Vol. 9, Issue 12 (2020)

Author(s):
Chenlep Yakha Konyak

Abstract:
This study addresses the need for a more sophisticated analysis of pharmacy performance data to uncover hidden patterns, correlations, and operational inefficiencies over traditional pharmacy evaluation methods. Applying machine learning techniques, specifically K-Means clustering and Principal Component Analysis (PCA), to a dataset of 100,010 examples, the research examines key performance metrics such as sales revenue, prescription volume, medication adherence rate, and customer satisfaction scores. The experimental approach segments pharmacies into distinct groups using K-Means clustering and employs PCA for two-dimensional visualization. The findings reveal four unique clusters, each with different operational strengths and weaknesses, highlighting the importance of technology adoption, online services, and community engagement in driving pharmacy performance. The study concludes that machine learning techniques provide valuable insights and strategic guidance for healthcare stakeholders, suggesting that investments in technology and community-focused services can optimize pharmacy operations and enhance patient outcomes.