and the pipe wall. This makes them well-suited for liquid pipelines and impractical for gas pipelines without significant operational adaptation.

Electromagnetic Acoustic Transducer (EMAT) technology generates ultrasonic waves in the pipe wall using electromagnetic induction, eliminating the liquid couplant requirement. EMAT is the primary ILI technology deployed for crack detection in gas pipelines, including stress corrosion cracking. The practical limitation of EMAT is anomaly characterization: the technology has shown good ability to detect features that MFL cannot, but it can struggle to identify precisely what it has found. EMAT is good at seeing that something is there; it is less reliable at characterizing what that something is. When EMAT anomalies cannot be characterized remotely, field excavation deploys a full suite of non-destructive examination (NDE) tools prepared to evaluate whatever is present.

Caliper tools detect and measure geometric deformations: dents, ovalization, and mechanical damage. High-resolution caliper tools also carry inertial measurement unit (IMU) sensors that map pipeline centerline position and bending strain, which is the basis for geohazard displacement monitoring and strain-based integrity assessment.

Piggability Constraints

ILI is only available for pipelines that can accommodate tool passage. The mechanical requirements are: consistent internal diameter, full-bore valves along the route, adequate bend radii, compatible launcher and receiver facilities, and sufficient operating pressure and flow velocity to propel the tool at acceptable speeds.

Common ILI infeasibility conditions include reduced-port valves that restrict passage, tight-radius bends and miter bends that exceed tool capability, diameter transitions of more than one nominal pipe size, manifold and station facility piping configurations, and operational constraints including insufficient flow velocity or operating pressure. When ILI is infeasible, operators must select a direct assessment method appropriate to the identified threats — or the specific combination of constraints that makes ILI infeasible may itself signal a segment that warrants heightened scrutiny about its assessment history.

Using ILI Data Effectively

ILI produces more than a list of anomalies for excavation. The full dataset, when properly analyzed, supports interval justification, threat population characterization, growth rate modeling, and program refinement.

A defensible interval justification built from ILI data starts with threat categories, not individual anomalies. Cycling through the threat populations present in the data set — corrosion features, crack indications, dents, manufacturing anomalies, girth weld and long seam populations — each receives a failure pressure analysis using recognized assessment equations: modified B31G or RSTRENG for metal loss, log-secant for cracking, B31.8 for bending strain. Anomalies that approach critical thresholds are identified for integrity excavation. Where ILI findings are inconsistent with the predicted threat picture based on the prior assessment, that inconsistency is itself an analytical trigger — it means either the threat identification, the growth rate assumptions, or the tool performance is not aligned with what the pipe is actually showing.

ILI tool measurement uncertainty must be factored into fitness-for-service decisions. Performance specifications are not guarantees — they are statistical statements about tool performance under defined conditions. Vendors are commercially incentivized to represent their tools as capable. The operator's obligation is to apply a technically appropriate level of scrutiny to vendor performance claims, particularly for features at or near detection thresholds.

Hydrostatic Pressure Testing

Pressure testing demonstrates that a pipeline can sustain a defined test pressure without failure. The test pressure is typically established as a multiple of MAOP, and the pipeline must hold the test pressure for a defined duration without a pressure drop indicating a failure.

The key characteristic of pressure testing is that it produces a binary result: the pipeline either sustains the test pressure or it does not. It does not produce quantitative data about the population of anomalies present along the inspected length. A pipeline that passes a pressure test has demonstrated that no defect of sufficient severity to fail at the test pressure existed at the time of the test. It has not demonstrated what defects remain below that threshold or how quickly those defects are growing.

Pressure testing is the method of record for many older pipelines that pre-date ILI availability and for segments that cannot be pigged. It remains an acceptable baseline assessment and reassessment method under both Part 192 and Part 195. However, for segments with active time-dependent threats — particularly corrosion — a pressure test that reveals no failure does not provide the quantitative condition data that supports a rigorous interval justification or growth rate analysis. This is a meaningful difference in program defensibility relative to ILI-based programs.

Direct Assessment

Direct assessment is a structured, four-phase evaluation process applicable to specific threat types when ILI is not feasible or as part of a confirmatory assessment program. The three regulatory direct assessment protocols are External Corrosion Direct Assessment (ECDA), Internal Corrosion Direct Assessment (ICDA), and Stress Corrosion Cracking Direct Assessment (SCCDA).

Each protocol follows the same four-phase structure: pre-assessment (reviewing existing data and determining applicability), indirect inspection (surface surveys to detect anomaly indications), direct examination (excavation and physical inspection at selected sites), and post-assessment (evaluating findings, establishing reassessment intervals, and documenting program effectiveness).

ECDA

ECDA is the most widely used direct assessment protocol, applicable to buried steel pipelines with cathodic protection systems. The indirect inspection phase deploys aboveground survey methods — close-interval survey (CIS) and direct current voltage gradient (DCVG) — to identify locations where CP effectiveness may be compromised, which serves as a surrogate indicator of external corrosion risk.

Survey data quality is the most critical variable in ECDA defensibility. The first checks when reviewing ECDA survey data are: CIS survey settings, waveform captures confirming that rectifier interruption is synchronized across the survey, and the date/time continuity of the survey record. Timestamp breaks — where the survey was interrupted and resumed — must be individually investigated, because equipment issues, operator errors, or changes in rectifier cycling during a survey gap can produce data that appears continuous but is fundamentally incorrect. Depth of cover is logged to the center of the pipe; on large-diameter pipelines (24 inches and above), the difference between centerline depth and top-of-pipe depth introduces a meaningful error into depth compliance assessments if not corrected.

Region definition is equally important. An ECDA region is only compliant if the indirect survey method used is applicable throughout the region. A CIS survey conducted over asphalt or concrete pavement may not produce interpretable data, making that portion of the survey non-compliant. Programs that do not account for survey applicability during region definition can produce survey coverage that appears complete on paper but is not analytically valid in those locations.

Severity criteria for identifying excavation sites must be pre-established in the procedure and applied consistently. Applying criteria after seeing the data introduces confirmation bias and creates an audit vulnerability.

ICDA

ICDA is applicable to dry gas pipelines where internal corrosion is a potential threat. It is less commonly used than ECDA and is most frequently deployed as part of a confirmatory direct assessment cycle that extends the effective reassessment interval — for example, an ILI every 14 years with ICDA performed at the seven-year midpoint.

The analytical foundation of ICDA is gas flow modeling to identify critical angles where water could drop out of the gas stream at operating conditions. Low points in the elevation profile become the primary direct examination targets, starting at the furthest upstream low point and proceeding downstream until internal corrosion is no longer observed. In practice, when ILI is feasible, it is generally preferred over ICDA — ILI simultaneously addresses both internal and external threats, while ICDA is limited to internal corrosion and requires that the pipeline be genuinely not susceptible to other threats or that those threats be addressed separately.

Situations where ILI is genuinely infeasible for mechanical reasons — reduced-port valves, miter bends, diameter changes of more than one nominal pipe size, low flow rates or operating pressures, vertical configurations — are the conditions that make ICDA a primary assessment method rather than a supplemental one.

SCCDA

SCCDA follows the same four-phase structure as ECDA but with the analysis framework adapted to target stress corrosion cracking rather than external corrosion. SCC is an environmentally assisted cracking mechanism, and the SCCDA protocol places increased emphasis on environmental conditions — soil chemistry, coating type, cathodic protection level, and operating temperature — that influence SCC susceptibility.

High-pH SCC tends to concentrate in the first 15 to 20 miles downstream of a compressor station, in areas with high-pH soil environments and near-neutral coating conditions. Near-neutral pH SCC does not follow the same geographic distribution — it can occur throughout the system and is more difficult to address because susceptibility correlates with combinations of coating disbondment, limited cathodic protection, and specific soil conditions that can exist across broad system segments. SCCDA programs that apply high-pH criteria to a near-neutral pH SCC environment will systematically miss the population they are trying to address.

Choosing the Right Assessment Method

The selection of an assessment method for a given covered segment should follow directly from the threat identification for that segment. Once threats are correctly identified and documented, the appropriate assessment method or combination of methods is substantially prescribed by the technical standards. The decision tree is not complicated when the threat identification is sound.

The most common method-selection error is not picking the wrong technology from a shortlist of plausible options — it is selecting a method that is appropriate for the documented threats while having an undocumented or underevaluated threat that the selected method cannot detect. The selection looks defensible on paper, but the program has a blind spot.

When ILI is available, it is generally preferred for piggable transmission pipelines because it simultaneously addresses multiple threat categories and produces quantitative condition data across the full assessed length. When ILI is infeasible, the direct assessment protocol appropriate to the primary threat on the segment should be selected, with attention to its specific applicability criteria and limitations.

Frequently Asked Questions

Can ILI satisfy the assessment requirement for all threats on a segment?

Not necessarily. Multiple ILI technologies may be needed to address all identified threats. Standard axial MFL covers metal loss but is not appropriate for crack detection. EMAT addresses cracking in gas pipelines. Caliper tools address deformation and mechanical damage. Running only one technology on a segment with multiple identified threats may leave those threats unaddressed for assessment purposes.

Is ECDA an acceptable substitute for ILI on a piggable pipeline?

ECDA is an acceptable assessment method under the regulations but is generally not the preferred choice when ILI is feasible. ILI provides quantitative anomaly data across the full pipe length; ECDA provides condition data at discrete excavation sites selected by the protocol. For a piggable pipeline with external corrosion as the primary threat, ECDA may satisfy the regulatory requirement, but it produces a materially different body of evidence than ILI. The program documentation should explain the basis for the method selection.

What makes a reassessment interval defensible when built on ILI data?

A defensible interval starts with threat-category analysis of the full ILI data set — not just the reported anomaly list. The supporting documentation should include failure pressure calculations for the identified anomaly populations, growth rate analysis, and consideration of accelerated time-dependent threats such as DC interference or known inhibitor program changes. Recognized engineering equations — modified B31G or RSTRENG for metal loss, log-secant for cracking — should be cited. Tool performance tolerances must be incorporated. Where anomalies are inconsistent with the predicted threat picture, the inconsistency should trigger program reanalysis rather than proceeding on prior assumptions.

What are the ILI data quality indicators that should be verified before accepting a run?

The first checks are tool performance verification against the specification for each anomaly type in the run, odometer consistency across the run, and confirmation that reported tool speed remained within the acceptable range throughout. For individual anomalies of interest, reviewing the raw sensor data rather than only the processed output provides a check on whether reported feature characteristics are consistent with the sensor signatures. Anomalies that approach threshold dimensions warrant independent analysis rather than reliance solely on vendor classification.

When is a combined ILI and direct assessment approach warranted?

Combined approaches are appropriate when a segment has multiple active threats that no single ILI technology can fully address, when ILI is partially infeasible along a route segment, or when an ILI run identifies anomaly indications that require ground verification to characterize. ICDA can also be used as a confirmatory midpoint assessment between full ILI cycles on gas pipelines, extending the effective interval while maintaining documented coverage of the internal corrosion threat.

Related Regulations and Resources

• Pipeline Regulations Hub (parent hub)
• 49 CFR 192.917 — Threat Identification for Gas Transmission Pipelines
• 49 CFR 195.452 — Integrity Management for Hazardous Liquid Pipelines
• Pipeline Integrity Management Overview
• PHMSA Enforcement Trends

---

Need a better way to navigate regulation context? Kondwit connects PHMSA inspection questions and enforcement cases to specific assessment method requirements so you can see exactly where audit focus lands. Learn more

---

See how Kondwit connects regulatory sections, enforcement history, and inspection questions. Request a demo

---

Matthew Brown, PE Licensed Professional Engineer | 15+ years pipeline integrity and compliance experience

Matthew Brown is a pipeline integrity engineer specializing in integrity management program development, regulatory compliance, threat assessment, and audit preparation. He has supported operators across transmission, distribution, and hazardous liquid systems with program development, documentation review, and PHMSA audit support.

This content is for informational and educational purposes only. It does not constitute engineering services, legal advice, or a professional engineering opinion. Operators should consult qualified professionals for system-specific compliance decisions.