Enhancing Decision Making in Critical Drilling Operations

Abstract

Abstract Many critical tasks in oil operations require cognitive skills for safe and efficient drilling. Oil well drilling is a complex process which frequently is leading to operational problems. One problem is wellbore cleaning. If the root cause of the problem was known, the treatment of the problem could be made more focused and efficient. However, to determine the root cause is difficult in this coupled process. One solution to this challenge is to apply knowledge intensive case based reasoning (KiCBR) methodology. KiCBR is a recent approach to problem solving and decision making where a new problem is solved by finding a set of similar previously solved problems, called cases, and reusing them in the new problem situation. This approach consists of two major parts: hierarchy of the concepts with different relation strength and case based reasoning. Concepts are abstracted to symbolic entities. Through the selected fields, data (real time logs, end of well plan and daily drilling reports, etc) are utilized to provide case data base. In this study, three main root causes were chosen; solids accumulation, problematic formation leading to future well cleaning and problematic formation causing well cleaning problems. This paper presents how to determine the root causes for poor hole cleaning episodes by means of the CBR (Case-Based Reasoning). All the involved parameters in each case were stored in the data base. Whenever an unexpected episode is encountered, highly similar case from the data base will come out for decision making. Furthermore, each case has specific and general lessons for those who are in charge of the operations. The match results were good enough but it needs to be improved through indicators that will reveal main cause more precisely. Introduction Oil well drilling is a complex process which frequently is leading to operational problems. In order to deal with the complexity of the oil well problems, knowledge intensive case based reasoning (KICBR) can be used. The main goals of KiCBR in drilling field are:Use similar experience and solution when the problems occur.○Solve problem in similar case that occurred before.○The predicted safe operation drillingAdding these solved cases to model for future prediction and improving quality of model.Borehole drilling is a costly process and when drilling crew face real problem during the drilling they have to react very fast and proper. An understanding of these problems, their causes, their anticipation and planning for solution is essential to control overall well cost and success in reaching the intended target zone. Hole cleaning problems. If hole cleaning is insufficient, it culminates in difficulties such as tight-hole, packing-off, stuck pipe, etc. The ultimate consequences of poor hole cleaning may include side-track or losing the hole completely. A lot of studies have been carried out by other researcher related to cleaning of deviated holes 1,2,3,4,5,6. But the results do not provide clear operational recommendations, perhaps mainly because such studies are focused on individual parameters. The specific application will be to reduce the risk of encountering downtime while drilling and hole control. The pre-drill prediction, drilling planning acts as guidance to expected instability, but real-time data are required to refine and revise the case base while drilling, matching if possible to improve planning in the future and reducing uncertainty ahead of the bit.