Dataset of retail industry segmentation in Indonesia

To guarantee customer satisfaction, retailers need to manage their product inventory considering factors such as demand volume, frequency, consistency, and variability. By classifying products according to demand trends and inventory turnover, retailers can improve efficiency and customer contentment. Overall, the clustering approach ensures that high-demand items are readily available while reducing the surplus of less sought-after products.Post-COVID-19 pandemic, the delivery and logistics sector have much attention from researchers. Small-and-medium enterprises (SMEs) are prioritizing cost reduction in shipping by deploying streamlined strategies to lower overall logistics expenditures. One such tactic is clustering, where regions or delivery requests are grouped for transportation between distribution centers. Clustering plays a vital role in location-driven decision-making. Clustering product delivery based on customers' location will enhance operational efficiency, customer contentment, and financial returns.This case study explores customer segmentation using three-dimensional time-series data: total spending, frequency of purchases, and average basket size.

为保障客户满意度，零售商需结合需求量、需求频率、需求稳定性与需求波动性等多维度因素管理商品库存。通过依据需求趋势与库存周转率对商品进行分类，零售商可提升运营效率并优化客户满意度。总体而言，聚类方法可确保高需求商品随时可供备货，同时减少滞销商品的库存积压。
新冠疫情后，配送与物流领域受到研究者的广泛关注。中小企业（Small-and-medium enterprises, SMEs）正将压缩航运成本作为优先事项，通过部署精简优化策略降低整体物流开支。聚类便是此类策略之一，即对配送区域或配送需求进行分组，以优化配送中心间的运输调度。聚类在基于区位的决策制定中发挥着关键作用。基于客户地理位置开展商品配送聚类，可提升运营效率、客户满意度与财务收益。
本案例研究借助三维时间序列数据——涵盖总消费金额、购买频次与平均购物篮规模——开展客户细分任务。