New Directions in Human Reliability Analysis

1
New Directions in Human Reliability Analysis

• Ronald Laurids Boring, PhDHuman Factors
  Scientist, Idaho National LaboratoryVisiting
  Scientist, Halden Reactor Project
• Vinh Dang, PhD
• Head, Risk and Human Reliability Group
• Paul Scherrer Institute

2
Human Reliability Analysis (HRA)

• Reminder Definition of HRA
• The use of systems engineering and human factors
  methods in order to render a description of the
  human contribution to risk and to identify ways
  to reduce that risk
• Typically considered to have from two to four
  phases

3
Whats New in HRA?

• Emphasis on data scrutability
• Use of simulation and modeling
• Use of dynamic HRA
• Emphasis on HRA for design

4
Data Scrutability

• Earlier HRA methods have not always been
  carefully validated
• The PSF multipliers and overall quantification
  may not have drawn heavily on human performance
  data sources
• A two-pronged problem
• Disconnect between human factors and HRA, such
  that most empirical results from human factors do
  not readily map to HRA
• Both are interested in probabilities Human
  factors reports significance levels (p lt .05),
  but not always the size of the performance
  enhancement or degradation (?HEP)
• Many HRA methods draw heavily on expert
  estimation to determine either the PSF
  multipliers or the overall HEP
• A normative model of risk should not be based on
  somebodys best guess!

5
Use of Simulation and Modeling

• Put the virtual back in reality!
• Simulators real humans virtual environments
• Simulation virtual humans virtual
  environments
• Human performance testing/determination of HEPs

6
Use of Dynamic HRA

• Dynamic PRA/PSA has been discussed for years
• Attempt to model how risk and safety change over
  time and contexts
• One PRA/PSA model fits all cases fails to account
  for emerging situations or even unanticipated
  situations
• Dynamic HRA has also been suggested
• Need to account for the dynamic progression of
  human behavior leading up to and following an
  event, accident, or incident
• Need to capture the dynamic decisions that can
  dramatically change the course of an event

7
Emphasis on HRA for Design

• Increasingly, human reliability needs to go
  beyond being a diagnostic tool to become a
  prescriptive tool
• US NRC and nuclear industry are looking at new
  designs for control rooms and want plants
  designed with human reliability in mind, not
  simply verified after the design is completed
• NASA has issued strict Human-Rating Requirements
  (NPR 8705.2) that all space systems designed to
  come in contact with humans must demonstrate that
  they impose minimal risk, they are safe for
  humans, and they maximize human reliability in
  the operation of that system
• How do we make reliable human systems?
• Design
• Test
• Model

classic human factors
human reliability analysis
8
Human Reliability Design Triptych
9
Three Generations of HRA?
10
Counter Arguments

• Do you really mean to argue that cognitive
  modeling can get us HRA?
• Cognitive modeling has clear limitations
• But, it is getting to the point where it can
  realistically model human decision making
• Also proving successful at modeling PSFs
• E.g., work by INL to map VACP to SPAR-H
  MicroSaint modeling of SPAR-H
• Important to baseline virtual performance to
  actual human performance
• Marriage of simulators and simulation
• Cautious extrapolation to new domains

11
More Counter Arguments

• Does HRA really need to be dynamic?
• New directions in HRA like resilience engineering
  are frustrated with the static nature of HRA/PRA
  models, which do not account for the full
  spectrum of human performance
• Trying to model human performance in a simulator
  or simulation reveals that current HRA models do
  not account for
• Evolving nature of PSFs (e.g., fatigue increases
  over time)
• PSF latency (e.g., stress does not instantly
  disappear just because a stressful event
  disappears)
• Complex dependencies between concurrent, parallel
  activities

12
Final Counter Arguments

• Do we need new generations of HRA?
• It offers some clear advantages
• Better understanding of the data behind human
  performance
• Ability to answer new questions
• Predict performance in novel domains
• Answer design questions using HRA as a human
  factors tool
• Create richer models for reconstructions of past
  events

13
The End, My Friend

• New HRA does not render other HRA obsolete
• Widely used first generation methods like THERP
  are still useful tools with a solid data basis
• Second generation methods offer good insights on
  PSFs and quantification
• Should continue to be developed
• Trickle up effect to third generation HRA
• New HRA is a framework that
• Improves HRA data quality
• Improves predictive ability (and applications) of
  HRA
• Needs facilities like HAMMLAB simulator studies
  to provide data