Oil and gas industry generates large datasets that often remain unused by operators and service companies. Key to improving data usage is converting this data into information to be able to easily answer pivotal questions on well performance enhancement and other improvement opportunities. This is quite difficult today, as raw data is typically not converted into information when it is stored. In this paper, we introduce the concept of spider bots, and demonstrate their effectiveness in converting this raw data into knowledge. Three classes of spider bots were identified and developed. First, scripts were created to automatically curate and clean the data, a vital step before the data can be further analyzed. Second, scripts were written to convert the cleansed data into information. Such scripts calculate metrics such as wellbore quality (tortuosity index, dog leg severity, etc.), drilling dysfunction type and severity, invisible lost time, etc. Finally, scripts were developed to create indices based on the information generated by the second set of scripts to rank well performance on the basis of the various KPIs. To test the effectiveness of the spider bots after deployment, we queried the database with relevant questions. One such question was "Which is the best drilled well out of the given wells?" Previously, answering this question effectively would have taken an excessive number of analyst human hours. On our system, this query was answered in a few seconds, chiefly due to the indexing performed by the spider bots. When we added new well data to the original dataset and posed the same question again, the answers changed to reflect the new data. A key advantage of our approach was the ability to easily update and improve the spider bots by modifying the existing scripts or by creating entirely new ones. These bots were primarily applied on drilling data, but can be applied universally on all kinds of oil and gas data, including for e.g. completions and production. These bots significantly facilitate and accelerate the analysis of big data and will become a necessity as the industry moves towards trying to generate more value from big data.
Skip Nav Destination
SPE/AAPG/SEG Unconventional Resources Technology Conference
July 23–25, 2018
Houston, Texas, USA
Spider Bots: Database Enhancing and Indexing Scripts to Efficiently Convert Raw Well Data Into Valuable Knowledge
Hong-Chih Chan;
Hong-Chih Chan
University of Texas at Austin
Search for other works by this author on:
Pradeepkumar Ashok;
Pradeepkumar Ashok
University of Texas at Austin
Search for other works by this author on:
Eric van Oort;
Eric van Oort
University of Texas at Austin
Search for other works by this author on:
Mojtaba Shahri
Mojtaba Shahri
Apache Corporation
Search for other works by this author on:
Paper presented at the SPE/AAPG/SEG Unconventional Resources Technology Conference, Houston, Texas, USA, July 2018.
Paper Number:
URTEC-2902181-MS
Published:
July 23 2018
Citation
Saini, Gurtej, Chan, Hong-Chih, Ashok, Pradeepkumar, van Oort, Eric, Behounek, Michael, Thetford, Taylor, and Mojtaba Shahri. "Spider Bots: Database Enhancing and Indexing Scripts to Efficiently Convert Raw Well Data Into Valuable Knowledge." Paper presented at the SPE/AAPG/SEG Unconventional Resources Technology Conference, Houston, Texas, USA, July 2018. doi: https://doi.org/10.15530/URTEC-2018-2902181
Download citation file:
Sign in
Don't already have an account? Register
Personal Account
You could not be signed in. Please check your username and password and try again.
Could not validate captcha. Please try again.
Pay-Per-View Access
$28.00
Advertisement
39
Views
Advertisement
Advertisement