The economy of production in flexible manufacturing systems (FMS) depends mainly on how effectively the production is planned and how the resources are used. This requires efficient and dynamic factory scheduling and control procedures. This paper addresses two knowledge-based scheduling schemes (work cell attribute oriented dynamic schedulers "WCAODSs") to control the flow of parts efficiently in real-time for FMS in which the part-mix varies continually with the planning horizon. The present work employs a hybrid optimisation approach in the generalised AI framework. A genetic algorithm that provides an optimal combination of a set of priority dispatching rules, one for each work cell "WC" (WCwise pdr set), for each of the problem instances characterised by their WC attributes, is used for generating examples. The WC attributes reflect the information about the operating environment of each individual WC. Two inductive learning algorithms are employed to learn the examples, and scheduling rules are formulated as a knowledge base. The learning algorithms employed are: the Genetic CID3 (Continuous Interactive Dichotomister3 algorithm extended with genetic program for weight optimisation) and the Classification Decision Tree algorithm. The knowledge base obtained through the above learning schemes generates robust and effective schedules intelligently with respect to the part-mix changes in real-time, for makespan criteria. The comparison made with a GA-based scheduling methodology shows that WCAODSs provide solutions closer to the optimum.