全渠道消费者行为协同决策研究

随着各大零售商全渠道营销战略布局,全渠道消费者数量呈爆炸式增长,对于全渠道消费者的消费行为研究成为热点。然而连锁零售供应链全渠道消费者消费数据呈现海量、高维的特征。针对这一特征,提出采用协同进化算法对连锁零售供应链全渠道消费者行为进行高维关联分析。利用粒子群优化算法和自适应遗传算法各自的优势,两个种群同时遍历,并在两种群间引入信息交互机制,使两种群协同进化。实证研究证明协同进化算法应用于连锁零售供应链全渠道消费大数据关联规则挖掘中,不仅算法的运算速度高,避免了遗传算法单独应用时容易陷入局部最优的缺陷,而且还提高了连锁零售供应链全渠道消费者行为关联规则的大数据挖掘质量,为全渠道消费者购买行为研究提供了新的方法。

Collaborative decision making of omni-channel consumer behavior

With the layout of the omni-channel marketing strategy, the number of consumers in the omni-channel witnesses an explosive growth, and the consumer behavior of the omni-channel becomes a research hotspot. However, the consumption data of the omni-channel consumers of the chain retail supply chain is massive and high dimensional. Given the abovementioned features, we propose a co-evolution algorithm to analyze omni-channel consumer behavior in the chain retail supply chain. Taking the advantages of the particle swarm optimization algorithm and adaptive genetic algorithm, the two populations are traversed simultaneously, and the information interaction mechanism is introduced between the two populations, which makes the two populations collaboratively evolve. Empirical research proves that when the collaborative evolution algorithm is applied to association rule mining of omni-channel consumer’s consumption data in the chain retail supply chain, the speed of the algorithm is faster, and it can also avoid the local optimum of the genetic algorithm when it is applied alone, and improves the quality of omni-channel consumer behavior association rule mining in the chain retail supply chain . It provides a new method for the study of the omni-channel consumer purchasing behavior.