Objective:

2

To carry out the risk assessment on the factors in the process of granulation fluidized bed of traditional Chinese medicine(TCM) by using failure model and effect analysis(FMEA) and Bayesian network(BN)

in order to effectively control risk factors and improve product quality.

Method:

2

The risk analysis of the fluidized bed granulation process was carried out by FMEA and the selected medium risk and high risk factors were taken as the main control points

the corresponding BN was established. The sensitivity analysis was used to screen out the main risk factors affecting particle fluidity

particle size uniformity

solubility and product cleanliness

the occurrence probability of each risk factor was determined by the evidence of unqualified particle quality

finally

taking fluidized bed granulation process of Sanye tablets as an example

the FMEA and BN were combined into the risk assessment process to verify the effectiveness and reliability of the method.

Result:

2

Based on the middle and high risk points of fluidized bed process

particle size of raw materials

moisture content and hygroscopicity of raw materials

dosage

concentration and addition amount of binder

cleaning degree and integrity of collection bag

and nozzle position

which were selected by FMEA

a fluidized bed granulation risk network with causality was constructed. Among them

hygroscopicity of raw materials

concentration and addition amount of binder

inlet temperature and atomization pressure were high probability risk factors

and the probability of occurrence were 55%

63%

59%and 58%

respectively. According to the Bayesian risk relationship network which controlled Sanye tablets fluidized bed granulation analysis results showed that the

P

values of inlet temperature

atomization pressure and concentration of binder were 0.003 4

0.032 6 and 0.041 8

respectively in the regression model of influencing factors and particle size uniformity

indicating that there was a significant correlation between the three factors and the particle quality

which was basically consistent with the conclusion obtained by FMEA-BN method.

Conclusion:

2

The combination of FMEA and BN for visualized risk assessment of fluidized bed granulation helps to effectively control the risk factors in the granulation process

reduce product quality risks and provide strong support for the improvement of granulation process of TCM.