Design of Experiments and Statistical Process Control for Enhancing Product Quality and Customer Satisfaction in E-Commerce Apparel

Customer reviews on e-commerce platforms offer a rich and real-time source of feedback for assessing and improving product quality. This study integrates Design of Experiments (DOE) and Statistical Process Control (SPC) to systematically analyze factors influencing customer satisfaction for high-waisted culotte jeans sold via an online marketplace. Review data—including ratings, comments, and timestamps—was automatically collected through e-commerce’s API using Python scripts, enabling efficient and large-scale acquisition. A 2³ factorial DOE evaluated three critical factors: delivery speed, fabric quality, and size accuracy, each tested at two levels. Results showed that size accuracy (main effect = 0.14) had the strongest influence on customer ratings, followed by fabric quality (0.13), while delivery speed (0.06) also contributed significantly. Interaction analysis revealed both positive synergies, such as between product size and material, and negative interactions, such as between delivery and size, indicating trade-offs in customer expectations. SPC analysis using X-bar, R, and S charts confirmed overall process stability, with all metrics within control limits. However, upward trends in mean ratings and fluctuations in range values indicated emerging variations requiring continuous monitoring. The decreasing standard deviation suggested improved consistency, though recurring patterns implied potential process changes. This research demonstrates the practical value of combining DOE and SPC with real-time, user-generated data to identify critical quality drivers and maintain stability in dynamic markets. The proposed framework offers apparel manufacturers and e-commerce retailers a replicable, data-driven approach for targeted product enhancement, operational monitoring, and sustained customer satisfaction.