In most products, a few components are responsible for most of the product failures, while the remaining components tend to function perfectly throughout the product’s life. This behavior is similar to the Pareto principle discussed in Reliability Analysis chapter.

As a product undergoes the diagnostic phase of repair, the technician will first examine the problematic components in an attempt to locate the fault as quickly as possible. This effort can be modeled using fault isolation.

To define the fault isolation scenario, you first define up to five groups of parts N1, N2N5. Each group is a subset of the next group that the technician examines. N1 includes the set of parts examined first, N2 includes the parts examined in N1 plus the next set of parts, and so on. If all sets of parts N1N5 are examined, then in effect the technician has examined the entire product tree.

Associated with each part grouping is a probability of fault isolation. This is the probability that the faulty component is a member of any one of the examined sets. Suppose you group all the problematic components for the first set N1. Then the probability (X1) that the fault will be detected while examining the set of parts in group N1, will be high. If the fault is not detected while examining N1, then the probability (X2) of detecting the fault while examining the set N2 will be higher. This continues until the technician examines the last set of parts - N5, whereupon the probability of isolating the fault after examining all components is X5 = 100%.

For example, if a failure is always isolated to a single part in group N1, then

X1  = = X5 = 100%

and N1 = 1.

If in 80% of the cases a failure is isolated to a single part, and in the remaining 20% to two parts, then X1 = 80%, X2 = 100%, N1 = 1 and N2 = 2.

In the Fault Isolation group box, enter your values for N1N5 and X1X5.

The value of the ambiguity factor S is calculated based on fault isolation and is used to calculate the MTTR for an assembly (see Ambiguity Factor paragraph).

RAM Commander is able to take the ambiguity groups data from FMECA module.