Being able to predict the performance of a Case-Based Reasoning(CBR) system against a set of future problems would provide invaluable information for design and maintenance of the system. Thus, we could carry out needed design changes and/or maintenance tasks to improve future performance in a proactive fashion.

In this talk, we will present a novel method for identifying regions in a case base where the system gives low confidence solutions to possible future problems. We will show the experiments performed at the RoboSoccer domain and how encountered regions of dubiosity help us to analyse our case base and the reasoning mechanisms of a CBR system.