Failure (engineering)

From Citizendium
Jump to navigation Jump to search
This article is developing and not approved.
Main Article
Discussion
Related Articles  [?]
Bibliography  [?]
External Links  [?]
Citable Version  [?]
 
This editable Main Article is under development and subject to a disclaimer.

A failure is any occurrence of a designed system not performing as expected or desired. Failures vary in severity (sometimes called criticality) between minor inconveniences (low tire pressure) to catastrophic disasters (passenger aircraft crash). System designers seek to minimize the potential causes of failure, the severity of failure, and the likelihood of failure--also known as maximizing reliability.

Overview

As engineered systems have grown more complex and more powerful with greater numbers of individuals participating in their design, anticipating the sources and consequences of failure has become both more difficult and more important. Failures can result in deaths, injuries, environmental damage, lost production, wasted resources, and lost time. In response to these risks formalized methods are regularly employed to minimize failures on new designs and to prevent the recurrence of failures on existing systems. The specialty interdisciplinary field of Reliability engineering is concerned with both minimizing the likelihood of failure and minimizing the potential severity of failures when they occur, but generally all engineers involved in system design employ methods specifically to mitigate failures.

Failure Mode

The failure mode is the manner in which the system does not perform as expected. Possible failure modes are based on the specific functions performed by the system in question, but generally can be grouped into one of these broad categories[1]:

  • Failure to function is the most obvious of failure modes and would includes failures such as automobile engine not starting or light bulb burnout.
  • Failure to function at the proper time is a failure mode of automated functions within a system that perform either too early or too late to permit the system to function properly. Examples include delay in real time indicators or timing discrepancies on automated assembly line operations.
  • Failure to cease function is similar to failure to function but with systems that should normally be quiescent. Examples include pressure release valves not closing once pressure returns to normal, or brakes not releasing when commanded.
  • Failure to function within specification includes devices that only partially perform their functions but not satisfactorily such as ball bearings with high friction or air filters that do not block the intended particle size.

Failure Analysis

Engineering analysis of failures includes both deductive and predictive methods with the goal being to produce systems that are more reliable, safe, and less likely fail

  • Failure Modes and Effects Analysis (FMEA) or sometimes referred to as Failure Mode Effects and Criticality Analysis (FMECA) describes each of the possible failure modes at a component or sub-system level tracing their probable effects up to the overall system performance[2].
For more information, see: Failure modes and effects analysis.
  • Reliability Analysis is a probabilistic prediction of the failure rate of a system based on statistical analysis of previous failures on that or similar systems, test data, or engineering design calculations as compared to predicted conditions to be experienced by the component [3].
For more information, see: Reliability Analysis.
  • Fault Tree Analysis begins with a given outcome of interest (e.g. fire occurs) and traces back proximate causes of each event in a tree structure until all of the possible root causes are identified in the leaf nodes [4].
For more information, see: Fault Tree Analysis.
  • Forensic Analysis is a physical analysis of the actual failed components or other debris to determine the physical reasons for the initiation of the failure. These can include findings of whether electrical components were exposed to excessive current or whether a structural member failure due to fatigue or overloading.
For more information, see: Forensic Analysis (engineering).
  • Root Cause Analysis examines all of the events and occurrences that led to the actual failure being investigated with the aim to produce recommendations to prevent recurrence of that failure [5].
For more information, see: Root Cause Analysis.


References

  1. National Aeronautics and Space Administration, International Space Station Program. (1999) Failure Modes and Effects Analysis and Critical Item, Requirements for Space Station. SSP 30234, Revision E
  2. National Aeronautics and Space Administration, International Space Station Program (1999). Failure Modes and Effects Analysis and Critical Item, Requirements for Space Station. SSP 30234, Revision E
  3. Ebeling, Charles (2009). An Introduction to Reliability and Maintainability Engineering. Waveland Pr Inc. ISBN 978-1577666257.
  4. Ericson, Clifton (2005). Hazard Analysis Techniques for System Safety. Wiley-Interscience. ISBN 978-0471720195
  5. Robitaille, Denise (2004). "Root Cause Analysis: Basic Tools and Techniques. Paton Press. ISBN 978-1932828023