Rule-Based Policy-Level Model of Nonsuperpower Behavior in Strategic Conflicts

摘要

This report describes the second-generation Scenario Agent, a rule-based, policy-level computer model of nonsuperpower behavior in strategic conflicts. Scenario Agent is a model within the Rand Strategy Assessment Center (RSAC) war gaming system. The report reviews the work completed on the model. hte ratonale behind the models 7s rules, and the need for refinement of the rules. A primary goal of the RSAC is to model crisies and conflicts that could arise between the United States and the Soviet Union, including strategic nuclear war. RSAC analysts are interested in learning more about the role of strategic nuclear forces in deterring war and the role of lesser conflicts as paths to major war. RSAC analysts are also interested in nonnuclear conflicts in their own right, so long as they involve the United States and the Soviet Union as the major contenders. Because nonsuperpowers can play significant roles in U.S.-Soviet conflicts, the Scenario Agent model describes their possible behavior when that behavior may have a major impact on U.S.-Soviet conflicts.