Much of the “unstructured” data collected for Daily Drilling Reports can be made available via Natural Language Processing (NLP) and put to use for analyzing/predicting the causes of Non-Productive Time (NPT). This could mean big savings in the cost of drilling operations!
Non-Productive Time Analysis in Drilling using Natural Language Processing
Dr. Avinash Wesley, Senior Technical Professional-Technologist, Halliburton Landmark
Dr. Geetha Gopakumar, Data Scientist, Halliburton Landmark
Drilling capital expenditure represents a significant portion of any oil and gas project, with drilling investment accounting for ~40% of the well cost. Therefore, monitoring the health of equipment and processes associated with rigs is critical in bringing down both the cost of the well and it’s Non-Productive Time (NPT). NPT accounts for about 10 to 20% of drilling costs for land and offshore rigs, respectively. Currently, there are several Key Performance Indicators (KPIs) to monitor NPT, but they are reactive in nature.
Proactive measures can be taken to avoid NPT and reduce Invisible Lost Time (ILT) by exploiting the power of Big Data. The scope of this work is to present a framework to predict NPT causes from the massive amount of unstructured data collected during drilling from the Daily Drilling Reports (DDRs). Natural Language Processing (NLP) has been used to create value from such unstructured data.
Using this framework, we were able to identify $23 million dollars attributed to NPT from customers’ drilling reports. In the future, the framework will “learn” NPT patterns from the offset wells and suggest preventive actions to avoid NPT during drilling operations.
For the full presentation click here.