Data Mining Applications in the Oil and Gas Industry

Abstract

Data mining and analytics The recent proliferation of resources exploitation, in both traditional and unconventional basins, has led to more upstream oil and gas industry activities in more regions than ever. These newer activities, when added to the already challenging work environments in such frontiers areas as deep water and the Arctic, place tremendous demands on the industry to work more efficiently and avoid risks to people, capital, and the environment. An emphasis on monitoring and assurance of the production operations during exploitation has caused the oil and gas industry to enter the digital age during these last two decades in a grand way and has generated what the information technology (IT) industry calls “big data.” Data acquisition in instrumented wells and monitoring of fields and operations processes are routinely carried out in both real-time and post-mortem modes. Management and use of this big data have become critical for the industry and its stake holders, including regulators and financiers. Integration of data analytics into the practice of petroleum engineers is essential to establishing a vision for the oil and gas industry to move toward data-based decisions in the production and operations arena. The largest amount of computing power in the industry so far has been dedicated to the collection and storage of information (databases and warehousing) as a result of the low cost and portability both of electronic sensors and computer memory. Great emphasis has been put on automated reporting, standardized presentations, and streamlined data transfer to cut costs for data processing and dissemination. Even when no immediate need for the data was present, the low cost of data collection has driven company decision makers to support data storing and sorting for later analysis. Current databases contain transactions that can be stored and retrieved without relying on employees’ memories. Unfortunately, while information banks are full, clients are not communicating efficiently and data sharing sometimes is inhibited by the size of the specialized warehouses. Integration of the data and coincidence of the information seen by all team members across the organization has been lacking. The more serious gap in this digital age has been in the transformation of this data into information and knowledge that could drive management actions or become the basis for corporate-level decisions.