论文部分内容阅读
研究电子商务零售中动态捆绑决策的优化问题.延迟购买行为会大大降低易逝品零售商的收益水平,动态捆绑策略是有效解决延迟购买效应的一种营销手段,计算复杂性和决策的实时性需求是实施动态捆绑策略的难点.对此,首先基于延迟购买效用分析提出了易逝品市场需求状态模型和动态捆绑的决策过程模型,并给出了决策过程模型中随机参数的估计方法.在此基础上,根据问题特点对传统Q学习算法加以改进,使之适应动态捆绑策略的优化问题.模拟实验结果表明:1)延迟购买程度越大,动态捆绑策略对收益的贡献也越显著;2)用所提出的改进型Q学习算法求解动态捆绑策略优化问题具有较高的效率和效用.
To study the optimization problem of dynamic bundling decision-making in e-commerce retailing, delay purchasing behavior will greatly reduce the return of perishable retailers, and dynamic bundling strategy is a marketing tool that effectively solves the delay buying effect, and computes the complexity and decision-making real-time Demand is a difficult point to implement dynamic bundling strategy.This paper firstly puts forward market demand state model and dynamic bundling decision-making process model of perishable goods based on delay purchase utility analysis, and gives a method of estimating stochastic parameters in decision process model. On this basis, the traditional Q learning algorithm is improved according to the characteristics of the problem to adapt it to the optimization problem of dynamic binding strategy.The simulation results show that: 1) the greater the delay purchase, the more significant the dynamic bundling strategy contributes to returns; ) It is more efficient and effective to solve the dynamic binding strategy optimization problem with the proposed improved Q learning algorithm.