Methods Of Single Disease Cost Variance Analysis Based On Software Algorithm And Its Application

If a hospital want to stand in an invincible position in the fierce competition and achieve sustainable development,It must actively participate in market competition, not only to improve the quality of medical services, and to reduce the price of medical services. So hospital should constantly reduce costs to gain competitive advantage. Cost variance analysis is a key technology to hospital cost control process. For the limitations of cost variance analysis model and the particularity of the hospital system, this paper proposed two models: a single disease cost variance analysis model based on Fuzzy Bayes rule and method of single disease cost variance analysis based on soft computing. The main work is as follows:1)For the present state that cost control model cannot predict the extent of variance, we propose A single disease cost variance statistical analysis model based on Fuzzy Bayes rule. By analyzing the level of cost variance to preventive control the costs. Because cost variance of single disease is very large, and abnormal and normal are some vague concept, So we using Fuzzy Bayes rules to calculate all state posterior probability, then using dynamic programming method to minimize the expected cost to determine how to make the decision. Finally, we apply this model into simple appendicitis surgery cost control and verify the the model is feasibility.2)We summarized a general form of cost variance analysis model on the basis of previous studies. Considering the faults of the general form, we developped a method of cost variance analysis based on soft computing. First, BP neural network is carried out to predict the probability of system status for the limitation that it‘s very hard to estimate every parameter accurately in the general model; secondly, differential evolution algorithm method is proposed to optimize the model to solve curse of dimensionality problem caused by dynamic programming method; finally, the model was applied to study the single disease cost variance analysis, comparing the results between general algorithm and soft computing integration algorithms shows that of soft computing integration algorithms can come to get approximate optimal solution of the general algorithm, verify the validity of the model.