Failure Modes Available for Function and Action Elements

The failure modes types available for both the Function and Action elements are described below.

Note that most failure modes are defined relative to a failure mode control variable. This is the variable that is referenced by the failure mode to determine when failure occurs. (For those failure modes that are defined as distributions, the control variable represents the x-axis of a failure distribution plot.) By default, the failure mode control variable A variable that is referenced by each failure mode to determine when failure occurs. For those failure modes that are defined as distributions, the failure mode control variable represents the x-axis of a failure distribution plot. Each variable is defined with respect to a base variable (e.g., total time, operating time, mileage). is the operating time since the simulation began (or the last time the component was replaced). However, for any given failure mode, it can be set to the total time since the simulation began, the number of actions completed (available only for Action elements), or a user-defined variable (e.g., mileage). You can also accelerate or decelerate failure with respect to the control variable.

Mathematical details of the failure modes are provided in Appendix A of the Reliability Module User’s Guide.

• Cumulative. For this failure mode, you define a custom failure distribution by specifying a table of the (cumulative) failure mode control variable value and the probability of surviving. The dialog for defining this table is accessed via an Edit... button, and looks like this:
  
  
  The table can only be defined using numbers (it does not accept expressions or links to other elements). You add and delete rows using the Add Row(s) and Remove Row(s) buttons.

Note: Probabilities must decrease monotonically as time increases.

• Defective Component. For this failure mode, you specify the probability that the component will be susceptible to the mode, and the (Poisson) failure rate if it is susceptible. As its name implies, it is used to simulate (typically rapid) failure due to a small fraction of defective components.
  
  
• Erlang multi-failure. This failure mode is used to represent the failure of a number of identical sub-components, each of which fails according to the same Poisson failure rate, and is used to simulate a system that has N-1 spare parts, that are replaced immediately.
  
  It is assumed that the first sub-component operates until it fails, at which time it is replaced by the second identical second sub-component, and so on until they have all failed. When all sub-components have failed, the component is assumed to have failed.
  
  
• Event-triggered failure. For this failure mode, you specify a trigger (via a triggering dialog) and the probability of failing whenever the triggering event occurs:
  
  
• Exponential/Poisson. This is identical to the default failure mode if the Failure Mode tab is not used. You define an exponential failure distribution by specifying the rate of failure (also referred to as the hazard rate) with respect to the failure mode control variable:
  
  
• LogNormal. For this failure mode, you define a log-normal failure distribution with respect to the failure mode control variable. Two LogNormal failure mode types are provided. This allows you to define the distribution in terms of either a Geometric Mean and Geometric Standard Deviation The square root of the variance of a distribution. The variance is the second moment of the distribution and reflects the amount of spread or dispersion in the distribution.:
  
  
  or an arithmetic Mean and Standard Deviation:
  
  
• Normal. For this failure mode, you define a normal failure distribution with respect to the failure mode control variable. To do so, you specify a Mean and Standard Deviation:
  
  
• Specified Value Exceeded. For this failure mode, you specify the value for the failure mode control variable which results in failure if it is exceeded:
  
  
• Uniform. For this failure mode, you define a uniform failure distribution with respect to the failure mode control variable. To do so, you specify a Minimum value (lower bound) and a Maximum value (upper bound):
  
  
• Weibull. For this failure mode, you define a Weibull failure distribution with respect to the failure mode control variable. Two Weibull failure mode types are provided. This allows you to define the distribution in terms of either a Mean life and Slope factor:
  
  
  or a Characteristic life and Slope factor:
  
  

Note: The input fields for Failure Mode Parameters can accept numbers, expressions and links from other GoldSim elements. They can also be specified as functions of time. Depending on the failure mode type, however, parameters that are defined as a function of time may not change instantaneously (i.e., they may only change when the mode is repaired or the component replaced).

There are two additional failure mode types that are used to model preventive maintenance.