Title:Proactive Decision Support using Automated Planning

Abstract:Proactive decision support (PDS) helps in improving the decision making experience of human decision makers in human-in-the-loop planning environments. Here both the quality of the decisions and the ease of making them are enhanced. In this regard, we propose a PDS framework, named RADAR, based on the research in Automated Planning in AI, that aids the human decision maker with her plan to achieve her goals by providing alerts on: whether such a plan can succeed at all, whether there exist any resource constraints that may foil her plan, etc. This is achieved by generating and analyzing the landmarks that must be accomplished by any successful plan on the way to achieving the goals. Note that, this approach also supports naturalistic decision making which is being acknowledged as a necessary element in proactive decision support, since it only aids the human decision maker through suggestions and alerts rather than enforcing fixed plans or decisions. We demonstrate the utility of the proposed framework through search-and-rescue examples in a fire-fighting domain.