The vehicle routing problems (VRP) in logistics distribution has far-reaching significance to logistics companies in reducing transportation costs and improving service quality and efficiency. A lot of research has been devoted to solve VRP. However, in practical applications, with the increase of constraints and optimization objectives, the dimension and difficulty of the problem increase exponentially, which will lead to a dramatic increase in the complexity of the algorithm and a decrease in performance. Therefore, a more efficient and simple algorithm is needed. Cuckoo Search Algorithm (CSA) is a new meta-heuristic algorithm proposed in 2009 with good global search capability and robustness, containing few parameters which has been validated to solve many optimization problems effectively. Therefore, this paper applies the CSA to solve the path planning problem of logistics vehicles, builds a vehicle routing problem with time windows (VRPTW) model, and validates that the CSA can effectively solve this optimization problem in an acceptable time by simulation.