In order to improve the accuracy of port logistics demand prediction, the improved Particle Swarm Optimization algorithm, Grey Model and Neural Network are combined to construct an Improved Particle Swarm Optimization Grey Neural Network(IPSO-GNN) prediction model, in which the improved Particle Swarm Optimization algorithm is used to find the weight and threshold of the Grey Neural Network to improve the accuracy of the prediction. Using the logistics demand data of Dalian Port, the prediction effect of the proposed IPSO-GNN model is compared with that of the BP Neural Network model, the Grey model, the Grey Neural Network model and the standard Particle Swarm Optimization Grey Neural Network model. The empirical results show that the IPSO-GNN model has high precision and strong stability, which can predict port logistics demand effectively.
Port Logistics Demand Forecast Based on Grey Neural Network with Improved Particle Swarm Optimization
LISS 2020 ; Kapitel : 10 ; 133-144
2021-04-11
12 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Forecast of sea-rail throughput of Xiamen Port based on improved grey prediction model
British Library Conference Proceedings | 2022
|Logistics Demand Forecast Based on Gray System Theory
British Library Conference Proceedings | 2007
|Logistics Demand Forecast Based on Gray System Theory
ASCE | 2007
|