From a Simple Electric Motor

When a manufacturer of electric motors noticed several shortcomings in their traditional reliability tools that prevented them from representing their product’s reliability, they came to us. The first problem was that all of the motor’s components had wear-out failure modes best represented by the Weibull distribution, and had lognormal repair distributions, but the company’s traditional tool could only model the exponential distribution. Raptor solved their problem by offering 18 different statistical distributions, plus the ability to enter raw data to build custom distributions.

The second problem was that the motor’s parts had intricate dependencies that could not be modeled with any other tool. If the power supply component failed, the brushes and armature components stopped functioning and, most importantly, stopped accumulating life. However, the failure of the power supply had no effect on the failure modes of the magnets or casing components since they failed independently of the motor's rotational effects.

Most reliability analysis tools overlook these important issues and generally solve an extremely small portion of real-life problems. Because those tools are equation based, they must assume exponential failures and independent components to ease the mathematical derivations. Raptor uses a powerful Monte Carlo discrete-event simulation engine, not closed-form mathematical models. Because of this, Raptor avoids the time-consuming development of path-based equations and can model configurations where closed-form solutions do not exist, such as complex systems that cannot be reduced into series or parallel subsystems.

Related to the issue of dependency are the concepts of adjacency and cascading failures. Adjacency is a design concept that allows components to use "nearby" alternates when its own support components have failed. Cascading is an extension of the adjacency concept in which the acquisition of nearby equipment causes the next component to acquire, in turn, its next nearest equipment, and so on, causing a chain reaction.

For example, component A’s support equipment fails and so it acquires component B's support equipment. Component B then acquires C’s and thus cascades the failure effect. Modeling such concepts can be extremely difficult in most reliability block diagram formats. Unfortunately for many reliability simulators other than Raptor, modern electronics commonly use those forms of advanced redundancy to improve their overall availability, safety, and performance.