Research has shown that harassed decision makers in decision support technologies for real-time and uncertain decision problems, often ignore crucial information, use inefficient strategies, and generate fewer alternatives. Examples of such problems are such as medical emergencies, traffic flow and military applications. Despite significant research on decision making support, researchers are limited by a narrow understanding of the most effectual methods to support decision-making in real-time and stressful situations. These decision problems often require up-to-the-minute information, dynamic response and conflict resolution. Intelligent software agents offer one approach to operationalizing a decision making support system in these circumstances by providing an environment that can adapt to speed and change. While there are no agreed upon criteria for determining whether a program is a software agent, in general, agents act on behalf of someone. Thus, an agent is a software package that carries out tasks for others autonomously, with the others being human users, business processes, workflows or applications.
Broadly stated, to achieve common goals, agents need to communicate, coordinate and cooperate with both the human user and other agents. This book is concerned with intelligent agents that are embedded in decision support systems and used to mediate the decision making process.