Risk-Based Inspection Assessment for Pressure Safety Valves at a Canadian LNG Facility
TES Canada completed a pilot API 581 Part 5 RBI assessment for 60+ pressure safety valves at a Canadian LNG facility, ranking PSV risk and identifying opportunities to optimize inspection intervals and move toward risk-informed integrity management.
Operational Context
A Canadian LNG facility operator operated a population of more than 60 pressure safety valves across a pretreatment process area. PSV inspection and overhaul programs at the facility were governed by fixed calendar-based intervals — a common approach that applies the same inspection frequency regardless of valve service severity, historical performance, or actual risk contribution.
The operator recognised that a one-size-fits-all inspection interval regime creates inefficiency in both directions: it can impose unnecessary cost and operational disruption on low-risk valves while failing to differentiate higher-priority devices that warrant closer attention. A pilot Risk-Based Inspection study was commissioned to determine whether inspection and bench overhaul intervals could be optimised using a structured, API 581-aligned risk methodology, while maintaining full compliance with safety, operation-on-demand reliability, and applicable regulatory requirements including CSA B51 and CSA Z276.
Engineering & Integrity Challenge
Pressure safety valves are the last line of defence against overpressure events. Any methodology applied to PSV inspection interval optimisation must be technically defensible, aligned with recognised standards, and able to demonstrate that safety and reliability are maintained — not compromised — by the proposed changes.
The specific technical challenges included quantifying the probability that a valve fails to open on demand, quantifying the leakage risk between inspection intervals, and assessing the consequence of failure for each valve relative to the protected equipment and process. Where full consequence data was unavailable, a semi-quantitative approach was required without sacrificing defensibility. The assessment also needed to produce practical, actionable interval recommendations rather than theoretical rankings alone.
Why the Situation Was Complex
PSV failure modes are distinct from conventional pressure equipment degradation. The primary concerns are failure to open on demand (FTOOD) and leakage — both of which are probabilistic in nature and influenced by valve type, service severity, process fluid characteristics, and historical overhaul performance. These failure modes require Weibull-based probability modelling rather than conventional corrosion-rate approaches.
Consequence assessment for PSVs depends on the characteristics of the protected equipment and process, not the valve itself. Where detailed consequence data for the downstream protected system was incomplete, conservative semi-quantitative consequence rankings were required — introducing conservatism that the assessment needed to manage transparently.
The pilot scope covered more than 60 valves across varying service conditions, valve types, and inspection histories. Maintaining consistency of methodology across a diverse valve population while accommodating individual data gaps required careful data management and assessment discipline.
Regulatory alignment with CSA B51 and CSA Z276 added a compliance dimension that had to be addressed explicitly — any interval changes needed to be supportable under the applicable Canadian pressure equipment regulatory framework, not just technically justified under API 581.
TES Engineering Thinking
TES Canada applied the API 581 Part 5 Section 6 methodology for pressure relief devices — the recognised industry standard for risk-based PSV inspection planning.
Data Gathering and PSV Asset Register Development: TES Canada began by developing a structured PSV dataset covering all valves in the pilot scope. Inspection and overhaul history was reviewed, service severity and valve type classifications were established, and demand scenarios for each valve were identified. Data gaps were documented and addressed through conservative engineering assumptions where direct data was unavailable.
Probability of Failure Assessment: Failure to open on demand probability was assessed using Weibull-based modelling consistent with API 581 Part 5. Leakage probability was assessed separately as a function of valve type, service severity, and time since last overhaul. Both failure mode probabilities were calculated as time-series functions to project how risk evolves between inspection events.
Consequence Assessment: Consequence of failure was assessed for each valve relative to the protected equipment and process. Where full quantitative consequence data was available, API 581 consequence methodology was applied directly. Where data was incomplete, semi-quantitative consequence rankings were assigned using engineering judgement and conservative assumptions — ensuring the assessment remained defensible without understating risk.
Risk Calculation and Interval Optimisation: Time-series risk was calculated for each valve by combining probability and consequence. Results were compared against the client's risk acceptance criteria and risk matrix. Valves were ranked by risk and inspection urgency. The assessment identified: valves with potential for inspection or servicing interval optimisation where risk remained within acceptable limits over an extended period; valves requiring continued shorter intervals pending consequence data refinement; and a group of higher-priority valves warranting near-term attention.
Future Recommendations: TES Canada recommended the use of pretesting during future PSV servicing to improve inspection effectiveness and build a richer historical dataset for subsequent reassessment. The assessment also recommended future software tool development to enable dynamic risk reassessment as new inspection and overhaul data becomes available — establishing a scalable platform for applying PSV RBI across the wider facility.
Technical Approach
Practical Outcome
TES Canada delivered a structured PSV RBI dataset covering the full pilot scope, with each valve ranked by risk and inspection urgency. The assessment identified more than half of the PSV population as lower-risk devices with potential for inspection or servicing interval optimisation — reducing unnecessary overhaul frequency without compromising safety or regulatory compliance.
A group of higher-priority valves was identified as warranting continued shorter inspection intervals, either due to service severity, historical performance, or conservative consequence assumptions pending data refinement. Selected valves with potential for moderate interval relaxation were identified with supporting technical justification.
The pilot assessment established a scalable, defensible framework for extending PSV RBI to the wider facility. It supported the operator's transition from time-based inspection — where all valves are treated equally regardless of risk — to a risk-informed program that directs resources to where they matter most. Recommendations for pretesting during servicing and future software-enabled reassessment provided a clear development pathway for the ongoing program.
Lessons Learned
PSV consequence data is often held in process safety documentation that is separate from inspection records. Early engagement with process safety and operations teams to gather protected-equipment consequence data significantly improves assessment quality and reduces the conservatism required in semi-quantitative approaches.
Weibull-based FTOOD modelling is only as reliable as the inspection and overhaul history behind it. A pilot assessment that identifies data gaps and recommends pretesting during future servicing creates compounding value — each subsequent overhaul cycle improves the dataset and the accuracy of the next reassessment.
Regulatory alignment must be addressed explicitly, not assumed. Demonstrating that risk-based interval changes are supportable under CSA B51 and CSA Z276 is a distinct step from demonstrating technical defensibility under API 581 — both are required for the assessment to be actionable.
A pilot scope is the right entry point for PSV RBI in a large facility. It validates the methodology, builds organisational confidence, identifies data improvement priorities, and produces a reusable framework — all at a fraction of the cost of a facility-wide first assessment.
Facing a Similar
Integrity Challenge?
TES brings practical engineering thinking to the integrity challenges that aging infrastructure and operational pressure create. If this case experience resonates with a challenge you are navigating, a technical discussion is the right first step.