1 Improved Efficiency of Oil Well Drilling through Case Based Reasoning Paal Skalle Norwegian University of Science and Technology, Dept. of Petroleum Technology, N-7491, Trondheim, Norway (pskalle@ipt.ntnu.no) Jostein Sveen SINTEF, Dept. of Safety and Reliability N-7465, Trondheim, Norway (jostein.sveen@indman.sintef.no) Agnar Aamodt Norwegian University of Science and Technology, Dept. of Computer and Information Science N-7491, Trondheim, Norway (agnar.aamodt@idt.ntnu.no) Abstract A system that applies a method of knowledge-intensive case-based reasoning, for repair and prevention of unwanted events in the domain of offshore oil well drilling, has been developed in cooperation with an oil company. From several reoccurring problems during oil well drilling the problem of ”lost circulation”, i.e. loss of circulating drilling fluid into the geological formation, was picked out as a pilot problem. An extensive general knowledge model was developed for the domain of oil well drilling. About fifty different cases were created on the basis of information from one North Sea operator. When the completed CBR-system was tested against a new case, five cases with descending similarity were selected by the tool. In an informal evaluation, the two best fitting cases proved to give the operator valuable advise on how to go about solving the new case. Introduction Drilling of oil wells is an expensive operation, costing around 150 000 US $ pr. day, and any loss of time caused by unwanted events is costly. Some unwanted events are repeatedly occurring but still so complex that they are not easily solved. The necessary experience obtained by individuals or by the organization is difficult to transfer efficiently to those that need it. Lost circulation during oil well drilling is an unwanted event characterized by not obtaining any or part of the drilling fluid (also called mud) back to the rig in spite of a running mud pump. The failure type may be sub-divided into four categories, related to possible fractures or other undesired properties of the geological formation: Induced fractures; natural fractures, cavernous formation and permeable formation. The strength of the sedimentary formation is surpassed due to an unfortunate combination of operational events. The problem occurs on an average base once every well drilled, and may last for less than an hour, but sometimes it takes 14 days to solve it. It is too complex a problem to predict or solve by means of a mathematical simulator. The many necessary simplifying constraints would make the simulator unrealistic. The necessary experience to deal with the problem may not be available at the time of occurrence; the ”expert” may not be available, written experience is usually partial or scattered, or the problem is too delicate to select the optimal procedure based on on- site available experience. A promising approach in such situations is case based