Postal and courier companies are under the pressure of minimizing operational costs while meeting requests for increasingly high service levels. We model their delivery problem as a dynamic multi-period general routing problem (DMPGRP) with the aim of minimizing the total cost over a given planning horizon. In most modern delivery systems, fluctuating demand volumes dynamically reveal themselves over time; in addition, service differentiation is highly important. Both these characteristics are considered in our DMPGRP. An effective solution strategy for this problem selects demands that must be fulfilled day by day, after distinguishing them according to priority classes. It then constructs the vehicle routes by using a heuristic algorithm based on the adaptive large neighbourhood search. The results of an extensive computational study are reported.

邮政和快递公司面临着将运营成本降至最低的压力，同时又要满足日益提高的服务水平的要求。我们将其交付问题建模为动态的多周期通用路由问题（DMPGRP），目的是在给定的计划范围内将总成本降至最低。在大多数现代交付系统中，需求量的波动会随着时间的流逝而动态地显示出来。此外，服务差异化非常重要。DMPGRP中考虑了这两个特征。针对此问题的有效解决方案是，在根据优先级对它们进行区分之后，选择必须每天满足的需求。然后使用基于自适应大邻域搜索的启发式算法构造车辆路线。