Minimizing total earliness and tardiness on a batch processing machine using a neural network approach

In this research, the problem of scheduling a single batch processing machine with non-identical job sizes is considered. The objective is to minimize the total earliness and tardiness penalties of all the jobs. Batch processing machine can process a group of jobs simultaneously as a batch as long as its capacity is not violated. The processing time of a batch is equal to the maximum processing time of all jobs in the batch. Since the problem under study is shown to be NP-hard, a neural network based approach which combines a heuristic algorithm and the iterative learning approach is proposed. The computational experiments are designed to compare the effectiveness of the neural network approach with the other heuristics. The results show that the proposed approach outperforms the other algorithms in terms of solution quality.