Artificial intelligence can be used to recognize and anticipate dynamic situations. Several computational methods based on mathematical tools already exist, but most of the time their implementation is complex and takes a long time to execute.
In this article, we propose another learning and anticipation method in order to assist users in dynamic situations. We call it ‘scenario-based reasoning’ algorithm. It is inspired by case-based reasoning. It works with symbolic data and its aim is to make real-time predictions. To do so, manipulated knowledge is specially designed to limit our solution's complexity and to facilitate learning and anticipation.