基于遗传BP神经网络组合模型的中药销售预测研究

Prediction of Traditional Chinese Medicine Sales Based on Combined Model of Genetic BP Neural Network

对某医院2010-2012年中药的销售及变化情况进行了对比分析。为了更好地保证医院中药的库存数量,提出了一种基于遗传算法优化BP神经网络组合模型,用改进的遗传BP神经网络进行中药销售量预测,并将预测结果和单纯使用BP网络的预测结果进行比较,实验证明遗传BP神经网络模型具有更高的预测准确度,为医院中药销售及科学管理中药库存量提供科学依据。

Sales and its changes of traditional Chinese medicine in 2010-2012in a hospital were compared and analyzed.To better ensure inventory quantity of traditional Chinese medicine in the hospital,we proposed a optimized BP neural network model based on genetic algorithm.Predicted sales of traditional Chinese medicine using optimized BP neural network was compared with the predicted result using a simple BP network.Experiments show that genetic BP neural network model has higher prediction accuracy,providing scientifical reference for sales of traditional Chinese medicine and scientifically managing inventory quantity of traditional Chinese medicine in hospitals.