The pressure to cut costs and optimize resources amid the oil price slump has made exploration and production companies analyze more carefully the potential of the shale assets they consider buying. The oil and gas industry has now started to use data analytics – comparing data from multiple sources – an approach that other sectors such as finance, for example, have been using for years.
Unconventional fields, especially those in West Texas—home to the Permian basin—have too many wells on a limited acreage, each with its own specific production-type curves, cost of drilling, geological formations, leasing costs, and completions optimizations. And each of those factors is changing from well to well, location to location, or from one period of time to another, because unlike conventional fields, unconventional resources are much more diverse in their formations. These variances make it difficult to compare assets.
Abhishek Gaurav, completions engineer at Texas Standard Oil, and his colleagues Edward J. Gibbon and Timothy M. Roberson, have created a methodology to aid in this comparison. The methodology would evaluate the economics and potential of shale assets by integrating data about the geology, completions, production, and leasing of the formations from public sources.
The idea behind this methodology is that the completions economics and optimization is a variable, rather than a constant. Therefore, an asset that previously looked unattractive may not always be ‘unprofitable’ given the recent advances in technology and efficiency. Gaurav and his colleagues believe that the completions parameters are key to helping a company identify the economic and production potential of a shale asset. The engineers combine all kinds of data into one huge data set in order to identify patterns with the best potentially performing wells.
The Texas Standard Oil team – who presented their data analytics paper at the SPE Unconventional Resources Conference in Calgary earlier this month – argue that big data analytics that considers all variables can help E&P companies identify the economics of an asset on their own instead of just “copying and pasting” the best practices of another operator in the same area. That’s because no well is exactly similar to another, and performance indicators vary depending on a number of factors, including geological formation and the way a well is drilled and/or positioned.
“But we also see that what really differentiates performance is the orientation—where you land the well,” Gaurav tells Journal of Petroleum Technology.
In addition, wells oriented in a particular way can deliver better performance than other wells in the same area.
For example, Gaurav has noticed that a change in the direction of the wellbores in Reeves Country, Texas, has made wells produce more than other wells in the same area.
Most of the data the engineer has been crunching is available in the public domain: in corporate press releases or in regulators and watchdogs’ data sets. The data not available in those sources is often found in the companies’ conference calls with analysts when E&P firms share the economics and details of their latest completion optimization advances to boast progress in efficiency. In addition, oilfield consulting reports are another source of data for Gaurav’s analytics methodology.
The engineer is also using artificial intelligence and machine learning to analyze shale assets, and aims to show how they can enrich his data analytics approach.
The oil and gas industry – once a rather conservative sector – has started to embrace the digital age with cloud computing and supercomputing, and big data analytics is another step in the direction that has been changing many other industries such as banking, retail or healthcare.