We present an overview of different theories of explanation from the philosophy and cognitive science communities. Based on these theories, as well as models of explanation from the knowledge-based systems area, we present a framework for explanation in case-based reasoning (CBR) based on explanation goals. We propose ways that the goals of the user and system designer should be taken into account when deciding what is a good explanation for a given CBR system. Some general types of goals relevant to many CBR systems are identified, and used to survey existing methods of explanation in CBR. Finally, we identify some future challenges.

No abstract

In this article we present DrillEdge — a commercial and award winning software system that monitors oil-well drilling operations in order to reduce non-productive time (NPT). DrillEdge utilizes case-based reasoning with temporal representations on streaming real-time data, pattern matching and agent systems to predict problems and give advice on how to mitigate the problems. The methods utilized, the architecture, the GUI and development cost in addition to two case studies are documented.

No abstract

Over the past decade, oil and gas service companies have implemented real-time operations and remote surveillance centers to reduce the non-productive time (NPT) associated with wellbore integrity and downhole failures. Despite these efforts, NPT remains at around 15 to 35% of the total well cost. Companies seek to reduce NPT with knowledge management systems collecting best practices, knowledge cubes along the wellbore, and lessons learned. On the other hand, drilling activities are inherently data intensive, and as the amount of available data increases, it makes it correspondingly difficult for engineers to interpret the situation in a short period of time. Case-based reasoning (CBR) is a method that can use the real-time drilling data to index and automatically recall the various experiences that are now only available through dedicated knowledge management systems. Using this method, human experience can be collected in a company case base, linked to observed data, and automatically brought forward in real time when it is again relevant. In this paper, we introduce the CBR method and discuss some of the challenges in applying this method to drilling data. The DrillEdge computer system, based on this methodology is introduced, which uses the WITSML standard for drilling data acquisition, integration, and mass storage, making it possible to use in an integrated environment with current real-time operations and remote centers. It will also highlight the importance of a unique user interface to interpret the real-time results during drilling operations. An evaluation of the system was performed in which the system learned to recognize and, subsequently, predict problems when historical data from several land-drilling operations in Latin America were played back as simulations. A rigorous testing routine was applied to evaluate the capability of the CBR system to correctly identify potential risks and provide the user with remedial actions within a specified time-critical window.