A Fuzzy Inference-Based Dynamic Discount Algorithm Proposal for Customer Loyalty Programs

Abstract

This study proposes a fuzzy inference-based loyalty program aimed at enhancing customer loyalty by considering shopping frequency, average spending, and subscription behavior. Developed to overcome the limitations of traditional loyalty programs, the proposed system analyzes customer behavioral data to offer personalized discounts. The model uses shopping frequency, average spending, and recency as input parameters, which are modeled using fuzzy logic to calculate a customer loyalty score. These scores are then utilized to determine tailored discount rates based on customer profiles. The performance of the model was validated through a case study, demonstrating its effectiveness in increasing customer satisfaction and encouraging repeat purchases. This approach provides businesses with a data-driven and flexible framework for designing loyalty programs.