THE TWO-ECHELON VEHICLE ROUTING PROBLEM WITH DYNAMIC SATELLITES FOR CITY LOGISTICS

摘要

The penetration of business-to-consumer online shopping increases the volume of home deliveries. This situation of more delivery points with smaller packages creates a unique logistics problem. In a big city, the efficiency of package delivery is highly affected by the traffic condition. However, most of the Vehicle Routing Problem (VRP) models assume that the traffic is under normal condition and do not consider traffic congestion. Many firms are not able to meet the desired delivery time or service level due to traffic congestion. Thus, some firms build more intermediate facilities, also known as satellites, to solve this problem. In reality, during traffic peak hours, smaller vehicles, such as motorcycles, have better mobility than the bigger ones. This is especially true in metropolitan areas where the road infrastructure is less developed. On the other hand, smaller vehicles mean more frequent cargo restocking. In this study, we formulated a routing problem that utilizes small size vehicles as carriers and trucks as dynamic satellites to handle more delivery points with smaller packages. This problem is named as two-echelon vehicle routing problem with dynamic satellites. In contrast to the traditional two-echelon vehicle routing problem model, city freighters replenish directly from the nearest trucks instead of the satellites. A heuristic algorithm was proposed to solve 2E-VRPDS. This algorithm was tested on several city logistics problems with different numbers of delivery points, package volume, and sizes of city freighters. The results show that the proposed model is more applicable to adapt to severe city logistics environment. It also produces shorter travel distance compared to the traditional two-echelon vehicle routing problem and eliminates the operating cost of satellite facilities.