Objective To establish a modeling method for an animal model simulating the decline of digestive function after a large amount of tissue necrosis of the pancreas due to acute injury after severe acute pancreatitis (SAP). Methods Twenty male SD rats were randomly divided into the model group and the sham operation group according to the random number table method, with 10 rats in each group. First, the SAP model was established by retrograde bile duct injection of sodium taurocholate in the model group, whereas the sham operation group received physiological saline injection. Fluid infusion began 2 hours later, twice a day, with an 8-hour interval, for 2 days. The traditional Chinese medicine Dachengqi Decoction without Decoction Granules was formulated into a suspension in proportion and administered by gavage once at 18 hours and once at 24 hours after the operation to ensure the blood volume of rats and reduce inflammatory damage. Normal drinking water was allowed 48 hours after modeling. After 72 hours, ordinary feed was given for feeding. The feeding lasted for 14 days (the total duration of the experiment was 17 days). The body weight, vitality status and stool characteristics of the rats were observed and recorded on the day of open feed feeding and 14 days later. Fourteen days after feeding, the animals were sacrificed and samples were collected for examination of blood glucose, fecal elastase and hematoxylin-eosin staining pathological scores. Results All 10 rats in the model group were successfully modeled with a 100% survival rate. The body weight of rats in the model group 14 days after ordinary feeding was lower than that on the day of open diet [(180.80±4.39) vs. (222.90±6.14) g, P<0.001], and lower than that of rats in the sham operation group 14 days later [(180.80±4.39) vs. (221.70±7.45) g, P<0.001]. Compared with the sham operation group, inflammatory cell infiltration injury still existed in the pancreatic tissue of the model group, and some pancreatic tissues showed pathologically related changes of chronic injury. The pathological score of the model group was higher than that of the sham operation group [7.5 (7, 9) vs. 0 (0, 0), P<0.001]; the blood glucose concentration increased [(13.000±1.531) vs. (8.070±0.851) mmol/L, P<0.001]. The secretion of fecal elastase, a metabolite of trypsin in vivo, was significantly decreased [(5.451±0.936) vs. (8.593±1.105) mg/mL, P<0.001]. Conclusion The use of short-term liquid supplementation, traditional Chinese medicine anti-inflammatory treatment, and early dietary stimulation can effectively combat early severe inflammatory damage in SAP, protect the life of model rats, and enable them to survive and experience digestive dysfunction, thus establishing an experimental animal model of digestive dysfunction in the late stage of SAP.