A neural-network approach to delivery time estimation for 300mm automatic material handling operations

This paper presents a neural-network-based approach for prediction of average delivery times of lots that move from one intrabay loop to the other in 300 mm AMHS. A neural network model is developed for each intrabay AMHS. Inputs to the proposed neural network model are the combination of intrabay loadings, automated material handling resources, and ratios of priority lots against regular ones. A discrete-event simulation model based on the AMHS in a local 300 mm fab is built. Its outputs are adopted as the inputs for training the neural network model with the back propagation method. This neural-network approach is applicable to implementing a transport time estimator in fab scheduling and dispatching functions in realizing fully automated 300 mm manufacturing.