Abstract: In the context of sea-rail intermodal transport, it is crucial to optimize the benth allocation of container terminal by taking into account the characteristics of port vessels and railway trains, in order to improve the efficiency of port operations. A berth allocation optimization model for sea-rail intermodal container terminals was established, taking into account resource constraints such as dock shoreline, berth depth, and quay bridges, as well as constraints such as ship service priority and preferred berths, and arrival and departure time of railway container trains, whose optimization objective was to minimize the dwelling time of ships in port and the penalty time of berth deviation. According to the features of the proposed model, an adaptive genetic algorithm was designed to solve the problem. The data of arriving ships within 48h of the planned period were selected for case study. Finally, the berth allocation scheme of sea-rail intermodal container terminal that met the requirements of ship service priority and berth preference as well as accorded with the train operation planning was obtained by MATLAB numerical experiment. The numerical experiment results show that under the condition of sea-rail intermodal transport optimization, the average waiting time of ships is 2.9 min, and the maximum waiting time is 11.4 min. At the same time, the distance between the berthing position of ships and the preferred berth is not more than 100 meters, which means the dwelling time of ships in the port and the difference between preferred berths are well controlled. This study can further optimize the berth allocation scheme of the sea-rail intermodal container terminal, which is conducive to improve the efficiency of the sea-rail intermodal container transshipment.

Key words: traffic and transportation engineering; sea-rail intermodal transport; collaborative optimization; berth allocation; adaptive genetic algorithm; train operation planning

摘要： 在海铁联运背景下，协同考虑到港船舶及铁路班列特点，优化集装箱码头泊位分配，对于提高港口作业效率至关重要。在满足码头岸线、泊位水深、岸桥等资源约束前提下，综合考虑船舶服务优先级和偏好泊位、铁路集装箱班列到发时间等约束，以船舶在港停留时间和泊位偏移惩罚时间最小为优化目标，建立海铁联运集装箱码头泊位分配优化模型；针对模型特点，设计自适应遗传算法进行求解，选取计划周期48 h内的到港船舶数据进行案例分析，通过MATLAB数值实验，得到符合船舶服务优先级和泊位偏好要求且契合铁路班列开行计划的海铁联运码头泊位分配方案。数值实验结果表明：在海铁联运协同优化条件下，船舶平均等泊时间为2.9 min，最大等泊时间为11.4 min，同时船舶的靠泊位置与偏好泊位之间的距离均不超过100 m，船舶在港的停留时间及偏好泊位差都得到了较好的控制。研究可进一步优化海铁联运集装箱码头泊位分配方案，有利于提高海铁联运集装箱在港转运效率。

关键词: 交通运输工程；海铁联运；协同优化；泊位分配；自适应遗传算法；铁路班列开行计划