Minimizing makespan for parallel batch processing machines with non-identical job sizes using neural nets approach

This paper aims at minimizing the makespan for parallel batch processing machines with non-identical job sizes (NPBM) using neural nets (NN) approach. NN approach was proved effective to solve combinatorial optimization problems but no application to NPBM problems hitherto. This research provides new methods to code the NN with the introduction of Master Weight Matrix (MWM). Computational experiments show that the NN approach outperforms other algorithms in various circumstances.