How to Use Big Data to Gain Oil Industry Insight

Lonnie Ayers

Have you ever ridden in a Tesla?  The all electric car?  I recently had the opportunity to test drive one and have to say, it was pretty impressive.  Especially since it gets infinite gas mileage! As a mechanic, I especially appreciated the reduced maintenance requirements, for example, if you're not burning gas, you don't need an exhaust system.   Then I read an article that said it may not be so green after all as the electrical generation plants that make the electricity that you use to recharge it are not all that clean.

As someone who has drilled a lot of wells in the past, I frequently refer to a chart produced by the US Government’s Energy Information Agency that shows retail gasoline deliveries month-by-month.


Not unexpectedly, gasoline consumption seems to be trending ever lower. But like any good analyst, I like to see the underlying numbers to see if there is any insight that is hidden by the chart.


Lo and behold, even in a year when the economy appears to be improving, if you look at any month in 2014 compared to any month in 2013, you see that consumption is down by about a 1/3rd YoY. Even more astonishing is that if you compare the 2014 numbers to the 1983 numbers, you see that monthly gasoline consumption is less than 50% of what it was back then.

What could account for that?  Who would want to know?  After all, gasoline prices have remained relatively high until very recently. 

If I were a business planner in an oil major, oil field services company or anyone involved in the upstream or downstream oil industry, I might want to know what is causing gasoline consumption to decline so dramatically, so I could make exploration, production and refinery capacity decisions.

If you thought this drop in consumption was a reflection of a drop in miles driven, you would be wrong. 


Data from St. Louis Federal Reserve.

What is clear is that the total number of miles driven, even though relatively flat since the onset of the 2008 Great Recession, are well above the 1983 levels, when gasoline consumption was almost triple what it was in 2014.

So how can we be driving more than twice as many miles, yet burning less than half as much gas?

Was there an event that would seem to have accounted for this. Yes, there was, a government program called  “Car Allowance Rebate System” or ‘Cash-for-Clunkers’ .

It turns out, that the average vehicle mileage of the vehicles bought under the program had:

An Average Fuel Economy of

  • New vehicles Mileage: 24.9 MPG
  • Trade-in Mileage: 15.8 MPG
  • Overall increase: 9.2 MPG, or a 58% improvement

Cars purchased under the program are, on average, 19% above the average fuel economy of all new cars currently available, and 59% above the average fuel economy of cars that were traded in. This means the program raised the average fuel economy of the fleet, while getting the dirtiest and most polluting vehicles off the road., Summary Statistics, 2009

But wait you say, this is just a small subset of the cars on the road. What about the rest of them? Turns out, that according to Automotive News,, the average age of the U.S. light vehicle fleet is at record highs, meaning that people have been keeping them on the road as long as possible, but that they are being forced to buy new, more efficient vehicles at an increasing rate.

What sort of mileage might those new vehicles get, even if they are not totally electric, like the Tesla I rode in?  The government mandated corporate average fuel economy ratings, or CAFE, are slated to rise from 37 MPG in 2013 to about 60 MPG by 2025.

What Conclusions Can You Draw From All This?

Roughly speaking, oil consumption is going to go down, way down, and as American oil production rises, we may indeed need less Saudi oil.

Most importantly, if I work in the oil industry, is there a system(s) that I could use to perform this type of analysis across these same web based information sources? If so, which one would be easiest to implement while offering the functionality I need, even when I don’t always know what questions I am going to try to answer?

To help answer that question, we recently did an evaluation of all the leading big data solutions and produced our own evaluation matrix.


We evaluated 21 solution providers and arrayed them across multiple dimensions.  There are solutions for virtually every environment, requirement and budget, and we believe, as a tool agnostic Business Intelligence company, that this tool can help you make the right decision when it comes to selecting and implementing a ‘Big Data’ solution.

You can easily get this valuable FREE report by clicking the button below.


Lonnie D. Ayers, PMP, SCM
Senior SAP Program Manager