ML19212A734

From kanterella
Jump to navigation Jump to search
04_15_Fire HRA Training_Atheana
ML19212A734
Person / Time
Issue date: 07/31/2019
From:
Office of Nuclear Regulatory Research
To:
D. Stroup 415-1649
Shared Package
ML19212A718 List:
References
Download: ML19212A734 (50)
v • d • e


Text

NRC-RES/EPRI FIRE HRA METHODOLOGY:

Task 12 - Fire HRA ATHEANA Approach to Detailed Fire HRA Quantification NRC-RES Fire PRA Workshop Module IV August 5-9, 2019 Rockville, MD

Course Overview

1. Introduction to HRA
2. Overview of the EPRI/NRC Fire HRA Guidelines
3. Identification and definition of fire human failure events
4. Qualitative analysis
5. Fire HRA Application Experience
6. Quantitative analysis a) Screening b) Scoping c) Detailed EPRI approach & ATHEANA (detailed)
7. Recovery analysis
8. Dependency analysis
9. Uncertainty analysis Fire HRA - ATHEANA Slide 2 Fire PRA Workshop 2019, Rockville, MD

ATHEANA - Outline

1. Introduction to ATHEANA
2. ASME/ANS PRA Standards Addressed
3. ATHEANA HRA Process
4. ATHEANA guidance for facilitating expert elicitation
5. ATHEANA - Whats Going To Be Different For Fire PRA?
6. Addressing Fire-Specific Issues With ATHEANA
7. Fire HRA Exercises Using ATHEANA Fire HRA - ATHEANA Slide 3 Fire PRA Workshop 2019, Rockville, MD

Introduction to ATHEANA ATHEANA is

- A Technique for Human Event ANAlysis

- A second-generation HRA method

- A development of NRC/RES and its contractors

- An input to NRCs Good Practices for Implementing Human Reliability Analysis (HRA), April 2005 ATHEANA is documented in:

- NUREG-1624, Rev. 1, Technical Basis and Implementation Guidelines for A Technique for Human Event Analysis (ATHEANA),

May 2000.

- NUREG-1880, ATHEANA Users Guide, June 2007.

Fire HRA - ATHEANA Slide 4 Fire PRA Workshop 2019, Rockville, MD

Introduction to ATHEANA (continued)

ATHEANA is

- A knowledge-base* for (mostly) at-power, post-initiator HFEs, including:

Relevant psychological literature Supporting analyses of historical events

- A multidisciplinary framework for understanding human error

- An HRA process (including detailed guidance for performing qualitative analysis)

- A search scheme for HFEs (including errors of commission)

- A quantification approach Also, ATHEANA provides a basis for performing retrospective analysis of historical events (including example analyses).

But, different knowledge bases* can be used or substituted.

Fire HRA - ATHEANA Slide 5 Fire PRA Workshop 2019, Rockville, MD

Multidisciplinary framework Error- PRA Forcing Human Error Logic Context Models Plant Design, Risk Operations Performance Error Unsafe Human Failure Shaping Management and Mechanisms Actions Events Factors Maintenance Decisions Plant Scenario Conditions Definition Fire HRA - ATHEANA Slide 6 Fire PRA Workshop 2019, Rockville, MD

Underlying model of operators behavior

Response

Implementation Human-System Monitoring/ Situation Response Interface Detection Assessment Planning I & C System (Plant Automation) Knowledge/

Situation Model Mental Model Internal to Operators Fire HRA - ATHEANA Slide 7 Fire PRA Workshop 2019, Rockville, MD

Introduction to ATHEANA (continued)

The basic premise of ATHEANA:

- People behave rationally, even if reason for an action (or inaction) is wrong.

- Often, when people make errors, they are set up.

- People can be set-up by contexts that can create the appearance that the wrong response is correct when, in fact, it is not.

Analyses of operating experience (particularly events with serious consequences) support this view, e.g.:

- Nuclear power plant events (e.g., TMI 2, Browns Ferry, Chernobyl)

- Incidents from a variety of other technologies (e.g., aviation, medicine, chemical processing, maritime)

Fire HRA - ATHEANA Slide 8 Fire PRA Workshop 2019, Rockville, MD

Introduction to ATHEANA (continued)

Across industries, the following contextual factors often have been involved in serious events:

1. The plant behavior is outside the expected range (as represented by procedures, training, and traditional safety analyses).
2. The plants behavior is not understood.
3. Indications of the actual plant state and behavior are not recognized (sometimes due to instrumentation problems).
4. Prepared plans or procedures are not applicable or helpful for the specific plant conditions.

Fire HRA - ATHEANA Slide 9 Fire PRA Workshop 2019, Rockville, MD

Introduction to ATHEANA (continued)

Consequently, the principal motivators for developing ATHEANA were:

1. HFEs modeled in most HRA/PRAs are not consistent with the roles played by operators in actual operational experience (including errors of commission and dependencies between actions).
2. The accident record and advances in behavior sciences both support a stronger focus on context.
3. Recent advances in psychology ought to be used and integrated with the disciplines of engineering, design, operations and training, human factors, and PRA in modeling HFEs.

Fire HRA - ATHEANA Slide 10 Fire PRA Workshop 2019, Rockville, MD

Introduction to ATHEANA (continued)

Overall, the goal of ATHEANA quantification approach is to:

- Develop an operational story (including plant conditions, operational aids such as procedures, and other influencing factors) to explain why an operator could failure to perform an action

- Explain and refine the operational story with plant-specific experts

- Use the expert judgment of plant-specific personnel (especially operator trainers) to develop failure probabilities for HFEs that require detailed HRA quantification (facilitated by the HRA analyst)

Fire HRA - ATHEANA Slide 11 Fire PRA Workshop 2019, Rockville, MD

Introduction to ATHEANA (continued)

Key characteristics are:

- Focuses on the error-forcing context (i.e., the context that sets up operators), but also addressed the nominal context

- Uses a structured search for problem scenarios (i.e., error-forcing contexts) and associated unsafe actions (i.e., operator failures)

- Links plant conditions, performance shaping factors (PSFs) and human error mechanisms through the context

- Is experience-based, both in its development and application (e.g., uses knowledge of domain experts such as operators, pilots, trainers)

- Uses multidisciplinary approach and underlying cognitive model of operator behavior

- Explicitly considers operator dependencies (including recovery actions) by developing entire accident sequences

- Uses a facilitator-led, expert elicitation approach for quantification (that allows the plant-specific experience and understanding from operators, operator trainers, and other operations experts to be directly reflected)

Fire HRA - ATHEANA Slide 12 Fire PRA Workshop 2019, Rockville, MD

Introduction to ATHEANA (continued)

Example ATHEANA applications:

- HRA/PRAs in a prospective analysis of regulatory and industry issues such as pressurized thermal shock (PTS) (3 plants -

Oconee, Beaver Valley, Palisades)

- International HRA Empirical Study (Steam Generator Tube Rupture and Loss of Feedwater scenarios)

- DOEs license application for Yucca Mountain waste repository (preclosure facility)

- Qualitative analyses of spent fuel handling (misloads and cask drops) (NUREG/CR-7016 and -7017, February 2012)

- Retrospective event analyses and development of a knowledge-base for fire-specific human performance issues (NUREG/CR - to be published)

- HRA/PRA to evaluate design features of a facility to dismantle chemical weapons Fire HRA - ATHEANA Slide 13 Fire PRA Workshop 2019, Rockville, MD

ANS/ASME RA-Sa-2009 Requirements for Fire - At Power High Level Requirements for HEP Quantification ATHEANA includes a fully capable detailed HRA quantification approach that satisfies requirements such as:

- Part 2, HLR-HR-F: Human failure events shall be defined that represent the impact of not properly performing the required responses, in a manner consistent with the structure and level of detail of the accident sequences

- Part 2, HLR-HR-G: The assessment of the probabilities of the post-initiator HFEs shall be performed using a well-defined and self consistent process that addresses the plant-specific and scenario-specific influences on human performances, and addresses potential dependencies between human failure events in the same accident sequence

- Part 4, HLR-HRA-B: The Fire PRA shall include events where appropriate in the Fire PRA that represent the impacts of incorrect human responses associated with the identified human actions

- Part 4, HLR-HRA-C: The Fire PRA shall quantify HEPs associated with incorrect responses accounting for the plant-specific and scenario-specific influences on human performance, particularly including the effects of fire and supporting level requirements such as:

- Part 2, SRs HR-F1, HR-G3, HR-G7, HR-G8; Part 4 SRs, HRA-B1 [Note 1] and HRA-C1 Fire HRA - ATHEANA Slide 14 Fire PRA Workshop 2019, Rockville, MD

The ATHEANA HRA process Step 1: Define and interpret issue of concern Step 2: Define scope of analysis Step 3: Describe base case scenarios Step 4: Define HFEs and unsafe actions (UAs)

Step 5: Identify potential vulnerabilities Step 6: Search for deviations from base case Step 7: Evaluate recovery potential Step 8: Quantification Step 9: Incorporation into PRA Fire HRA - ATHEANA Slide 15 Fire PRA Workshop 2019, Rockville, MD

Steps in the ATHEANA Process Fire HRA - ATHEANA Slide 16 Fire PRA Workshop 2019, Rockville, MD

The ATHEANA HRA process Not all of these steps are needed for every HRA/PRA job.

For fire HRA/PRA, certain steps will not need to be performed by ATHEANA, e.g.,

- NUREG/CR-6850 [EPRI 1011989] and the ANS/ASME PRA Standard already address Steps #1 and #2 (i.e., define and interpret the issue of concern, define the scope of analysis)

- Deviations from the base case scenario (i.e., Step #6) are usually not needed for fire; most fire scenarios are generally challenging enough for operators that we do not have to look for even more unusual conditions So, later when we talk about ATHEANA steps, well highlight those needed specifically for fire HRA/PRA.

Fire HRA - ATHEANA Slide 17 Fire PRA Workshop 2019, Rockville, MD

ATHEANA guidance for facilitating expert elicitation ATHEANA uses an expert elicitation approach to develop failure probabilities for HFEs:

- Described in NUREG-1880, ATHEANAs Users Guide

- Based on previous expert elicitation approaches, especially:

NUREG/CR-6372, Recommendations for Probabilistic Seismic Hazard Analysis:

Guidance on Uncertainty and Use of Experts (1997) (a.k.a., NRCs SSHAC report)

- ...and consists of:

A six-step process, leading to quantification of HFE (and its distribution)

Description of who the experts should be General guidance for the facilitator Guidance on addressing uncertainty, controlling for unintentional bias in experts, information to discuss, how to lead discussions and build distributions Guidance on how to educate experts on probabilities and context Guidance on how to build a consensus HEP and its distribution, and perform sanity checks Fire HRA - ATHEANA Slide 18 Fire PRA Workshop 2019, Rockville, MD

ATHEANA - Whats going to be different for fire PRA?

1. NUREG/CR-6850 [EPRI 1011989] and supporting documents indicate the need for adjustments for a fire-specific knowledge-base (e.g., fire-specific human performance issues).
2. EOCs are limited to those stated in the ASME/ANS PRA Standard.
3. Many Fire HRA Guidelines qualitative analysis tasks overlap; may already be performed or started before detailed quantification is performed.
4. The fire context may already be sufficiently challenging for operators; ATHEANA steps and activities related to finding an error-forcing context may not be needed.

Fire HRA - ATHEANA Slide 19 Fire PRA Workshop 2019, Rockville, MD

Mapping ATHEANA process steps to Fire HRA Guidelines process ATHEANA Process Step Fire HRA Guideline Process Step Steps 1 & 2: Define issue & scope Defined by fire PRA & its scope of analysis - no additional work needed of analysis Step 4: Define HFEs and unsafe Covered* by Chapter 3: Identification and actions (UAs) Definition Steps 3 & 5: Describe PRA scenario Some additional information needed for

& assess human performance detailed HRA; but, mostly covered by information, etc. Chapter 4: Qualitative Analysis Step 6: Search for deviation Probably not needed; fire scenarios are scenarios already deviations Step 7: Assess potential for Similar to Chapter 6: Recovery recovery Step 8: Quantification (explicitly Different approach than scoping trees addresses dependencies &

develops uncertainty distributions) (Chapter 5) or CBDT (Appendix C);

different approach to dependency &

uncertainty (Chapters 7 & 8)

Fire HRA - ATHEANA Slide 20 Fire PRA Workshop 2019, Rockville, MD

ATHEANA HRA process - highlighting needs for fire HRA Step 1: Define and interpret issue of concern Step 2: Define scope of analysis Step 3: Describe base case scenarios*

Step 4: Define HFEs and unsafe actions (UAs)*

Step 5: Identify potential vulnerabilities*

Step 6: Search for deviations from base case*

Step 7: Evaluate recovery potential Step 8: Quantification Step 9: Incorporation into PRA

  • Previous fire PRA tasks provide a start on these ATHEANA tasks; for example qualitative analysis is continuous through detailed HRA quantification Fire HRA - ATHEANA Slide 21 Fire PRA Workshop 2019, Rockville, MD

Additional ATHEANA needs for fire HRA

1. Some additional qualitative analysis to support Steps 3, 5, (6), 7, and 8, including:
  • Information collection
  • Interviews of operator trainers
2. ATHEANA approach for quantification and recovery
  • With dependency considerations embedded
  • With uncertainty distribution being explicitly developed as part of quantification
3. Adjustments to knowledge-base (per considerations in NUREG/CR-6850 [EPRI 1011989] and others)

Fire HRA - ATHEANA Slide 22 Fire PRA Workshop 2019, Rockville, MD

Additional ATHEANA needs for fire HRA (continued)

So, in this presentation, we will only discuss the following steps in the ATHEANA process:

- Step 3: Describe the base case scenario*

- Step 5: Identify potential vulnerabilities*

- Step 6: Search for deviations from base case (often not needed)

- Step 7: Evaluate recovery potential

- Step 8: Quantification As for the entire process in applying the Fire HRA Guidelines, these steps are iterative.

Note: If Step 6 is needed, HFEs may need to be redefined (as in any HRA/PRA, if warranted by plant conditions, timing of plant behavior, etc.). But, Fire HRA Guidelines can address this situation without using Step 2 of ATHEANA explicitly.

Fire HRA - ATHEANA Slide 23 Fire PRA Workshop 2019, Rockville, MD

Step 3: Describe base case scenario (i.e., PRA scenario and its nominal context)

The base case scenario:

- represents most realistic description of expected plant and operator behavior for selected issue and initiator

- provides basis to identify and define deviations from such expectations (found in Step 6)

Ideally, base case scenario:

- has a consensus operator model (COM)

- is well-defined operationally

- has well-defined physics

- is well-documented

- is realistic Scenario description often based on FSAR or other well-documented analyses In practice, the available information defining a base case is usually less than ideal -

analysts must supplement information deficiencies or simply recognize them.

Fire HRA - ATHEANA Slide 24 Fire PRA Workshop 2019, Rockville, MD

Step 3: Describe base case scenario (continued)

Initial plant conditions Sequence of events and expected timing before and following reactor trip Plant system and equipment response What the operators will see

- usually trajectories of key plant parameters and indications Key operator actions during the scenario progression NUREG-1921, Section 4 is good resource for this step Fire HRA - ATHEANA Slide 25 Fire PRA Workshop 2019, Rockville, MD

Step 3: Examples of information sources Plant-specific FSAR (and other design basis documents)

Safety analyses (e.g., plant-specific, vendor)

Procedures (e.g., plant-specific EOPs, vendor, basis documents)

Operator experience (actual and simulator)

Operator training material and its background documentation Plant staff, especially operators, operator trainers, T-H experts Plant-specific and industry generic operating experience Again, NUREG-1921, Section 4 is a good resource Fire HRA - ATHEANA Slide 26 Fire PRA Workshop 2019, Rockville, MD

Step 5: Identify potential vulnerabilities Identify and characterize factors (e.g., performance shaping factors (PSFs)) that could contribute to crew performance in responding to the various accident scenarios

- Factors that might increase the likelihood of the HFEs and UAs of interest

- Helps focus later deviation searches Operators and trainers must play a role in this step

- directly or through question/answer sessions

- observation of simulator exercises (with relevant scenarios if possible)

Fire HRA - ATHEANA Slide 27 Fire PRA Workshop 2019, Rockville, MD

Ways to identify potential vulnerabilities Investigation of potential vulnerabilities due to biases in operator expectations (training, experience)

- review training materials, interview trainers, operators Understanding of base-case scenario timeline and any inherent difficulties associated with required response Identification of operator-action tendencies based on

- standardized responses to indications of plant conditions

- informal rules Evaluation of formal rules and EOPs

- critical decision points, ambiguities, sources of confusion, timing mismatches, special cases such as preemptive actions, etc.

Fire HRA - ATHEANA Slide 28 Fire PRA Workshop 2019, Rockville, MD

Ways to identify potential vulnerabilities (continued)

Guidance given in NUREG-1921, Section 4, is good starting point Additional tools or guidance can be found in ATHEANA documentation, e.g.,

- NUREG-1624, Revision 1 and NUREG-1880:

Summary of operator tendencies (for off-normal plant conditions in PWRs and BWRs), e.g.,

- steam generator pressure too low or decreasing=> operators decrease steam dump (i.e., cooldown) or isolate tube rupture Examples of informal rules (e.g., believe your indications, protect pumps (i.e., stop if no lube oil pressure))

Scenario characteristics that are challenging to operators (e.g., missing information, impasses, tradeoffs, double binds)

Parameter characteristics that are challenging (e.g., small change in parameter, slow rate of change in parameter, one or more false indications)

Fire HRA - ATHEANA Slide 29 Fire PRA Workshop 2019, Rockville, MD

Ways to identify potential vulnerabilities (continued)

Additional tools and guidancein ATHEANA (continued)

- NUREG-1624:

Table 9.18, Examples of information problems (e.g., display failures, human factoring issues)

Table 9.19, Physics algorithms in instruments that can confuse operators (e.g., drive versus stem position for valve position indication)

Table 9.15b and 9.16b, scenario or parameters characteristics and associated error mechanisms, error types, and potential PSFs Appendix A: Retrospective analyses of six events Appendices B - E: ATHEANA example applications (e.g., SLOCA, loss of service water)

- NUREG-1880:

Table 3.5-1, Relevant time frames for large LOCA and Loss of Main Feedwater (MFW) example scenarios Section 3.5.2.2, Descriptions of PSFs (and associated discussion)

Appendix A: Example of an EOP flowchart for loss of MFW scenario Fire HRA - ATHEANA Slide 30 Fire PRA Workshop 2019, Rockville, MD

Step 6: Search for deviations from base case scenario Identify deviations from base case likely to result in risk-significant unsafe acts Deviations are plant behaviors or conditions that set up unsafe actions by creating mismatches between the proposed plant behavior and:

- operators knowledge, expectations, biases and training

- procedural guidance and timing ATHEANA search schemes guide analysts to find real deviations in plant behavior and conditions

- not just false perceptions in the operators minds Fire HRA - ATHEANA Slide 31 Fire PRA Workshop 2019, Rockville, MD

Step 6: Four search schemes for deviation scenarios Identify deviations from the base case scenario using HAZOP guide words to discover troublesome ways that the scenario may differ from base case

- more, less, quicker, slower, repeat ...

Identify deviations for vulnerabilities associated with procedures and informal rules

- e.g., changes in timing, sequencing of decision points, etc.

Identify deviations caused by subtle failures in support systems

- cause problems for operators to identify whats happening Identify deviations that can set up operator tendencies and error types leading towards HFEs/UAs of interest Fire HRA - ATHEANA Slide 32 Fire PRA Workshop 2019, Rockville, MD

Step 7: Evaluate potential for recovery Possibility of recovering from UAs is considered in this step; similar to recovery analysis when using other detailed HRA quantification methods However, for ATHEANA, recovery always considers both the complete EFC and the occurrence of other UA(s)

- in qualitative analysis

- in quantification (i.e., probability of failed recovery is conditional on probabilities of other operator failures and successes)

Deviation description is extended to include the scenario characteristics up to the last opportunity for recovery Performance of this step linked with quantification - iteration between these steps is likely Fire HRA - ATHEANA Slide 33 Fire PRA Workshop 2019, Rockville, MD

Guidance for recovery analysis Define the possible recovery action(s) given the initial error corresponding to the HFE/UA has occurred Consider the time available to diagnose the need for and perform the recovery action so as to avoid a serious or otherwise undesired condition Identify the existence and timing of cues as well as how compelling the cues are that would alert the operators to the need to recover and provide sufficient information to identify the most applicable recovery action(s)

Identify the existence and timing of additional resources (e.g., additional staff, special tools), if necessary, to perform the recovery Fire HRA - ATHEANA Slide 34 Fire PRA Workshop 2019, Rockville, MD

Step 8: Quantification Very structured, facilitator led, expert opinion elicitation process

- leads to consensus distributions of operator failure probabilities Considerations in elicitation process (covered in NUREG-1880):

- Forming the team of experts (include experts familiar with important relevant factors during fire conditions, operator trainers, etc.)

- Controlling for biases when performing elicitations

- Addressing uncertainty Fire HRA - ATHEANA Slide 35 Fire PRA Workshop 2019, Rockville, MD

ATHEANA quantification: Asks experts two questions

1. Does the operational story make sense?
  • given the specific PRA scenario or sub-scenario
  • given what is known about operators and operations at this plant
2. What is the likelihood that operators will fail as described in the operational story?

Fire HRA - ATHEANA Slide 36 Fire PRA Workshop 2019, Rockville, MD

Basic formulation for quantification process P (HFElS) = P(EFCilS) x P(UAjlEFCi,S) ij HFEs are human failure events modeled in PRA

- Modeled for a given PRA scenario (S)

- Can include multiple unsafe actions (UAs) and error-forcing contexts (EFCs)

First determine probability of the EFC (plant conditions and PSFs) being addressed Determine probability of UA given the identified EFC If multiple EFCs identified, then quantify a UA given each EFC separately Fire HRA - ATHEANA Slide 37 Fire PRA Workshop 2019, Rockville, MD

Six steps to quantification process

1. Discuss HFE and possible influences / contexts using a factor checklist as an aid
2. Identify driving influencing factors and thus most important contexts to consider
3. Compare these contexts to other familiar contexts and each expert independently provide the initial probability distribution for the HEP considering:

- Likely to fail ~ 0.5 (5 out of 10 would fail)

- Infrequently fails ~ 0.1 (1 out of 10 would fail)

- Unlikely to fail ~ 0.01(1 out of 100 would fail)

- Extremely unlikely to fail ~ 0.001 (1 out of 1000 would fail)

Fire HRA - ATHEANA Slide 38 Fire PRA Workshop 2019, Rockville, MD

Six steps to quantification process (continued)

4. Each expert discusses and justifies his/her HEP estimate
5. Openly discuss opinions and refine the HFE, associated contexts, and/or HEPs (if needed) -

each expert independently provides HEP (may be the same as the initial judgment or may be modified)

6. Arrive at a consensus HEP for use in the PRA Fire HRA - ATHEANA Slide 39 Fire PRA Workshop 2019, Rockville, MD

Addressing fire-specific issues with ATHEANA ATHEANA should be applied in the same way for fire HRA, as for any other HRA/PRA However, the fire-specific operator performance issues should be considered in performing ATHEANA steps (e.g.,

identifying potential vulnerabilities, quantification)

- Again, Section 4 provides good basis for issues to address and tools for performing qualitative analysis tasks (e.g., how to collect and interpret timing information, fire-specific issues with respect to use of procedures)

Plus, some of the information needed to apply ATHEANA may be collected and analyzed already in order to have used either the screening values or scoping approach provided in the Fire HRA Guidelines Fire HRA - ATHEANA Slide 40 Fire PRA Workshop 2019, Rockville, MD

Addressing fire-specific issues with ATHEANA (continued)

Additional guidance/discussion on addressing operator response to spurious indications is provided in NUREG-1921, Appendix C, e.g.,

- Development of uncertainty ranges in timing estimates (as discussed in Section 4.6.2) can (and have) been developed directly with expert elicitation

- EOCs due to spurious indications (both recovery, as for scoping approach, and initial failure)

- Impact of spurious indications as distractions (see Section 4.10)

Fire HRA - ATHEANA Slide 41 Fire PRA Workshop 2019, Rockville, MD

Example qualitative analysis results - Chapter 4 In applying the Fire HRA Guidelines, the following are examples of information already collected and/or analyzed:

- Procedures used in fire scenarios

- Usage of procedures

- Potential fire effects and their impacts on human performance

- Fire PRA scenarios with associated equipment and indication failures

- Possible crew responses to fire scenarios Errors of Commission Errors of Omission Fire HRA - ATHEANA Slide 42 Fire PRA Workshop 2019, Rockville, MD

Examples of additional qualitative analysis to support ATHEANA

1. Identify:

- important decision points or branching, and other possible places in procedures where operators may make different choices

- plant-specific informal rules and other guidance that may supplement or slightly deviate from relevant procedural guidance

- tradeoffs (e.g., impromptu choices between alternatives) or other difficult decisions that operators may need to make

- potential situations where operators may not understand the actual plant conditions (e.g., spurious indications)

- different ways by which an HFE could occur, starting with the fire PRA scenario description, different procedural paths or choices, and the reasons for these different choices Fire HRA - ATHEANA Slide 43 Fire PRA Workshop 2019, Rockville, MD

Examples of additional qualitative analysis to support ATHEANA (continued)

2. Develop:

- insights from training, experience, or demonstration of fire-related operator actions (in- and ex-MCR), including use of specialized equipment

- timelines or other ways of representing the time sequencing of events in fire scenarios

3. Objective or final result of ATHEANA qualitative analysis:

- A full operational scenario description, or operational story, including accident progression and as many bells and whistles as are reasonable, such that operator trainers can put themselves into scenario Because, in quantification, you will be asking them, what would your crews do in this situation?

Fire HRA - ATHEANA Slide 44 Fire PRA Workshop 2019, Rockville, MD

Examples of additional qualitative analysis to support ATHEANA (continued)

The resulting operational scenario description may include:

- Additional plant conditions that will need to be quantified as part of the HFE (unless accident sequence analyst wants to revise event trees or fault trees).

- Distinctions on timing of plant behavior (that might need to be addressed as part of the HFE, unless logic is revised).

- Instrument or indication issues (including failures) that will need to be reflected (for fire, might be explicitly part of PRA model, or may not).

- Different possible procedure paths or response strategies that operators might rationally take.

- Reasons why operators might take different procedure paths.

- Credible recovery actions.

Likely to need help from operational experts on the last three elements.

Fire HRA - ATHEANA Slide 45 Fire PRA Workshop 2019, Rockville, MD

RememberBasic quantification formula?

First, lets simplify; only one EFC for each scenario, S.

So, we have:

P (HFElS) = P(UAjlEFC,S) j S = Full operational story (might not be equivalent to PRA scenario)

UAs = Different procedure paths leading to undesired outcomes, and associated reasons for taking them EFCs = Plant conditions, behavior, PSFs, etc., that are not explicitly modeled in PRA, but needed to represent S Probability of each UA is conditional on EFC/S Fire HRA - ATHEANA Slide 46 Fire PRA Workshop 2019, Rockville, MD

ATHEANA - Iterating between qualitative analysis and quantification Development of operational scenario descriptions should be both for and by operational experts (e.g., trainers).

Even during quantification, the analyst should be alert to the need to modify, refine, and/or add details to the operational description of the scenario. For example:

- During quantification, very different failure probabilities are provided by the expert panel of trainers.

- When explaining answers, one trainer brings up a possible influence (e.g., a specific plant condition or equipment failure) that no one else has considered.

- Because everyone agrees to the validity and importance of this factor, the analyst either:

Has everyone include this factor in their quantification, or Defines a new HFE to address this newly defined scenario Fire HRA - ATHEANA Slide 47 Fire PRA Workshop 2019, Rockville, MD

ATHEANA - Iterating between qualitative analysis and quantification (continued)

Based on experience in applying ATHEANA, most of the effort is in identifying and developing the elements of an operational story that represents what the experts think is important to operator behavior.

Once this agreement is reached, reaching a consensus in final quantification by the operational experts is usually not difficult (if using the tools and techniques for facilitating expert elicitation, such as that given in the ATHEANA Users Guide.)

Fire HRA - ATHEANA Slide 48 Fire PRA Workshop 2019, Rockville, MD

ATHEANA - Addressing uncertainty in fire HRA/PRA Performed as usually would, i.e.,

- Expert elicitation process for quantification includes:

Detailed qualitative discussions to ensure all the available information (evidence) is brought to the table, shared, and agreed upon to the extent possible Detailed identification of the key factors contributing to aleatory and epistemic uncertainty

- The HEP developed for an HFE in a fire scenario (as for any other scenario) may be made up of combinations of distributions of multiple unsafe actions that have been evaluated separately.

- Individual distributions combined mathematically into a single distribution.

Fire HRA - ATHEANA Slide 49 Fire PRA Workshop 2019, Rockville, MD

Questions?

Fire HRA - ATHEANA Slide 50 Fire PRA Workshop 2019, Rockville, MD

Retrieved from "https://kanterella.com/w/index.php?title=ML19212A734&oldid=1902161"