Sales Forecast for Garment Companies with Cluster Analysis and Modified Neural Networks

Abstract:

Article Preview

For the companies of the garment industry, managers often dedicate their efforts to forecast the sales accurately while making decisions for marketing resource allocation and scheduling. Based on the historical database, this paper constructs a method to investigate the relationship of the relating factors and sales values. The proposed method combines the cluster analysis and modified neural networks to fulfill the sales forecast task. Firstly, the average linkage cluster algorithm is applied to cluster similar sales values. Secondly, a modified neural network is used to investigate the mapping relationship between those influencing factors and the sales clusters. The method employs a self-adjust mechanism to determine the structure of the neural network. The effectiveness of the proposed method is illustrated with a case study of a garment company in Shanghai.

Info:

Periodical:

Edited by:

Quanjie Gao

Pages:

490-495

DOI:

10.4028/www.scientific.net/AMM.127.490

Citation:

L. Yu and Y. Chen, "Sales Forecast for Garment Companies with Cluster Analysis and Modified Neural Networks", Applied Mechanics and Materials, Vol. 127, pp. 490-495, 2012

Online since:

October 2011

Authors:

Export:

Price:

$35.00

In order to see related information, you need to Login.

In order to see related information, you need to Login.