Abstract:

In recent years, international public health emergencies have occurred frequently, which severely affect shipping and trade exchanges between the affected areas. In particular, the stagnation and conversion of refrigerated containers affected by the pandemic prevents the demand for cold chain goods from being continuously met. At the same time, viruses adhere to the cold chain environment. As a result, the risk of transmission is relatively high. Therefore, it is necessary to study the optimization of international cargo transportation routes in a pandemic. Based on the designated cold storage mode of the first station, scientific planning for the transportation of refrigerated containers is conducted, and the safe and efficient transportation of cold chain goods is finally realized, aiming to minimize the transportation cost of refrigerated containers and the shortest transportation completion time, and establish a dual-objective optimization model for reefer container transportation route planning under the epidemic situation. The transportation volume of refrigerated containers on routes and ship schedules is determined by the real-time transportation volume sent from the destination port to the secondary cold storage during the land transportation phase. Asian ports A and B that have cold-chain goods with China are selected as the import locations, and domestic ports C and D that have a large overlap in hinterland demand are selected as the transportation route nodes for refrigerated containers. According to the established transportation routes, schedules, ship information and freight rates and other data are used to optimize the route planning of the refrigerated container transportation route in the marine and land transportation stages, and the non-dominated sorting genetic algorithm with elite strategy (NSGA-II) is used to solve the problem. The calculation results show that when the detection capability is 20TEU, the transportation cost of the refrigerated container is USD $44 210 to 75 350, and the transportation time is 1 422 to 3 000 h. The route analysis is conducted on one group of Pareto optimal solutions, and it is found that the route selection in each cycle is different. The first stop sensitivity analysis is performed on the detection capability of the designated cold storage. When the detection capability A_j is 20, the maximum transportation completion time is 3 000 h. When the detection capability A_j is expanded to 50, the maximum transportation completion time is 2 280 h, and the transportation completion time is reduced by 24%. The results show that the greater the detection capability of the first station designated cold storage, the smaller the total transportation time, but when the detection capability reaches a certain value, the impact on the entire cold chain transportation process is no longer significant.

Key words: shipping management, normalization of pandemic, first station designated cold storage, multi-objective optimization, NSGA-II algorithm

摘要：

近年来，国际突发性公共卫生事件频发，疫情波及区域间的航运及贸易往来受到严重影响，尤其是冷藏箱滞港及转换受阻使冷链货物需求无法得到持续满足，同时冷链环境下病毒附着及传播的风险较大，故研究疫情情况下的国际货物运输路径优化问题十分必要。因此，基于首站定点冷库模式，对冷藏箱运输路径进行科学规划，最终实现冷链货物安全高效的运输。以冷藏集装箱运输费用最小与运输完成时间最短为目标，建立疫情情景下进口冷藏箱运输路径规划双目标优化模型，对海运阶段各航线、班次上冷藏集装箱的运输量以及陆运阶段由目的地港发往二级冷库的运输量进行决策。选取与我国有冷链货物往来的亚洲港口A和港口B作为进口地，选取腹地需求有较大重叠的国内C港和D港作为冷藏集装箱运输节点，根据既定运输航线、班次、船舶信息、运价等数据对海运及陆运阶段的冷藏集装箱运输路径进行优化，并用带精英策略的非支配排序的遗传算法（NSGA-II）进行求解。计算结果表明：当首站定点冷库的检测能力为20 TEU时，冷藏集装箱运输费用为44 210~75 350美元，运输时间为1 422~3 000 h。对其中一组Pareto最优解进行具体航线分析，发现每个周期的航线选择各不相同。对首站定点冷库的检测能力进行敏感性分析，当检测能力A_j为20时，最大运输完成时间为3 000 h；当检测能力A_j扩大至50时，最大运输完成时间为2 280 h，运输完成时间降低了24%。结果表明：首站定点冷库的检测能力越大，运输总时间与运输总成本越小，但当检测能力达到一定数值时，对整个冷链运输过程的影响不再显著。

关键词: 航运管理, 疫情常态化, 首站定点冷库, 多目标优化, NSGA-II算法