A variable neighborhood search algorithm for the location problem of platoon formation center

Autonomous platooning technology has been considered a promising solution for reducing costs in the trucking industry. In trucking networks, the operation of autonomous truck platoons is constrained by many factors such as network topology, travel distance, and demand distribution, indicating the importance of deciding when and where to form and decompose platoons. This study investigates a novel location problem for platoon formation center (PFC) in a trucking network, which plays an important role in platooning operations. PFCs are specifically constructed as infrastructures for the formation and decomposition of truck platoons. Semi-automated truck platoons traveling between PFCs save labor and fuel. Therefore, each origin–destination (OD) pair in the trucking network can choose a beneficial transportation route via PFCs. This study aims to find the optimal PFC location scheme and OD demand routing scheme to minimize the total cost, which includes PFC construction, driver labor, and truck fuel costs. Accordingly, a mixed-integer linear programming model is developed. To effectively solve large-scale problems in practical applications, a variable neighborhood search algorithm embedded with a heuristic routing allocation algorithm is designed. We construct 20 instances with different characteristics, such as the number and distribution of customer nodes and PFC candidate nodes; construction cost of PFC candidate nodes; and amount and distribution of OD demand. We evaluate the efficiency of the algorithm based on the experimental results. Furthermore, the benefits of the new transportation mode are quantified, and planners are provided with insights into PFC construction.