衣用洗涤剂是每个家庭常用的洗涤用品，它的去污能力是消费者购买时普遍关心的问题，但是洗涤剂在日常使用过程中，往往没有达到预期的清洁效果。衣用洗涤剂清洁能力受很多因素影响，本文着重分析衣用洗涤剂的各种成分和洗涤环境对各种污渍去污能力的影响。本文数据采用响应面设计方法进行试验，并根据试验数据建立响应面模型来分析洗涤剂的各种成分对各种污渍去污能力的影响关系。本文主要对可漂白性污渍和特殊衣物两类共7种污渍进行分析，以去污性能作为因变量，洗涤剂的主要化学成分和洗涤环境作为自变量来拟合模型，得到的模型用于预测洗涤剂对每种污渍的去污能力。在模型拟合过程中，我们运用逐步回归的方法对变量进行筛选，将不显著的变量剔除，从而得到对洗涤剂去污能力有显著影响的因素。回归得到的模型具有很好的预测能力，可用于优化后的洗涤剂配方去污能力的预测。将相同类别污渍的模型进行加权平均得到综合模型，模型中包含了一阶主效应和二阶交互作用及平方作用，各项前的回归系数对洗涤剂配方的优化提供了重要的信息。

Cleaning product is commonly used for each family. Consumers always concern the decontamination ability of detergent during their purchase. When we use the detergent, it often does not achieve the performance as we expect. We want to know what affect the decontamination ability of detergent. This article focuses on the research of the effect on decontamination ability by the detergent composition and washing conditions for the different kinds of stains. We fit each model by the response surface method and analyze the effect of detergent composition and washing conditions.This article mainly discusses the bleachable and special stains which contain seven kinds of stains. In the model, decontamination performance of each stain is the dependent variable and the main chemical compositions and washing detergent environments are the independent variable. The model can be used to forecast the decontamination ability of detergent. The method of Stepwise regression is used to select the significant variables. The model of each class stains is fitted by averaging the coefficient of the model each stain weighted by their standard deviation. It can provide some useful information for detergent formulation optimization.