What is the probability of throwing 6 with a single die?
Given some simple assumptions, we could all agree that the answer to this question should be 1/6 or about 16.7%. But how do know? How can we be sure? In short, where does the answer 16.7% originate from?
We could perform an 'experiment' to determine this. We could e.g. throw the die many times over and then count the number of times we end up throwing 6. Over the long run this should amount to about 16.7% of the time.
In reliability engineering, we rarely have the luxury of performing experiments to obtain insight into the number of equipment failures in a certain time interval.
This is especially true for critical, capital equipment for which we need to determine a maintenance strategy. We need to analyse trends and optimise, but our maintenance regimes in general do a good job in preventing us from collecting data. The results is that the data we do obtain is probably related to equipment we don't care about very much.
This phenomenon is called Resnikoff's conondrum and has been recognized since the late 1970's. In a future post, I will dive deeper into this phenomenon to further explore its impacts.
Search by tags
Let's talk about the needs of your organization.
.png){kind=small}
