DynaFLIP++: a knowledge interpreter that matches dynamic domain knowledge with static fuzzy constraints
Wolfgang Slany · 1996
Fuzzy constraints must be fine-tuned for a certain task. However, such fine-tuning cannot be done for all possible instantiations of a problem, and thus is generally done on protypical constraints. These templates are applied during run-time to a certain actual situation, i.e., dynamically adapted and applied only where necessary. This paper describes issues arising from the existance of such dynamic constraints and how they are handled in the FLIP++ optimization libraries. Examples from a steelmaking scheduling application are given. 1 Introduction Real-world decision making is optimizing under vague constraints of different importance, often using uncertain data, where compromises between antagonistic criteria are allowed. Typical application areas include scheduling, design, configuration, planning, and classification. To address such real-world decision making problems, we have designed and partly implemented a family of C++ libraries: *FLIP++ (pronounce: StarFlipPlusPlus; this r...