What Is Fault Tree Analysis?

Fault tree analysis (FTA) is a deductive failure analysis method from top to bottom. It uses Bollinger logic to combine low-order events to analyze undesired states in the system. Fault tree analysis is mainly used in the fields of safety engineering and reliability engineering to understand the cause of system failure and find the best way to reduce the risk, or to confirm the occurrence of a safety accident or a specific system failure. Fault tree analysis is also used in aerospace, nuclear power, chemical process, pharmaceutical, petrochemical and other high-risk industries. It is also used in other areas for risk identification, such as the failure of social service systems. Fault tree analysis is also used in software engineering. It is used in debugging and has a close relationship with the technique of eliminating the cause of the error.

Fault tree analysis

Fault tree analysis (FTA) is from top to bottom

Fault tree analysis (FTA) is from top to bottom

Fault tree analysis can be used:

• Understand the relationship between the topmost event and the unwanted state below.

• Shows system compliance with system safety / reliability specifications.

• Prioritize the causes of the top event: Create a list of key equipment / parts / events for different importance measurements.

• Monitor and control the safety performance of complex systems (eg: Can a certain aircraft fly safely when the fuel valve x behaves abnormally? How long can the aircraft fly in this situation?)

• Minimize and optimize resource requirements.

• Assist in designing the system. Fault tree analysis can be used as a design tool to create outputs or requirements for lower-level modules.

• Diagnostic tools can be used to identify and correct the cause of the top event, and help to create a diagnostic manual or diagnostic program. ^[1]

  Many industrial and government technical standards mention the methodology of fault tree analysis, including NRC NUREG0492 for the nuclear energy industry,

  There are many different ways to perform fault tree analysis, but the most common and most used method can be organized into several steps. A fault tree can analyze an unwanted event (or the topmost event), or only one. The result can be connected to other fault trees and become a basic event. Although the nature of the unwanted events may be very different, the events may be 0.25ms later than the power generation system, an undetected cargo compartment fire, or random accidental launches of ICBMs, but the fault tree analysis procedures are the same . Because of labor cost considerations, fault tree analysis is typically performed only for the most serious of unwanted events.

  Fault tree analysis can be divided into five steps:

  1. Define unwanted events to explore

  • The definition of unwanted events can be very difficult, but some events are easy to analyze and observe. An engineer who fully understands system design or a system analyst with an engineering background is best suited to define and enumerate unwanted events. Unwanted events can be used for fault tree analysis. A fault tree analysis can only correspond to one unwanted event.

  2. Get information about the system

  • If an undesired event is selected, all the causes that affect the undesired event and their probability of occurrence must be studied and analyzed. To know the exact probability requires a high cost and time, which is mostly impossible. Computer software can be used to study related probabilities and allow for lower cost system analysis. The system analyst can understand the entire system. The system designer knows everything about the system. This knowledge is important to avoid missing any cause of unwanted events. Finally, list all events and probabilities in order to draw the fault tree.

  3. Plotting the fault tree

  • After selecting an unwanted event, and analyzing the system, knowing all the reasons that may cause this event (may also include the probability of occurrence), you can draw a fault tree. The fault tree is composed of OR gate and AND gate, which defines the main characteristics of the fault tree.

  4. Evaluate the fault tree

  • After drawing a fault tree for unwanted events, it is necessary to evaluate and analyze all possible improvement methods. In other words, it is to manage risk and try to improve the system. This step leads to the next step, which is to control the identified risks. In simple terms, this step will find ways to reduce the chance of unwanted events.

  5. Controlling identified risks

  • This step will vary from system to system, but the main focus is to confirm that all feasible methods are used to reduce the incidence of events after all risks have been identified. ^[1]

    Fault tree analysis is

    Fault Tree Analysis (FTA) was originally developed by Bell Watson's HA Watson. It was initially commissioned by the US Air Force's 526th ICBM System Group to evaluate the launch control system of the Volunteer Type 1 Intercontinental Ballistic Missile (ICBM) . Since then, fault tree analysis has become a tool for failure analysis by reliability analysts. In 1962, the research on the launch control safety of the Volunteer Type I Intercontinental Ballistic Missile was announced for the first time using fault tree analysis technology. After that, Boeing and Avco began using the fault tree analysis on the Volunteer Type II complete system from 1963 to 1964. . In 1965, sponsored by Boeing and the University of Washington, in a system security seminar in Seattle, extensively reported the related technology of fault tree analysis. Boeing began using fault tree analysis in the design of civil aircraft in 1966.

    Later, the US military's Picatinny Arsenal began using fault tree analysis for lead applications in the 1960s and 1970s. US Army Equipment Command began integrating fault tree analysis into the Engineering Design Handbook on Design for Reliability in the 1976s. The Rome Laboratory's Reliability Analysis Center and subsequent organizations under the United States Defense Technology Information Center have published documents for fault tree analysis and reliability block diagrams since the 1960s. More recent references are available in MIL-HDBK-338B.

    The United States Federal Aviation Administration (FAA) issued a revision of 14CFR25.1309 in Federal Register 35 FR 5665 (1970-04-08) in 1970, which is a regulation for airworthiness of transport aircraft. This amendment adopted the failure probability criterion of aircraft systems and equipment, so civil aviation operators began to use fault tree analysis widely. The FAA issued Order 8040.4 in 1998 and established a risk management policy that includes hazard analysis, including many key activities after aircraft certification, including air traffic control and modernization of the US National Airspace System, and later Federal Aviation Management The agency also published the FAA System Safety Handbook, which describes many formal hazard analysis methods, including the use of FTA.

    In the early days of the Apollo program in the United States, the possibility of sending astronauts to the moon and returning to Earth safely has been analyzed. Based on the results of some risk (or reliability) calculations, the probability of the task being successful is unacceptably low. Therefore, NASA does not conduct subsequent quantitative analysis or reliability analysis, and only relies on failure mode and impact analysis and other qualitative system safety assessment tools until the Challenger event occurs. After that, NASA experienced the importance of fault tree analysis and probabilistic risk assessment (PRA) in system safety and reliability analysis, and began to use it widely. Later, fault tree analysis became one of the most important system reliability and safety analysis technologies.

    In the nuclear energy industry, the US Nuclear Energy Regulatory Commission began to use probabilistic risk assessment (PRA), including fault tree analysis, in 1975. After the Three Mile Island nuclear accident in 1979, it significantly expanded its research on probabilistic risk assessment. Finally, the U.S. Nuclear Energy Regulatory Commission published the NRC Fault Tree Handbook NUREG0492 in 1981, which also mandated the use of probabilistic risk assessment techniques within the jurisdiction of the Nuclear Energy Regulatory Commission.

    After the 1984 Bhopal incident and the 1988 Alpha rig explosion, the US Department of Labor Occupational Safety and Health Administration (OSHA) issued Federal Register 57 FR 6356 (1992-02-24) in 1992. , Which mentions the process safety management (PSM) standard in 19 CFR 1910.119. The Occupational Safety and Health Administration's program safety management system treats fault tree analysis as a viable approach to process hazard analysis (PHA).

    At present, fault tree analysis is widely used in system safety and reliability analysis. Fault tree analysis is also used in all major engineering fields. ^[1]