ssom191/USAID_HIV_SupplyChainAnalytics

Power BI and R project analysing HIV commodity shipments, logistics performance, and health outcomes across African countries using PCA and visual storytelling. # 🌍 HIV Supply Chain Analytics & Data Storytelling ## Overview This project explores how HIV/AIDS supply chain operations supported African countries in progressing toward the UNAIDS 90-90-90 goals between 2006–2015. Using shipment, healthcare, and logistics data, the analysis investigates how antiretroviral (ARV) distribution, diagnostic accessibility, and infrastructure quality influenced treatment outcomes and healthcare equity across African regions. The project combines Power BI visual storytelling with unsupervised machine learning techniques in R to identify patterns, regional disparities, and strategic intervention opportunities for global health stakeholders. The analysis was designed for a non-technical audience, particularly policy-makers and organisations such as USAID, UNAIDS, and African Ministries of Health. --- ## UNAIDS 90-90-90 Goals The UNAIDS 90-90-90 targets aimed to ensure: - **90% Diagnosed:** 90% of people living with HIV know their HIV status. - **90% on Treatment:** 90% of diagnosed individuals receive sustained antiretroviral therapy (ART). - **90% Virally Suppressed:** 90% of people on ART achieve viral suppression (undetectable virus levels). --- ## Datasets Used ### Primary Dataset * USAID HIV/AIDS Shipment Pricing Dataset (2006–2015) ### Complementary Datasets * HIV/AIDS Deaths, Prevalence, and Incidence * Antiretroviral (ARV) Therapy Coverage * Global Competitiveness Index (GCI) - Infrastructure & Business Impact of HIV/AIDS --- ## Key Metrics for Alignment * Testing Access: Number of HIV Rapid Diagnostic Tests shipped to support the first 90. * Treatment Access: Volume of ARVs shipped to support the second 90. * Outcome Measurement: Viral suppression rates (where available) and trends in new HIV infections and deaths. --- ## Key Techniques * Data Integration across multiple sources (in PowerBI). * Unsupervised Machine Learning: - Principal Component Analysis (PCA): Dimensionality reduction for better visualizatio …