目的 评价古方工艺与现代工艺样品质量的一致性，优化泻白散的提取工艺。方法 选取加水量、提取时间和提取次数为关键影响因素，采用单因素结合Box-Behnken响应面设计，以桑皮苷A、指纹图谱相似度、出膏率以及地骨皮乙素为评价指标，通过模糊层次分析法（fuzzy analytical hierarchy process，FAHP）-熵权法结合基准关联度确定各指标权重系数，计算综合评分，优选出与泻白散基准样品最为接近的现代提取工艺。结果 建立的工艺评价函数模型：综合评分＝SR_i,_j桑皮苷A×0.22＋SR_i,_{j指纹图谱相似度}×0.21＋SR_i,_j出膏率×0.34＋SR_i,_{j地骨皮乙素}×0.23，模型预测最佳提取工艺为加10倍水量，煎煮2次，每次90 min。验证试验综合评分均值为98.44，RSD为0.37%，符合基准样品标准的质量要求。结论 该提取方法稳定可行，能够为其他经典名方开发提供理论依据。

Objective To evaluate the consistency of the quality of ancient and modern process samples and optimize the extraction process of Xiebai San (泻白散). Methods The amount of water added, extraction time and extraction times were selected as the key influencing factors, and the single factor combined with Box-Behnken response surface design, mulberroside A, fingerprint similarity, paste yield and kukoamine B were used as evaluation indexes. The weight coefficient of each index was determined by fuzzy analytic hierarchy process (FAHP)-entropy weight method combined with benchmark correlation degree, and the comprehensive score was calculated to optimize the modern extraction process closest to the benchmark sample of Xiebai San. Results Establishment of process evaluation function mode: Comprehensive score = SR_{i,j mulberroside A} × 0.22 + SR_{i,j fingerprint similarity} × 0.21 + SR_{i,j extract rate} × 0.34 + SR_{i,j kukoamine B} × 0.23. The model predicted that the best extraction process was to add 10 times amount of water, decocted twice, 90 min each time. The average comprehensive score of validation test was 98.44, RSD was 0.37%, which met the quality requirements of the reference sample standard. Conclusion The extraction method was stable and feasible, which can provide a theoretical basis for the the development of other classic prescription.

R283.6

吉林省科技计划重点研发项目（2020040491YY）