目的 基于质量源于设计（quality by design，QbD）理念，建立红花温浸提取工艺的设计空间并进行验证。方法 以红花为模型药，通过查阅文献及前期研究经验获得关键评价指标；采用鱼骨图结合失效模式与效应分析（failure mode effects analysis，FMEA）确定影响红花温浸提取过程的关键工艺参数（critical process parameters，CPPs），通过Box-Behnken实验设计建立了CPPs与关键评价指标的数学模型。结果 以总黄酮提取量、羟基红花黄色素A（HSYA）提取量以及总固体提取量为关键评价指标，通过鱼骨图结合FMEA确定了加水倍量、提取温度、提取时间及提取次数为CPPs。Box-Behnken实验方差分析结果显示所建立模型的P值均小于0.000 1，表明所建立的模型具有较好的预测能力，得到推荐的操作空间为提取次数2次，提取时间2.5 h，加水倍量为23.5～25 mL/g生药，提取温度65～71℃。结论 红花温浸提取设计空间的建立，提高了提取工艺参数与提取液质量之间的关联性，为设计空间法在中药领域的适用性提供了参考。

Objective To establish and verify the design space for the extraction process of Carthami Flos (CF) based on the concept of quality by design (QbD). Methods The safflower was used as a model drug. The critical evaluation indicators were determined through literature and previous research experience. Fishbone diagram and Failure Mode Effect Analysis (FMEA) were carried out to determine the critical process parameters (CPPs). The mathematical models of CPPs and critical evaluation indicators were established using the Box-Behnken experimental design method. Results The critical evaluation indicators were the extraction amount of total flavonoid, hydroxysafflor yellow A (HSYA), and total solids. The water-adding amount, extraction temperature, extraction time, and extraction times were determined as the CPPs by the fishbone diagram and FMEA. The variance analysis results of Box-Behnken experiments showed that the P values of established models were less than 0.000 1, indicating that the models have good prediction ability. The recommended operating space were as follows:the extraction times were twice and 2.5 h for each time, the water-adding amount was 23.5-25 mL/g crude drug and the temperature was 65-71℃. Conclusion The establishing of design space for CF extraction improves the correlation between the extraction process parameters and extract quality, which provides a reference for the applicability of the design space approach in the field of CMM.