面向班型动态生成的地服人员排班算法

面向班型动态生成的地服人员排班旨在动态生成班型，并将员工分配到班型中，以班型为航班地面保障单元.现有面向班型的人员排班算法是建立在固定班型数和班型内的员工资质数，未考虑班型数未知的重要前提.为此，提出面向班型动态生成的地服人员排班算法，其核心思想是通过block Gibbs抽样迭代优化班型内人员构成、班型内航班集和班型生成.在某机场值机人员的数据集中验证此算法，实验结果表明，在满足员工层次资质、员工白夜班和班型动态生成的约束下，算法能够生成合理的班型.

Dynamic Group Generation for Airport Ground Staff Scheduling

The problem of scheduling airport ground staffs is to dynamically generate groups and assign employees to groups. The existing group-specific staff scheduling algorithms are built on the fixed number of groups. Nevertheless, it is impractical to know the number of groups in advance. An algorithm is proposed to address the issue via dynamically generating groups. Its key idea is to employ Gibbs sampling with replacement to iteratively optimize three sub-problems: assigning staffs to groups, allocating flights to groups and group generation. Experimental results on a real dataset demonstrate that the proposed algorithm can generate reasonable groups under the constraints of skill requirements for flights and shifts for employees.