Personalization in Dynamic Assortment Planning: An Analysis Based on Multi-Agent Simulation Method

Assortment planning is one of the most critical issues in retail category management. This research aims to introduce personal data into traditional assortment planning problem. Firstly, we analyze decision process of customer and retailer, and formulate consumer choice model and retailer assortment planning model. Based on that, a multi-agent simulation system is constructed, to evaluate effect of personalization and explore assortment strategies in different scenarios. Result of simulation experiment shows that, personalized assortment planning model and algorithm proposed in this research can effectively improve retail performance.