Research On Revenue Maximization Method Of Logistics System Based On Artificial Neural Network

This thesis abstracts the basic characteristics of the logistics system through the development process and manifestations of the traditional logistics system and the modern logistics system.It quantitatively analyzes the logistics operation cost based on combining the time characteristics of the logistics system,and qualitative analysis based on functional features of logistics system.A comprehensive time and cost optimization strategy was proposed,combining the effects of time and price based on cost on customer satisfaction.A dynamic cost input and revenue analysis method was proposed,and a realization model for maximizing the revenue of the logistics system was established.Combining the basic characteristics of the operation of logistics and the maximization model of profit,this thesis introduces the realization of the objective of maximizing the revenue of the logistics system assisted by artificial neural network.The neural network is used to forecast the future revenue data and cost data of the logistics system,and the revenue maximization management model is used to help the logistics system make reasonable management decisions.This thesis discusses the construction of a logistics system revenue maximization model based on BP neural network.It mainly includes the following aspects: It discusses the background knowledge of BP neural network and the steps to maximize the benefits of logistics system.Discussing the key technologies for establishing a predictive model based on BP neural network.Including the selection of samples and data preprocessing,the selection of input variables and output variables,the determination of the number of nodes in the hidden layer,the selection of initial weights and thresholds,the selection of activation functions,training algorithms and training parameters.Finally establish a reasonable network model.Finally,using the simulation data of the virtual simulation platform as a sample,based on the MATLAB neural network toolbox to establish a neural network prediction model.Comparing and analyzing the prediction values and the simulation data of the virtual simulation platform,it verifies the feasibility of the realization of the goal of maximizing the benefits of neural networks for logistics systems,and further enriches the management theory of the logistics system.