Periodic and Collaborative Allocation of Berth-Yard-Gate Resources at Container Terminals

In order to improve the efficiency of import/export container terminals in periodic environment, the periodical and collaborative allocation problem of key resources at container terminals, including berth, yard and gate is investigated. A mixed-integer linear programming model is first established for the integrated scheduling of three types of resources. This model takes into account the capacity restrictions, inter-relationships and periodicity requirements, and sets the objective of minimizing total dwelling time of all vessels. Furthermore, an adaptive genetic algorithm is proposed to find a solution. As for the algorithm, the upper level performs evolutionary search within the space consisting the encoded vessel priority lists, while the lower level decodes each vessel priority to generate a complete resource allocation plan by heuristics, and returns its evaluation for upper level iteration. Numeral experiments shows that with the collaborative allocation of berth-yard-gate resources, the dwelling time of vessels in the execution of periodic plan is shortened by 20% comparing to traditional two-stage method.