A distributed recommendation system to generate insights and optimize oil production

Abhay Dutt Paroha is an experienced Software Engineer and Technology Leader at SLB. Abhay leads a team of engineers to develop software applications to optimize oil production.

Oil and gas companies are utilizing digital technologies to automate workflows and increase production in real time. However, these applications often require on-premises systems, which can be challenging due to scalability, accessibility, and maintenance. Data collection in production operations can become complicated due to disparate sources, data frequencies, software conventions, and maintenance. These challenges can lead to non-productive time, data quality issues, inconsistencies, and missing or incomplete data.

I invented 3 novel patented solutions (as first author, second author, and individual author) and built a distributed recommendation system to provide insights and help production engineers diagnose problems and confidently make informed decisions by quickly seeing and understanding the reasons for oil well underperformance. Here is a summary of inventions:-

• A cloud-hosted autonomous software agent system was developed to efficiently ingest oil well production operation data from various sources (historians, edge devices, sensors, mobile apps, and simulation models), reducing latency and enhancing data flow efficiency, enabling applications or workflows to generate results on-the-fly.
• Created a bi-temporal structure and time-series cloud storage for consistent, reproducible long-running calculations, enabling the unification of data from various sources and representing both raw and calculated data.
• Developed multiple source data change journal system to receive data from a source; detecting a change in the data; generating an aggregate change journal based on the change; and providing access to information in the aggregate change journal by a computational framework that consumes the data in a time-dependent manner.

The recommendation system, used by five major oil companies in South America, South Asia, and Australia, managed 14,450 oil wells, generated $XX million in revenues, and resulted in 98% time savings to execute automated production forecasting, 88% cost savings for the execution of oil well review programs, and increased well uptime by 20%.
I received a patent grant award, two patent filing awards, and two organization’s global recognition awards for these significant contributions.