基于云计算的车辆调度系统_濮阳.pdf

摘要 : se of this thesis is to use an improved swarm intelligence algorithm to solve the vehicle scheduling problem in the logistics field, improve the overall satisfaction of logistics delivery services by improving the mathematical model of vehicle scheduling, and develop a vehicle scheduling system that can be deployed in cloud environment. In swarm intelligence optimization algorithm, ant colony algorithm uses a distributed computing method in its calculation, and has good compatibility with other algorithms. Cuckoo search algorithm is also easy to integrate with other algorithms. Hybrid algorithm can combine the excellent mechanisms of both. Therefore, this thesis proposes a hybrid adaptive ant colony algorithm that combines the Levy flight mechanism of cuckoo search algorithm to optimize the global search ability of the algorithm and prevent premature convergence in the search process. Finally, multiple test functions are used to verify the optimization characteristics of the hybrid adaptive ant colony algorithm, and a test case of path optimization is used to verify the effectiveness and feasibility of the proposed algorithm in solving the vehicle scheduling problem in logistics delivery. In the case of very limited delivery vehicles, prioritizing and differentiating services for customers can overall improve customer satisfaction. This thesis introduces a customer priority factor by analyzing customer characteristics and priority in logistics delivery vehicle scheduling problems, provides a logistics delivery strategy with priority division, and proposes a path optimization model based on customer priority for the delivery process. This method participates in the path optimization process under the condition of priority division, and designs and provides the best delivery path solution. Finally, a path optimization algorithm in vehicle scheduling problem is selected for simulation experiments to verify the effectiveness of the model in improving customer satisfaction in logistics delivery process. Finally, this thesis designs and implements a vehicle scheduling system based on cloud computing, providing a cloud computing environment for the system by building a Hadoop cluster, and applying the hybrid adaptive ant colony algorithm and vehicle scheduling based on customer priority in the system. The system has the main functions of information management, path optimization, and scheduling arrangement, and has passed basic system testing, verifying that the system meets design requirements and further confirming the strong practical value of the proposed algorithms and models. Keywords: cloud computing, swarm intelligence optimization algorithm, vehicle scheduling, path optimization