新型ARIMA-BP组合模型在医药企业销售管理中的应用
2016-05-09吴磊徐怀伏
吴磊+徐怀伏
摘 要 随着医药企业间的市场竞争加剧,企业销售管理的重要性逐渐凸显。医药销售预测是个复杂的非线性系统,为提高企业销售预测的准确性,本文选取医药上市企业处方药七叶皂苷钠历史销售数据,分别建立ARIMA线性模型和BP神经网络非线性模型并加以验证。证明了在销售预测上采用ARIMA-BP组合模型可以有效降低误差,为医药企业的销售管理和企业决策带来新的思路。
关键词 销售管理 ARIMA模型 BP神经网络
中图分类号:F406.69 文献标识码:A 文章编号:1006-1533(2016)07-0068-05
Application of ARIMA -BP model in the sales management of the listed pharmaceutical companies
WU Lei, XU Huaifu
(School of International Pharmaceutical Business, China Pharmaceutical University, Nanjing 211198, China)
ABSTRACT The importance of sales management in pharmaceutical enterprises has become increasingly prominent with the intensification of market competition among them. Since the forecast of pharmaceutical sales is a complex nonlinear system, the data of historical sales for sodium aescinate, a prescription drug from the listed pharmaceutical companies were selected, and an ARIMA linear model and a BP network nonlinear model were established and verified based on the theory of linear and nonlinear prediction in order to improve the accuracy of sales forecast. It is proved that adoption of a combination of two models can effectively reduce errors and bring in some new ideas for the sales management and policy decision of pharmaceutical enterprises.
KEY WORDS sales management; ARIMA model; BP neural network
随着经济全球化以及市场竞争的加剧,销售管理工作逐渐成为企业管理工作的核心。目前,我国医药企业销售管理工作中存在销售预测脱离实际,预测主观性较强,缺乏科学的方法等问题。准确的药品销售预测不但可以指导生产、销售工作来防范风险,还是科学的销售绩效指标的重要前提条件。随着现代理论以及数据挖掘工具的发展,销售预测和实际数据吻合将成为现实。由于药品销售数据实际是非线性、时变的时间序列数据,本文通过对医药上市企业历史销售数据的分析,建立ARIMA模型,预测结果作为线性预测能力,同时建立BP神经网络模型,预测结果作为非线性拟合能力,将两者结果通过权值来组合模型进行预测,提高了药品销售预测的准确率,为我国医药企业优化药品销售决策方案提供新思路
1 药品销售预测理论概述
1.1 销售预测
药品销售预测是根据药品的历史销售数据和企业发展战略,充分考虑市场需求、市场环境、医药政策等因素,对未来市场进行定量预测,并有合理的预见结论。……
