Analytics, to me, has always been about continuous education and giving people the chance to learn new insights. The SAS Day at Houston CityCentre earlier this month is a shining example.
Hosted by the Department of Statistics as Texas A&M University, the event showed how analytics guides decisions across many industries, including oil and gas, energy supply and distribution, government, and sports (yes, even sports, and in this case football). For those who could not attend, most of the SAS Day presentations are now available online.
Below are my top 10 takeaways from the presentations I attended.
#1 - The U.S. and the world will not be running out of gas, or oil for that matter, any time soon. Some of the information regarding the amount of world-wide natural gas is based on a study done back in 1997. For more recent specifics, see Dr. Stephen Holditch's presentation on "Shale Gas Development." (Spoiler alert: North America has around 269 years of technically recoverable resources (TRR) worth of gas while the world has 488 years of TRR gas. And TRR only represents about 25 percent of the total gas available, according to theAugust 2012 data/reports in Holditch's presentation).
#2 - Horizontal drilling has been done since the 1950's so this is not a new or unproven technique.
#3 - Ditto for fracking. Water fracture treatments have been going on for at least 25 years.
#4 - Water table pollution has occurred while drilling water wells, since this is shallow drilling and not as well regulated as drilling for gas, which is deeper and is highly regulated.
#5 - The boom occurring in North Dakota has a high probability of occurring world-wide.
#6 - SAS understands and has the technology to use a combination of multivariant, multidimensional, multivariate, and stochastic (models) to help solve complex subsurface system problems or, in simpler terms, help drill better holes. See Keith Holdaway's "Competing on Analytics: Case Studies." (Note while viewing the presentation: SEMMA is a best practice process for doing data mining which stands for Sample, Explore, Modify, Model, and Assess.)
#7 - Oil companies have used SAS to incrementally increase annual production by 2-3 percent with lower costs, reduce the overall cost of proppant used by 30 percent, automate well surveillance for a 5 percent improvement in forecast accuracy and a 2 percent increase in recovery factor.
#8 – You can take proven analytic techniques from one industry (telcom) and apply it to another industry (football) to help your staff beat the competition. See Joe Indelicato's presentation "Predictive Analysis - Out Thinking the Competition" to see how predictive analytics was applied to help make the right offense and defense based on the players on the field and the play being called by the opposing team.
#9 - One definition for a "data scientist" is a data analyst who lives in California, according to Josh Wills from Cloudera. Take a look at his "Data Science with SAS and Cloudera" presentation.
#10 - Using Hadoop to solve iterative algorithms (advanced analytic algorithms that are interactive) is like cutting pizza with a saw. What is the right tool? Well of course, it is SAS High-Performance Analytics , based on our new SAS LASR Server which is an in-memory analytics engine - NOT an in-memory database.
Ok, so I already knew #10, but it was nice to hear the same advice coming from someone outside of SAS (Josh Wills from Cloudera) and it also would have been weird to provide a top 9 list instead of a top 10.
Dont' forget, you can find all the presentations mentioned here at the SAS Day site. Or, continue your learning at another great event by following the A2013 conference coverage for the next few days.
