Is the forecasting of spare parts requirements hard, especially intermittent parts?
Is your demand for parts based on random failure?
These issues lead to forecasts that don’t achieve their goal. Having the past you need, when you need them.
In this short video Phillip Slater explains what works – and what doesn’t!
Forecasting Spare Parts Requirements
Video Transcript
Hi, name is Phillip Slater and for the next couple of minutes I’d just like to give you something to think about.
All spare parts inventory management stocking decisions can ultimately be resolved as a forecasting problem.
Because if we think about why we’re holding that stock, we are holding that stock in order to attend to future needs, so, therefore we need to have an idea of what those future needs might be.
And the only way we can try to estimate or approximate those needs is to provide some kind of forecast.
So, the way that you forecast your future needs is actually really important.
Not only that, when you think about spare parts that are used for maintenance and operations support, a lot of the demand for those parts is based on random failure.
Of course, randomness also implies a degree of intermittence, that is, the failures will be intermittent, almost by definition. So there will be a variation in terms of frequency of demand and quantity of demand.
Now this means that some standardized processes such as the normal bell shaped curve are most likely inappropriate for these types of parts. That is, random and intermittent failures.
And you’ll need to look at some other types of forecasting techniques and one that I’ve seen that works particularly well is known as bootstrapping.
Bootstrapping takes a series of your past data and runs thousands of iterations in order to be able to come up with a reliable forecast within certain constraints of risk.
It’s actually one of the best approaches I’ve seen for managing or forecasting intermittent spare parts.
So, when it comes to managing your spare parts inventory, think carefully about the type of forecasting technique that you are using or considering using.
That’s something the think about.
My name is Phillip Slater.
Posted by: Phillip Slater