An improved memetic algorithm to enhance the sustainability and reliability of transport in container terminals

This paper improves our previous attempts in which we studied a combination of an evolutionary algorithm (EA) and Monte Carlo simulation (MCS). Results of those studies showed the process of sampling in MCS is very time consuming. This prevents the EA from producing an accurate estimation of the robust solutions within reasonable time. Thus the present work improves the performance of the EA to make it possible to reach high quality solutions in reasonable time, therefore yielding a number of more practical solutions in real cases. Firstly, it proposes a new sampling technique to generate samples that better reflect the worst-case scenarios. This helps the EA to find more robust solutions using smaller sample sizes. Secondly, it proposes a new adaptive sampling technique to adjust the sample size during evolution. Subsequently, to evaluate the proposed algorithm we tested it in a typical environment with shuttle transport tasks: container terminal. Experimental results show that such improvements led to a significantly improved performance of the EA, thus making the proposed algorithm perfectly usable for empirical cases.