Abstract: Estimating assembly time of configuration product accurately is not only the basis of product sales quotation and delivery time, but also the key to gain a competitive advantage in mass customization environment. This paper analyses the configuration of the company’s existing products with Rough set and defines a time coefficient. Then a regression model on the relationship between the time coefficient and assembly time is built. According to the time coefficient of a new product, its assembly time can be estimated by using this model. In addition, the assembly time of transformers in a company is given to verify the effectiveness of this estimation model, and the error analysis is provided simultaneously.

Key words: mass customization, configuration product, Rough set, dependency of attributes, assembly time

摘要： 准确地估算配置产品装配时间是确定产品销售报价和交货期的基础，是在大规模定制生产环境中取得竞争优势的关键。应用粗糙集的方法对企业现有产品的配置进行分析，定义了一个时间系数，并用回归分析的方法建立起该系数与装配时间的数学关系模型。此模型可通过新产品的时间系数对其装配时间进行估算。此外，以某企业的变压器装配时间为例对所提出的估算模型进行了验证，并对其误差作出了分析。

关键词: 大规模定制, 配置产品, 粗糙集, 属性依赖度, 装配时间