Background
The optimization of process parameters in the pharmaceutical industry is often carried out according to the Quality by Design (QbD) concept. QbD also emphasizes that continuous improvement should be performed in life cycle management. Process parameters that are difficult to control in actual production could be regarded as noise parameters. In this study, a noise parameter was considered, an example of continuous improvement in the design space was provided.
Methods
The ethanol precipitation process of Carthami Flos (Honghua) extract was optimized based on the QbD concept. The critical process parameters (CPPs) were identified using a definitive screening design. Considering that the refrigeration temperature of industrial ethanol precipitation is often difficult to control with seasonal changes, the refrigeration temperature was treated as a noise parameter. The design space was then calculated using an exhaustive search-Monte Carlo method. The mathematical models were reestablished when more data were obtained and then the calculated probabilities of reaching the process standards were updated.
Results
The calculation procedure of design space based on an exhaustive search-Monte Carlo method was proposed. The density of the concentrated extract, ethanol concentration, the volume ratio of ethanol to concentrated extract, stirring time after ethanol addition, and refrigeration temperature were selected as CPPs. The mathematical models of CPPs and evaluation indicators were established, and the coefficient of determination of each model was greater than 0.81. The predictive performance of the models was good. After continuous improvement, the recalculated probability values were more reliable, the design space became larger.
Conclusions
The calculation of design space and the continuous improvement strategy considering a noise parameter was developed. In industrial production, it is also recommended to adopt this similar idea, that is, continuing to collect industrial data and regularly updating the mathematical models, which can further update the design space and make it more stable and reliable.