Abstract
This paper describes a study of the container truck routing problem for pickup and delivery of fully-loaded and empty containers in a local area, considering its operational flexibility. In a container truck operation, an empty container can be transported to a container yard or customer. In addition, since a container can be separated from a truck, two consecutive tasks, a delivery task followed by a pickup task at the same customer can be served by a truck or different truck. Such features of container trucks bring up street turn and decoupling strategy in container truck routing and scheduling. Contrary to pervious studies, this paper considers all pickups and deliveries of containers as independent tasks and two types of consecutive tasks from the perspective of customers incorporating the flexible operation of truck. In a simplified truck routing network, the total operating cost traveled by a truck with an empty container and without a container are minimized. The problem is formulated as an asymmetric multiple- vehicle Traveling Salesman problem with Time Windows (am-TSPTW) including driverâ??s work-shift hours. A heuristic approach is developed based on an insertion heuristic and a Variable Neighborhood Search (VNS) to solve the problem more efficiently. Result from numerical experiments showed that container truck routes reflecting the two flexibilities reduced the total cost of operation compared with routes without the flexibility. Hence, the proposed problem and solution approach enable a trucking company to accomplish the tasks under two strategies with reduced cost and improved operational efficiency.