The use of many-valued and nonmonotonic logics for design decisions

Abstract

It is shown how to use techniques and properties of many-valued and nonmonotonic logics for the construction of programs. The design decisions of program development depend on nonfunctional properties (like complexity, influences of the application area and target systems, etc.). If one develops a program using a transformation system, a design decision corresponds to the choice of a transformation rule. The connection between nonfunctional properties and design decisions is described by the use of logic. An inference mechanism helps in choosing a transformation rule. A classical inference mechanism is not sufficient, because the knowledge on the connection between nonfunctional properties and transformation rules is uncertain and incomplete. It is shown that the use of a many-valued and nonmonotonic inference mechanism leads to more adequate results.<>