Description
Estimate a human error probability by multiplying a nominal rate for diagnosis or action by multipliers for eight performance-shaping factors, then adjust for dependence and uncertainty.
SPAR-H is the simplest widely used multiplier-based human reliability analysis. A human failure event is classed as diagnosis (nominal error probability 0.01), action (0.001) or both. Eight performance-shaping factors are each rated at defined levels with tabled multipliers: available time, stress, complexity, experience and training, procedures, ergonomics and HMI, fitness for duty and work processes. For example, low experience multiplies diagnosis error by 10 and action error by 3, while high experience halves it. The nominal probability is multiplied by the product, with a correction if the result would exceed 1. Dependence on preceding failures is assessed with a matrix of crew, time, location and cues using THERP's dependence levels, and uncertainty is expressed with a constrained non-informative beta distribution. Petro-HRA adapts it for petroleum, adding full qualitative guidance and a single nominal rate of 0.01.
Attributes
- Discipline
- Human Factors & Error
- Loss type
- Siloed
- Stage
- Design, Evaluation
- Data
- Analytical / expert
- Effort
- Days
Steps
- Define the scenario and the human failure event, and decide whether it involves diagnosis, action or both.
- Take the nominal error probability for each part (0.01 for diagnosis, 0.001 for action).
- Rate each of the eight performance-shaping factors at its defined level and look up the multipliers.
- Multiply the nominal probability by the product of multipliers, applying the correction if the result exceeds 1.
- Assess dependence on preceding failures with the crew, time, location and cue matrix and compute the conditional probability.
- Derive uncertainty bounds from the beta distribution and report which factors drive the estimate.
Sources
- Handbook of Human Factors and Ergonomics (Salvendy & Karwowski, 2021)
Also known as
- SPAR-H
- Standardized Plant Analysis Risk-Human Reliability Analysis
- Petro-HRA
Categories
- Accessibility
- Human Factors & Error