What is Statistical Process Control?

Hi All!

Thank you all for the feedback from our last post on Digital Oilfield! Our goal is to continue to provide useful content from our own experiences from the energy industry.

This week’s post was supposed to be about Open Oilfield, but it’s actually our technology team’s turn to pick a topic. We realized it was important to talk about an area of engineering that could provide a tremendous “array” of solutions on how to operate at 50 USD/barrel oil: statistical process control.

Without even trying to attempt how to explain statistical process control, we’ll be millennials and resort to Wikipedia to define it (sorry academic mentors, we know better not to): 

“Statistical process control (SPC) is a method of quality control which uses statistical methods. SPC is applied in order to monitor and control a process.”

But what does that really mean and how does that apply to you and I?

In the simplest of terms, Statistical Process Control  is a way to optimize processes using statistical data. In other words, use math to influence the world.

By measuring the output, it uses what are called special and common cause variations to detect whether a process is in control or not. Let’s put that in normal words.

While the very thought of it causes us discomfort, let’s talk about driving here in Houston.

If you live in a city like Houston and drive to work everyday, you probably take a certain amount of time to drive from your home to work, assuming you follow a discipline schedule. For example, it takes me around 20 minutes to drive each day at 7 am.

However, it’s not exactly 20 minutes now is it?

Some days, I might take 23 minutes and on other days maybe 18 minutes. This is what we call common cause variation. Maybe some days traffic is a bit heavy, on other days I might have bad luck with some of the traffic lights. On other days, I just might be lucky with no traffic and all traffic lights going in my favor.

Now say one day unfortunately there is a traffic accident and I take an hour to get to work.

This is “out of the blue” and cannot be explained by any unknown factors such as traffic light timing.  This scenario is defined as special cause variation.

Assuming driving to work follows a pretty consistent schedule or process we can now determine external factors and what we can improve on, or see what we can control.

This is how we see the world at RigBasket.

Seeing the world with this perspective is key towards improving the world around us. We just happen to have a strange passion for fixing inventory. So much so that we have spent many weekend early mornings over the years at warehouses, bulk plants, and parts trailers tallying with paper clipboards.

We believe that inventory control is just another type of process control. Like most processes, inventory should follow a pattern around known numbers. Not too bold a claim now is it?

What if demand is changing all the time?

What if suppliers produce unreliable parts?

What if suppliers have a highly variable lead time?


Say I have a part with a certain part number and the inventory numbers for that part are moving all over the place every month. One month I have a shortage and then three months later I have an over-supply. Using statistics, we can figure out what exactly is causing your inventory to go “nuts” and reduce the volatility.

How does this all apply to the energy industry?

Say you’re in the energy industry and your demand fluctuates with oil prices causing you to be over or under-stocked all the time. We claim that through statistics, you could figure out how much inventory you will need based on the oil price trend.

In fact, RigBasket can do that now. 

Through machine learning and predictive analytics, we can understand how demand is going to change through statistics.

So without getting into the real technicalities of control theory by deleting half of the remaining original article, we want to conclude with an emphasis on more statistical methods to the madness that is energy operations to make both our academic & professional mentors proud after bringing shame by citing Wikipedia.

“Math happens” in the digital oilfield people, and it’s about to matter even more.

Until next time, feel free to drop us a line at www.rigbasket.com or tweet to us @rigbasket. We are open to suggestions about what topic to cover next.

Leave a Reply

Your email address will not be published. Required fields are marked *