Exploring rfm model and fuzzy c-means clustering in dynamic

Abstract

Dynamic customer segmentation plays a critical role in modern business analytics by enabling firms to adapt to rapidly changing consumer behaviour. Unlike static models, it leverages real-time and behavioral data to update customer groupings, supporting personalization, predictive analytics, and efficient resource allocation. Advancements in big data, machine learning, and customer data platforms have made dynamic segmentation increasingly accessible and effective. This thesis examines its strategic value in enhancing customer engagement and decision-making agility. The findings demonstrate that businesses utilizing dynamic customer segmentation are better positioned to anticipate customer needs, optimize marketing strategies, and sustain competitive advantage in rapidly evolving, data-intensive economic environments characterized by continuous change such as short-term financial shocks to longterm economic cycles. This dynamic environment also includes seasonal fluctuations, technological advancements, and wider socioeconomic developments. Consequently, a critical challenge for organizations across sectors, including government agencies, small businesses, and large corporations, is the ability to rapidly detect and respond to these economic and technological changes. Effective adaptation to emerging trends and seasonal variations is essential for maintaining competitiveness and operational efficiency. In response to these demands, dynamic data mining techniques have gained prominence as vital tools for analyzing and navigating complex, evolving environments. These methodologies have found widespread applications across diverse fields such as science, marketing, engineering, and management. Within the retail industry, data mining approaches are increasingly employed to uncover customer purchasing behaviors, thereby enhancing service quality, customer loyalty, and retention. One widely used method for customer segmentation is the Recency, Frequency, and Monetary (RFM) model. When integrated with machine learning algorithms, the RFM model provides effective segmentation and valuable insights into purchasing behavior. However, the traditional RFM model is limited by its reliance on only three variables and does not incorporate demographic information or detailed purchase