A receding horizon approach for berth allocation based on random search optimization

An approach to address the berth allocation problem is presented that is based on the receding horizon paradigm. In more detail, berthing decisions are computed by solving an optimization problem at each time step aimed at minimizing the waiting times of vessels, exploiting predictions on the ship arrivals and berth occupancy over a moving window starting from the current time instant. A discrete time dynamic model is devised to forecast the state of the terminal in the forward window, and a computationally-efficient approximate solution method based on random search is proposed. The considered framework can be used either for real time planning or scheduling in advance. Simulation results are reported to show the effectiveness of the method in different terminal configurations, forward horizons, and traffic intensities, in comparison with state-of-the-art approaches.