Technical Overview of HAZOP and LOPA Integration:
A Detailed Approach to Risk Assessment
In the first part of this series, we discussed how HAZOP and LOPA are integrated to identify, assess, and manage risks in industrial processes. In this second post, we will explore the technical details behind these methodologies, focusing on how they determine severity and likelihood and how factors like initiating causes, propagating causes, and Independent Protection Layers (IPLs) contribute to the risk analysis process.
Here’s what you’ll discover in this article:
Severity and Likelihood in HAZOP and LOPA
In both HAZOP and LOPA, the severity of an event and the likelihood of its occurrence are key factors in assessing risk. These elements help prioritize hazards and determine the required level of risk reduction.
Severity
The severity of an event refers to the consequences if the hazard were to occur. It could involve different outcomes, such as:
- Injuries (e.g., fatalities, long-term health effects)
- Environmental damage (e.g., chemical releases, contamination)
- Financial impact (e.g., equipment damage, downtime)
The severity rating is determined based on the potential impact, with higher severity reflecting more significant consequences. For example, in a hazardous chemical release, the severity could range from minor injuries to catastrophic damage, depending on the amount and type of chemical released.
Likelihood
In HAZOP and LOPA, likelihood is quantified using the Probability of Failure on Demand (PFD). This is based on:
- The frequency of the initiating event (e.g., human error or equipment failure)
- The chance of that event propagating into a more severe consequence
- The probability of the scenario being enabled by certain conditions (e.g., startup or shutdown conditions)
By combining these factors, HAZOP and LOPA enable a comprehensive view of potential risks, helping to prioritize the most dangerous scenarios and determine what protective measures are necessary.
Determining the Likelihood of an Event Using HAZOP and LOPA
The HAZOP and LOPA methodology uses different steps to calculate the likelihood of a hazardous event occurring:
- Select credits based on industry-published data for each scenario.
- Determine the Unmitigated Risk (UMR Likelihood) by assessing the risk prior to applying any safeguards.
- Determine the Mitigated Risk (MR Likelihood) by assessing the risk after accounting for protection measures [Independent Protection Layers (IPLs)].
- Combine the UMR Likelihood and MR Likelihood to determine the final likelihood of the event occurring.
Determining the Unmitigated Risk (UMR Likelihood) Using HAZOP-LOPA
To calculate the Unmitigated Risk Likelihood (UMR) within the HAZOP-LOPA methodology, we assess several factors that contribute to the likelihood of a hazardous event occurring without any protective measures in place. These factors include:
- Initiating Cause: This is the event that starts the hazardous scenario, and can include:
- Human error (e.g., improper operation, failure to follow procedures)
- Instrument malfunction (e.g., sensor failure, control system error)
- Equipment failure (e.g., pump failure, valve leakage)
- External events (e.g., natural disasters, power outages)
- Human error (e.g., improper operation, failure to follow procedures)
- Propagating Cause: Once the initiating event occurs, there may be a propagating cause for the scenario to continue. This could include:
- Additional system failures, errors, or conditions that contribute to the risk.
- Note: It’s important to avoid “double jeopardy”—the idea that two independent failures are unlikely to occur simultaneously.
- Additional system failures, errors, or conditions that contribute to the risk.
- Enabling Events: Enabling events are time related conditions that must exist for the scenario to progress, such as:
- Environmental factors (e.g., extreme weather conditions, like lightning)
- Hidden failures (e.g., unnoticed sensor malfunctions)
- Environmental factors (e.g., extreme weather conditions, like lightning)
- Probability of Ignition: The probability that a hazardous mixture will ignite, should the scenario involve flammable materials. This includes:
- The probability that an ignitable mixture will be created (e.g., a leak or spill).
- The probability that an ignition source with sufficient energy will be present (e.g., spark, heat, or friction).
- The probability that an ignitable mixture will be created (e.g., a leak or spill).
Assumptions: This calculation assumes no ignition controls are in place, such as:
- Secondary containment to limit the spread of hazardous materials.
- Electrically rated or grounded equipment within containment areas.
- Secondary containment to limit the spread of hazardous materials.
- Probability of Worker Presence: This factor assesses the likelihood that workers will be in proximity to the hazard when the incident occurs. It is expressed as the Probability of Presence on Demand (PPD), which takes into account:
- Worker schedules, operating conditions, and proximity to the hazard at the time of the event.
Determine the Mitigated Risk Likelihood (MR Likelihood) Using HAZOP and LOPA
The Mitigated Risk Likelihood (MR) accounts for the reduced probability of an event occurring due to the presence of Independent Protection Layers (IPLs), which are safety measures designed to reduce risk. This step is integral to the HAZOP and LOPA process, as it helps to quantify the impact of safety controls in reducing overall risk.
Examples of IPLs include:
- Basic Process Control System (BPCS) Interlock
- Hardwired Interlock
Each IPL is assessed using its Probability of Failure on Demand (PFD). A low PFD indicates a highly reliable safeguard, while a high PFD suggests reduced reliability. When evaluating this, it must be remembered that PFD is based on exponents.
For example: A safeguard with a:
- PFD = 1E-2 (failure rate of once in 100 years)
– is more reliable than a –
- PFD = 1E-1 (failure rate of once in ten years)
Independent Layers of Protection (IPLs) in HAZOP and LOPA
Independent Protection Layers (IPLs) are safety measures designed to reduce the likelihood of a hazard or its consequences. To be effective in the HAZOP and LOPA study, an IPL must function independently from other systems and be adequately maintained.
Examples of IPLs include:
- Automatic shutdown systems (e.g., based on control systems)
- Pressure relief valves (mechanical safeguards)
- Operator intervention (manual safeguards)
- Detection systems (e.g., fire or gas detectors)
The reliability of an IPL is assessed using its PFD. A low PFD indicates high reliability, while a higher PFD suggests the safeguard is less dependable.
Evaluating IPLs in a HAZOP-LOPA study involves:
- Identifying IPLs: Listing all available safeguards.
- Assessing PFD: Determining the PFD value for each IPL.
- Crediting Safeguards: Effective IPLs reduce risk and are credited in the mitigated risk analysis. Safeguards that lack integration into safety programs (e.g., maintenance procedures) cannot be credited.
Combining Unmitigated and Mitigated Risk Likelihood:
The final Risk Likelihood is determined by combining the Unmitigated Risk Likelihood (UMR) and Mitigated Risk Likelihood (MR), then adjusting the combined value based on the severity of the event.
Example Final Risk Calculation:
- Unmitigated Risk Likelihood: -3
- Mitigated Risk Likelihood: -3 (credits from IPLs)
- Final Risk Score: The combined likelihood is adjusted based on the severity (e.g., severity = 6). In this example, the final risk is reduced to 0, which is within acceptable limits based on most company’s risk criteria.
Example Scenario: Risk Calculation with IPLs in HAZOP and LOPA
To demonstrate the application of severity, likelihood, and Independent Protection Layers (IPLs) in HAZOP and LOPA, let’s revisit an example scenario involving an alcohol distillation column. In this scenario, if the overhead valve in the distillation column closes unexpectedly, pressure may build up. If the pressure relief system cannot handle the excess pressure, an overpressure condition could result in equipment damage or hazardous material release.
This example will demonstrate how we calculate both unmitigated and mitigated risk likelihoods by analyzing initiating causes, propagating causes, and enabling events, as well as how the effectiveness of IPLs reduces risk.
Step 1: Severity: We previously determined that the severity of a vapor cloud explosion from the distillation column rupture is 6 on the severity scale (indicating multiple fatalities).
Step 2: Unmitigated Risk Likelihood: Without any safeguards in place, the unmitigated risk likelihood is determined by:
- The frequency of the initiating cause (e.g., operator error or valve malfunction)
- The enabling events (e.g., startup)
- The frequency of the initiating cause (e.g., operator error or valve malfunction)
In our example, the unmitigated likelihood has a rating of -3.
Step 3: Mitigated Risk Likelihood: Now, let’s credit the Independent Protection Layers (IPLs) that are in place:
- Basic Process Control System (BPCS) Interlock: -1 PFD
- Hardwired Interlock (with SIS calculations): -2 PFD
- Basic Process Control System (BPCS) Interlock: -1 PFD
The combined mitigated likelihood is -3, resulting in a final mitigated risk score of 0, which is within most companies acceptable limits.
In this second part of the HAZOP and LOPA series, we have explored how severity and likelihood are calculated, with a specific focus on the role of Independent Protection Layers (IPLs). These safeguards are essential in reducing risk, and their proper evaluation can help ensure the safety of industrial processes.
By combining HAZOP for qualitative analysis and LOPA for quantitative risk evaluation, organizations can comprehensively understand their risk profile and take appropriate steps to mitigate potential hazards. Integrating IPLs with a mechanical integrity program ensures that safeguards are considered and maintained effectively.
This concludes the second blog in our series. If you missed the first part, you could catch up by reading Part 1: Integrating HAZOP and LOPA for Effective Risk Assessment. Stay tuned for more insights on safety risk management in future posts!
