Fault Tree Analysis, FTA, Event Tree Consulting


Fault Tree Analysis

A fault tree analysis (FTA) is a systematic, deductive methodology for defining a single, specific undesirable event and determining all possible reasons that lead to the event. The undesired event is the top even in the fault tree diagram and generally represents a safety-related failure. However, any event can be analyzed with a fault tree. Fault tree analysis focuses on a subset of all possible system failures. Contrast with the FMEA, which analyzes all possible failure modes regardless of severity.

When properly applied, a FTA is extremely useful during the initial product design phases as an evaluation tool for driving preliminary design modifications. After product release, the FTA can be used as a troubleshooting tool. Through an FTA, a product can be evaluated from both a reliability and a fault probability perspective. From a reliability perspective, the FTA can estimate whether or not a product will meet performance reliability requirements. Using probabilistic evaluation, the FTA emphasis shifts in the likelihood of the occurrence of the undesired event.

The results of a FTA may be expressed qualitatively (minimum cut-sets, qualitative importance measures, and common cause potentials) or quantitatively (numeric probabilities of events, quantitative importance measures, and sensitivity evaluations).

FTA Features

  • Functional analysis of highly complex systems.
  • Allows observation of combined effects of simultaneous, non-critical events on top event.
  • Deductive analysis.
  • Graphical analysis.
  • Can consider human error as a cause of the top event.
  • Can consider software failures as causes of the top event.
  • Can be either qualitative or quantitative.

FTA Uses

  • Evaluation of safety requirements.
  • Evaluation of system reliability.
  • Evaluation of human interfaces.
  • Evaluation of software interfaces.
  • Identification of potential design defects and safety hazards.
  • Evaluation of potential corrective actions.
  • Simplifying maintenance and troubleshooting.
  • Logical elimination of causes for an observed failure.