An Intelligent Decision Support System (IDSS) for Public Decisions Using System Dynamics and Case Based Reasoning (CBR)

摘要

This paper presents the design of an IDSS that allows the decision makers to identify key issues that matter for the future of a social system and helps them to improve the policy-making processes. The implementation is in process, so this paper presents the achievements up to the present. It combines IA techniques with qualitative models and Systems Dynamic Simulation. The selection of strategies and policies for complex social systems needs to take into account non-quantifiable variables. For this reason, we build models that allow the treatment of these kinds of variables. We propose a methodology divided into three phases. In the first one we build a model and simulate dynamic systems of particular scenarios, using this module as an analysis tool. This phase allows the detection of issues that matters. The results obtained by simulation are stored in a database and are used as entries in the reasoning process. So, they are the start point of the second phase. For this phase we use the CBR (Case Based Reasoning) technique, where each case is defined by a set of norms (common attributes), cases and indexes (attributes for discriminating cases), problem, solution and explanation. Different values of these attributes will be new cases. The last phase produces different solutions, giving to the decision maker explanations about pros and cons of these alternatives. In the case that none of these alternatives are accepted by the users of the IDSS, they can incorporate a new solution explaining that, for them, it is the best alternative to follow. It is important to emphasize that the IDSS is an instrument to promote and facilitate the attainment of a coherence and consensus between the decision makers.