As of today, no well is drilled without problem. Oil and gas companies spend about $ 25 billion annually on drilling. Unfortunately, not all of that money is well spent. A significant portion, about 20% is attributed to losses. These include loss of materials & drilling process continuity. In particular, wellbore stability related problem remedy cost is substanting and such problem 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. The objectives of this paper is to a problem analysis method called Case Base Reasoning( CBR) and its application in terms of a new level of active computerized support for information handling, decision support for wellbore instability cases and prediction of potential unwanted events in the drilling engineering domain.CBR is well suited for dealing with the complexity of wellbore stability problems, and the large number of parameters involved, experience is stored in cases which are linked to a model of general domain knowledge. The general domain knowledge is containing information with explanatory support to case knowledge. A knowledge Tool named "Creek" (research version) has been used here to create the ontology framework. This paper mostly addresses how to make a case from real case data, create symbolic entities including case retrieval methodology and finally diagnosis similar problem cases.The cases resented are all related to improving the drilling plan, in such a way that the problem can be planned away throughts the next drilling plan. The resulting outcome from pased cases acts as input for making better drilling plan to avoid similar problems in the future.
Introduction:Wellbore instability problems are pack off, stuck pipe, lost circulation, poorhole cleaning and down hole equipment failure. These unwanted events are repeatedly occurring but still so complex that they are not easily solved. In our daily drilling operation, most practical problems need to be solved fast. Since most practical problems have occurred before, the solution to the problem is hidden in past experience, experience which either is identical or just similar to the new problem. Such problems can be solved efficiently by storing and then reusing similar experience, i.e. CBR. A similar, previous experience is a good initial approach to solving the problem.