Scheduling as a fuzzy multiple criteria optimization problem

Real-world scheduling is decision making under vague constraints of different importance, often using uncertain data, where compromises between antagonistic criteria are allowed. The author explains in theory and by detailed examples a new combination of fuzzy set based constraints and iterative improvement repair based heuristics that help to model these scheduling problems. The mathematics needed for a method of eliciting the criteria’s importances from human experts are simplified. The author introduces a new consistency test for configuration changes. This test also helps to evaluate the sensitivity to configuration changes. The implementation of these concepts in the fuzzy logic inference processor library FLIP¸+¸+, in the fuzzy constraint library ConFLIP¸+¸+, in the dynamic constraint generation library DynaFLIP¸+¸+, and in the heuristic repair library DéjàVu is described. All these libraries are implemented in a layered framework enhanced by the common user interface InterFLIP¸+¸+. The benchmark application to compare the fuzzy constraint iterative improvement repair heuristic with constructive methods based on classic constraints is a scheduling system for a continuous caster unit in a steel plant.