A New Algorithm for Demand Prediction of Fresh Agricultural Product Supply Chain

Xinwu Li

doi:10.19026/ajfst.6.80

Abstract

Demand prediction plays a key role in supply chain management of fresh agricultural products enterprises and its algorithm research is a hotspot for the researchers related. A new algorithm for demand prediction of supply chain management of fresh agricultural products is advanced based on BP neural network and immune genetic particle swarm optimization algorithm. First, the deficiencies of traditional BP demand prediction models are analyzed. Second, the BP neural network and immune genetic particle swarm optimization algorithm are integrated and some measures are taken to overcome the deficiencies of traditional BP demand prediction models and calculation flows of the presented algorithm are redesigned. Finally, the presented algorithm is realized with the data from certain fresh agricultural products supply chain and the experimental results verify that the new algorithm can improve effectiveness and validity of demand prediction for fresh agricultural products supply chain.

Highlights

Along with the changes of market environment, the focus of supply chain management begins to keep its eye to market demand management, pay much more attention to market demand
For fresh agricultural product enterprises, due to strict requirement for the storage period of products, in case of prediction error on inventories demand, enterprises will suffer heavy loss, so the research of demand prediction for fresh agricultural products based on supply chain management has become the hotspot for the researcher and corporations related (Rudulf, 2012)
The study takes Immune Genetic Particle Swarm Optimization algorithm (IGPSO) to modify and improve BP Neural Network (BPNN) model to overcome the question of slow convergence speed of original BPNN

Summary

Introduction

Along with the changes of market environment, the focus of supply chain management begins to keep its eye to market demand management, pay much more attention to market demand.

Results

Conclusion