Cost Effective Approach of Big Data Analytics in Oil and Gas Industry: Case Study for a Giant Offshore Field in Middle East

The subject offshore field is undersaturated and spans approximately 35 km from east to west and 20 km from north to south. Field development commenced with seawater injection since the beginning. The initial development phase included drilling a series of peripheral water injector wells to maintain pressure and improve oil recovery. A five-spot pattern based on wellhead towers with a spacing of 2.8 km was employed later for the development. This spacing was subsequently reduced to 2 km and then to 1.4 km through infill drilling. Recently, artificial islands have been constructed to facilitate the re-development of the field, shifting from a wellhead-tower-based five-spot pattern to an islands-based line-drive pattern. Currently the field is producing through thousands of well strings. As can be seen from Fig. 1, these strings extract hydrocarbon from different reservoirs, then gather on dozens of well head towers and multiple artificial islands at the surface for further processing before export. The subject field contains multiple reservoirs, each with its standalone subsurface simulation model. One of the key tasks the team works on is the field development business plan. The main objectives of this task include: 1) obtaining the predicted short-term and long-term fluids profiles (i.e. oil, gas, produced water, injected water) for each reservoir and the entire field, 2) obtaining the predicted short-term and long-term fluids profiles (i.e. oil, gas, produced water, injected water) for each surface location to check surface facility adequacy and for planning purposes, 3) designing the oil/water/gas treatment facilities based on the fluids behavior at each surface location, 4) given the limited rig resources, obtaining and optimizing the well drilling schedule for the entire field based on the drilling sequence priority of each reservoir, 5) designing the artificial lift strategy based on the wells' bottom-hole pressure performance and field operation strategy, and 6) obtaining profiles for different wedges which represent different activities for economic evaluation. The subsurface simulation model is run separately for each reservoir in this field, and it is a challenging process to extract each well’s data from each reservoir and then couple and analyze the data manually using the conventional tool: e.g. combining a large volume of related data in Excel using multiple functions, which resulted in tens of thousands of rows of formulas. The spreadsheets are resource-intensive and the calculations need to be performed every time for the entire field even with minor updates. Additionally, the volume of data that needs to be processed is substantial: there are several million rows of data generated from the simulation for wells from all reservoirs, which exceeds the limit of conventional tool’s capability.