Abstract: The escalating demand for maritime transportation has given rise to congestion challenges in both domestic and international ports, resulting in significant economic losses for the shipping industry. In practice, ports must strategically plan vessel scheduling in advance, during periods of heavy vessel traffic. Vessel scheduling requires not only the consideration of objective port conditions and pilot availability but also the randomness of a scheduling plan to mitigate uncertainties in port operations. This paper introduces a two-stage stochastic programming model to address the vessel scheduling problem considering uncertain port operation time. An algorithm that combines sample average approximation with adaptive large neighborhood search is proposed to solve the model. Additionally, an internal program is developed to automatically generate optimal timetables based on the sequence of waterways and the scheduling of pilots, tackling issues related to dynamic changes in waterway capacity, pilot transportation, and vessel departure time. Finally, numerical experiments provide evidence that the proposed algorithm outperforms the commercial solver CPLEX in terms of both speed and accuracy, showing practical significance in solving large-scale instances of such problems. Furthermore, sensitivity analysis conducted on the parameters of the sample mean approximation method and the coefficients of uncertain port operation time verifies the rationality and practical value of the proposed model and algorithm.