Please visit the COVID-19 response page for resources and advice on managing through the crisis today and beyond.
A leading pharmaceuticals company needed to accurately track the market share of medications and consumer health products it manufactures and sells. It wanted to identify the future business prospects for each product, its contribution to the total market and key factors influencing the growth of that market.
The existing Microsoft Excel-based classifications model required significant manual intervention that led to rework and human error. So, the manufacturer looked for a scalable, artificial intelligence-based solution to categorize market share analytics automatically.
We designed and developed an AI-based intelligent system using Azure Machine Learning to auto-classify medications and consumer health products into a variety of market-related categories. Prior to classification, the application processes high-volume data from multiple sources using Azure Data bricks and Azure Data Factory.
Our solution analyzes the output in Power BI (business intelligence) and generates different metrics from that data, enabling users to create customized, self-service reports. This allows the business to understand and respond to big picture market trends and how each product performs against its competition.
We leveraged Azure MLOps to build a data pipeline for continuous model training. As testing and training progress, new data is incorporated and models become more accurate. A scalable infrastructure monitors and alerts analysts about any operations issues.
This engagement increased operational efficiency, allowing the manufacturer’s personnel to focus on more value-added and less repetitive tasks. It also greatly improved the customer experience by making it simple to categorize and sort sales of specific medications and consumer health products.
reduction in processing time using scalable Azure AI/ML services
reduction in manual efforts using machine learning approach
onboarding effort for new market sources achieved through automated ingestion framework
SPREAD THE WORD