采用基因库构建的季节性服装需求预测

发布时间：2018-01-08 17:21

本文关键词：采用基因库构建的季节性服装需求预测　出处：《纺织学报》2014年05期 　论文类型：期刊论文

【摘要】：季节性新款服装的历史销售数据少,生命周期短,故利用数据序列的趋势信息进行预测的方法不合适。鉴于服装企业的销售点(POS)数据中包含许多特征信息,利用层次聚类与动态时间弯曲距离从POS数据中进行特征信息抽取,以构建服装销售基因库。并提出基于基因库进行季节性服装需求预测的方法,根据新款服装信息、销售信息计算其与基因库中基因的相似度,再以相似度查找相似基因,用相似基因的信息预测新款服装的需求。这种方法可为服装企业利用POS数据进行定量预测,构建快速响应系统提供解决方案。最后,对某知名女装企业2011年POS数据抽取了基因库,并对其2012年3月1日到2012年7月31日的新款库存单位进行预测,结果表明该预测方法是有效的。
[Abstract]:Seasonal new clothing history sales data is small, life cycle is short. Therefore, it is not appropriate to use the trend information of the data series to predict. In view of the clothing enterprises' point of sale (POS) data contains a lot of characteristic information. The hierarchical clustering and dynamic time bending distance are used to extract the feature information from the POS data in order to construct the gene bank of clothing sales and the method of seasonal clothing demand prediction based on the gene bank is proposed. According to the new clothing information, the sales information calculates its similarity with gene in gene bank, and then looks for similar gene by similarity. Using the information of similar genes to predict the demand of new clothing. This method can provide a solution for garment enterprises to use POS data to quantitatively predict and build a rapid response system. The gene bank was extracted from the POS data of a well-known women's clothing enterprise on 2011, and its new inventory units from March 1st 2012 to July 31st 2012 were forecasted. The results show that the method is effective.
【作者单位】：嘉兴学院商学院;同济大学经济与管理学院;嘉兴学院图书馆;
【基金】：国家自然科学基金资助项目(71071113) 教育部人文社会科学研究青年基金项目(12YJCZH284)
【分类号】：F407.86;F224
【正文快照】： 很多服装企业采用自有品牌服装专业店零售商(speciality store retailer of private label apparel,SPA)营销模式。知名品牌服装企业Zara的SPA营销流程是:品牌概念确定—季节概念确定—商品构成立案—制作样衣—生产—销售—生产、销售[1]。目前,我国多数服装企业的SPA营销流

【参考文献】