Publications

After producing for more than 40 years, many of the large reservoir tanks in the Semberah Field were already depleted and close to ultimate recoverable reserve. In the current situation of high efficiency in the oil and gas industry, the approach used in finding new potential zones for rigless candidates will need to be more robust in order to sustain the production of this mature field. A petrophysical approach using open-hole log data interpretation is one of the most used methods in the oil and gas industry to identify pay zones in a well. Nevertheless this conventional petrophysical interpretation still leaving many potential zones unproduced. Evidence of such bypassed potential zones identification using a more advanced approach of a petrophysic methodology does exist in VICO. Still a great deal of new opportunity of potential zones present in this mature Semberah field and waiting to be discovered. Artificial Neural Network (ANN) is one of the methods in Artificial Intelligence that can be used to generate computational model based on existing data provided. ANN works by mimicking how the human brain works through training of sample data sets to build a model. This ANN model later on can be used to perform prediction of outcomes from a larger and different set of input data. ANN itself has been proven to be working on the other fields such as medical and business in performing prediction of cancer and stock market respectively. Currently, implementation in the oil and gas industry has been started around the world. Owing to the subsurface database available at VICO, data required to implement this ANN method is statistically sufficient. This paper will provide the workflow and methodology on using ANN to identify bypassed zones in VICO’s Semberah Field starting from input selection, result validation up to future rigless candidate.