Determine knowledge model and root cause of 12 reported incidents of poor hole cleaning
MetadataShow full item record
Case based reasoning can be applied to drilling problems which are too complex to be solved by conventional mathematical or statistical models, due to the complexity and uncertainties involved in drilling a well. People use the principles of case based reasoning in many aspects of daily life, in order to avoid mistakes done in the past. For an oil company it is extremely useful to systematically store the knowledge gained from their engineers so that when they eventually retire the knowledge they posses is not gone to waste. The basic goal of this study was to use the principles of case based reasoning, and through observations made from the events find the root cause, and thus the solution. Each symptom or observation from a case was directly linked to a root cause, and assigned a relation strength, which in the end was used to find the most probable failure type. The magnitude of the relations ranged 0 to 1. For the model to work properly it was important to obtain a large number of symptoms in order to efficiently separate the different root causes from each other. However, this proved to be a difficult challenge since most observations were based on real time data. The model used in this study was developed in Excel, and was used to test the most probable root cause of 10 previously solved and 2 unsolved cases of poor hole cleaning on a large oil and gas field in the Russian sector of the Barents Sea. The model returned the same root cause as the previously stated cause, in 8 of the 10 cases.