A support vector machine based scheduling approach for a material handling system

This paper considers scheduling a multiple-load vehicle of a material handling system (MHS) in a manufacturing system. The orders of handling materials are generated dynamically by the manufacturing system. This paper proposes a support vector machine (SVM) based scheduling approach to tackle the scheduling problem. The approach generates training samples and uses the samples to train an SVM offline. The trained SVM is then used online. Online information at decision points are structured as an input to the SVM. The output of the SVM is used to make decisions such as “wait for next order” or “deliver immediately”. Simulation results show that the proposed approach outperforms other approaches.