Rule-oriented methods in problem solving

A heuristic, in the usual sense and certainly in the sense that this term is used in this paper, is a rough guide or principle, or underlying knowledge which is not expressed explicitly. Heuristic methods for solving problems can involve the incorporation of such knowledge into algorithms. This paper describes two methods for finding the solution to problems using heuristic methods. A solution to a problem is represented by a sequence of operators which transpose one problem situation to another. Problems are categorized into four types: reversible regular, irreversible regular, reversible irregular and irreversible irregular problems. The rule-oriented method described were is useful particularly for the reversible regular problems to which heuristic knowledge is applicable. Some examples of the application of heuristics are shown in solving the Missionary-and-Cannibal problem and the Tower-of-Hanoi problem.

Authors

• Masamichi Shimura Seiichiro Sakurai
• Frank H. George

Keywords

No keywords are indexed for this paper.