Abstract: The multi-stage decision process of integrated resource reallocation at import/export container terminals is addressed, considering the uncertainty of vessel arrival time at operational level. To satisfy both robustness and flexibility requirement, a dynamic decision framework based on 2-stage approximation is proposed. At each decision point, the dynamics of information and operation are analyzed, based on which the vessel classification is proposed and a stochastic-scenarios-based mixed integer programming model is established. The model is to determine both the 1st-stage fixed decisions and the 2nd-stage adjustable pre-decisions, with the objective of minimizing expected dwelling time of all vessels. Dependent on such decision logic, a two-stage Tabu Search is proposed to solve the optimization problem at each decision point. Numerical experiments verify the efficiency and effectiveness of the proposed decision method, which makes better utilization of updating certain and uncertain information.

Key words: container terminal, dynamic decision-making, scheduling, uncertainty