Knowledge Acquisition and Qualitative Reasoning for Flood Control

In this paper, we will develop an inference system for supporting flood control by using qualitative reasoning which is one of the methods of artificial intelligence. First we will refer to difficulties which occur when we apply the qualitative reasoning to flood control as well as general difficulties inherent to qualitative reasoning. We will propose how to solve the former type of difficulties. In our system, a method of controlling inference flow eliminates these difficulties. Finally, we apply the system to a real flood caused in the Managawa river.