{"slug":"novo-nordisk-uses-machine-learning-and-computer-vision-to-automate-pharmaceutical-manufacturing-quality-checks","url":"https://findausecase.com/use-cases/novo-nordisk-uses-machine-learning-and-computer-vision-to-automate-pharmaceutical-manufacturing-quality-checks","title":"Novo Nordisk uses machine learning and computer vision to automate pharmaceutical manufacturing quality checks","description":"Novo Nordisk, a Danish pharmaceutical company that supplies nearly 50 percent of the world's insulin, built an automated ML pipeline on AWS using Amazon SageMaker Pipelines to train, deploy, and monitor computer vision models on edge devices. The system automates previously manual tasks including counting drug cartridges in boxes and detecting bacterial anomalies on agar plates, using a robotic arm, camera rig, and edge inference, with results monitored via an Amazon QuickSight dashboard. The solution deploys ML models at scale to different edge devices, automates quality-assurance tasks, improves time to market, and was successfully repurposed from the cartridge-counting use case to the agar-plate anomaly detection use case with minimal changes.","company":"Novo Nordisk","industry":"Life Sciences","country":"Denmark","aiCapabilities":["Computer Vision","Machine Learning"],"technology":["Amazon SageMaker","Amazon S3","AWS IoT Greengrass","Amazon QuickSight"],"deployment":"Unknown","problemStatement":"Novo Nordisk had explored ML to automate time-consuming manual manufacturing tasks, but the parts of its ML-development process ran locally on individual machines and were not interconnected, making it difficult to deploy models at scale and maintain them in production.","solutionApproach":"On AWS, Novo Nordisk used Amazon SageMaker Pipelines to build an automated ML pipeline covering data processing, model training and tuning, evaluation, and deployment to edge devices via Amazon SageMaker Edge and Edge Manager, using AWS IoT Greengrass as the edge runtime and monitoring models in production with Amazon QuickSight and Amazon Timestream.","businessValue":"The solution deploys ML models at scale to different edge devices, monitors them in production, and automates quality-assurance tasks such as cartridge counting and agar-plate anomaly detection, improving time to market; the cartridge-counting pipeline was successfully repurposed for the agar-plate use case with minimal changes, proving the approach could scale to other quality-assurance use cases.","evidence":{"band":"high"},"sourceUrl":"https://aws.amazon.com/solutions/case-studies/novo-nordisk-case-study/","dates":{"publishedAt":"2026-09-15T09:05:20.823Z","publishedAtSource":"pipeline","updatedAt":"2026-09-15T09:05:20.823Z"},"license":"Open for reading and citing with a link to https://findausecase.com/use-cases/novo-nordisk-uses-machine-learning-and-computer-vision-to-automate-pharmaceutical-manufacturing-quality-checks. Bulk republication requires permission."}