Case-based decision support
Albert A. Angehrn, Soumitra Dutta · Communications of the ACM · 1998
Decision support systems (DSS) are interactive, computer-based systems helping decision-makers (individuals and/or groups) to solve various semi-structured and unstructured problems involving multiple attributes, objectives, and goals. The majority of DSS today are either computer implementations of mathematical models (e.g. optimization algorithms) or extensions of database systems and traditional management information systems. More recently, researchers [1, 2, 12] have argued that the role of DDSs should not be limited to “imposing relatively simple and rigid mathematical (or formal) artifacts on the rich, natural, selforganizing, and knowledge-producing processes of individual and social decision making ” [12]. They have proposed that research in DSS should focus on defining “flexible environments in which learning about a decision situation can take place [1],” i.e. support decision makers (DMs) in recursively redefining their problems and updating their aspirations, until a form of stability (cognitive equilibrium [12]) is obtained. Such an emphasis on learning has led to a different type of “learning-oriented ” DSS [1, 2, 12] to support DMs in incrementally exploring a decision situation and in reaching a cognitive equilibrium while avoiding typical decision-making biases observable in practice. Learning Oriented DSS In operational terms, learning-oriented DSS are symbiotic systems in which the human DM and the computer-based system are viewed as resources in the decision making process. As illustrated in Figure 1, the role played by the human component consists