Water Injection Optimized with Statistical Methods

The complexity of the water injection process is often underestimated and asset managers often focus on controlling the water injection costs rather than maximising the injection water, value creation. This paper discusses a several methods to evaluate the techno-´conomic issues surrounding water injection, ranging from discounted cash flow analysis to portfolio analysis. It is shown how quantitative technical and economic analysis of the available options allows optimum decisions for process management. Advance planning ensures that changes in water injection performance do not surprise managers since they are prepared for such possible events, appreciate the likelihood of their occurrence and are able to quickly identify the optimal response. In many situations, statistical methods are the most appropriate toos to identify these optimal responses. The industry has experience with these methods and the required input data is often available. Monte Carlo methodology is shown to be a powerful alternative to Portfolio Analysis when the input data for the latter is incomplete or there are insufficient alternatives available. Portfolio Analysis emphasises the value of diversity. This can be important as the presumption of the simpler methods that all events are independent is often not true. E.g. the. chance of a success of a stimulation treatment is not neccessarily independent of the chance of success in next well. A decision tree analysis of a water injection case study is used to quantify the "Value of Information" and to show how spending significant resources on extended technical studies can be justified. This extra data might lead to a more "advanced" project design, ensuring that operations staff are better prepared for upsets later in field life. Monte Carlo Analysis shows that, although this "advanced" option is likely to create significantly more value, there is still a small possibility that the "low cost" option is more profitable.