Based on the concept of quality by design(QbD), the Box-Behnken design-response surface methodology combined with standard relation(SR) and analytic hierarchy process(AHP)-entropy weight method(EWM) was applied to optimize the extraction process of the classic prescription Yihuang Decoction. The content of geniposidic acid, phellodendrine hydrochloride, and berberine hydrochloride in Yihuang Decoction, the extract yield, and fingerprint similarity were used as the critical quality attributes(CQAs) of the extraction process. The extraction time, water addition, and extraction times were used as the critical process parameters(CPPs). After determining the levels of each factor and level through single-factor experiments, response surface experiments were designed according to the Box-Behnken principle, and the experimental results were analyzed. The SR between each sample and the reference sample under various evaluation indicators of different extraction parameters was calculated. The weights of the five evaluation indicators were determined using AHP-EWM, followed by comprehensive evaluation. A function model between CPPs and CQAs characterized by comprehensive scores was established to predict the optimal extraction process parameters. In the final comprehensive weight coefficients, the yield rate accounted for 43.1%, and the content of berberine hydrochloride, phellodendrine hydrochloride, and geniposidic acid accounted for 35.1%, 6.3%, and 15.5%, respectively. After comprehensive score analysis with SR, the established second-order polynomial model was statistically significant(P<0.01, and the lack of fit was not significant). The predicted optimal extraction conditions for Yihuang Decoction were determined as follows: 8-fold volume of water, extraction time of 1.5 h, and extraction once. The mean comprehensive score of the validation experiment was 85.77, with an RSD of 0.99%, and it met the quality control stan-dards for the reference sample of Yihuang Decoction. The results indicate that the optimized extraction process for Yihuang Decoction is stable and reliable, and the water extract is close in quality attributes to the reference sample. This can serve as a foundation for the research and development of granules in the future. Box-Behnken design-response surface methodology combined with SR and AHP-EWM can provide references for the modern extraction process research of other classic prescriptions.
Keywords: Box-Behnken design-response surface methodology; Yihuang Decoction; analytic hierarchy process; classic prescription; entropy weight method; extraction process; quality by design; standard relation.