An improved genetic algorithm for the berth scheduling with ship-to-ship transshipment operations integrated model
Marwa Al Samrout,Abdelkader Sbihi,Adnan Yassine
DOI: https://doi.org/10.1016/j.cor.2023.106409
IF: 5.159
2024-01-01
Computers & Operations Research
Abstract:In this paper, we consider a dynamic variant of Berth Allocation Problem with ship-to-ship transshipment in a container terminal, namely (DBAP STS ), where ships can arrive after the start of the planning plan. A novel mixed integer linear program (MILP) is developed to optimize the berthing schedule and build a transshipment connections planning between feeder and mother vessels. The aim is to reduce vessels dwell times in the terminal, and the penalty due to late vessels and decide the mode of transshipment needed. First, we develop a packing heuristic, and later we use an improved genetic algorithm based heuristic (GA) to solve the problem efficiently. We conducted a statistical analysis to identify relative importance and effective settings for parameters control of the GA, and we applied this algorithm with the control parameters settings determined to carry out the computational experiments on random generated instances. The proposed tailored method is able to solve the problem in an acceptable computing time for medium and large instances while a commercial solver was able to solve only small size instances.
computer science, interdisciplinary applications,engineering, industrial,operations research & management science