The Application of Taguchi Method and Artificial Neural Network on Retail Apparel Markdown Strategy

• Main Authors: Yi-Chun Hung, 洪宜君
• Other Authors: Peitsang Wu
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/21311961126434126108

碩士 === 義守大學 === 工業工程與管理學系碩士班 === 98 === While facing globalization marketing issues, the retailers have to face the pressure of product diversities, short life cycles, and worldwide competitions. Retailers must now think of how to allocate resources, establish global supply chain, use flexible business model, increase product value and revenue, and establish long-term strategy to get a greater position in the world. If the retailers couldn’t predict the demands of the marketplace accurately, the consequence is to markdown the prices of goods with 10% discount, 20% discount or 30% discount. Different markdown levels will affect customer’s intensions to buy goods. Large discount means more buying. On the contrary, no discount means no buying. How much to markdown and when to markdown becomes very important for the retailers. Retailers can face from other retailers competition, markdown pressure all over the world at the same time, the mistake predict can lead to decreasing of product sales and revenues. The objective is to find how much quota a branch should have in order to pursue the maximal revenue. The main purpose of this research is to: (1) apply the Taguchi methods to seek optimal parameter design for the output performance of various inventory models, which include Quick Response, VMI, Newsboy and Target Weekly Supply, by using the signal to noise ratio to evaluate the output variations of these models; (2) incorporate the input parameters from the orthogonal array of Taguchi methods, and implement the input-output relationship by using artificial neural network to determine the better markdown strategies for the various inventory models.