Interpreting formulae and not just generating numbers …

Abstract

Chris and Fred discuss how formulae (or formulas!) can be very important … but it is way more important to understand what formulae represent, and how they work. Listen to this podcast if you want to learn more!

Key Points

Join Chris and Fred as they discuss how important it is to understand what formulae represent, including all those pesky parameters that are represented with Greek letters. Why is that?

Topics include:

What is the ‘mean’ or ‘average’ of something? Some of us might be able to quote a formula for the mean. But do you actually know what it is? Put simply … the mean is the ‘balance’ point of your data points if they sit on a pivoting plank. Think of the ‘balance point’ of the histogram of your data points. That’s it. What about the standard deviation? Well … once you get that histogram, find its moment of inertia (i.e. how hard it is to spin around) and then take its square root. Again … that’s it. But did you know that?
OK … but why do I need to know how to interpret formulae? Take (for example) Weibull plotting. For those who don’t know what it is, it is just a special way of visualizing data that allows us to see failure characteristics if we know what to look for. If you know what the shape of these data points means, you can see where you need to service (i.e. when things start to wear out), or how many things are suffering from infant mortality. If you don’t know what equations mean, then all you can calculate are parameters, means, and other numbers that you won’t know how to help you make better decisions.
Then there is error checking. If you know what your formulae and equations represent, then you can have a good guess at what the answer should be before you get the number. And this can help you make sure you didn’t make a simple error. If you are modeling the time to failure of something, and a computer tells you that the parameter ‘μ‘ of the normal distribution you use to model time to failure is negative, then you should immediately know you have a problem. Why? Because that parameter is the ‘mean’ or the balance point of data points that must be greater than zero. This approach has saved so many reliability engineers over the journey!

Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches.