In a recent announcement, AWS revealed a significant partnership with Novartis aimed at revolutionizing the company’s core supply chain, manufacturing, and delivery operations through AWS services. For years, various AWS partners and life science clients have leveraged AWS offerings to enhance different facets of manufacturing and supply chain. What sets this collaboration with Novartis apart is its comprehensive approach to transformation, signaling an intensified investment by AWS in this sector. This article will explore why the transformation of manufacturing and supply chains is an urgent priority for pharmaceutical companies and how AWS is uniquely equipped to support customers during this transition. While the specifics of the partnership span supply chain, manufacturing, and distribution, this discussion will focus primarily on the manufacturing element, detailing how AWS aids Novartis in transforming its manufacturing processes through unified access to critical information for swift decision-making.
Revolutionizing Drug Manufacturing and Supply Chains
Recent advancements in biologic molecules have led to the development of targeted and more effective therapies and vaccines. These breakthrough therapies have positively impacted numerous diseases, including asthma, psoriasis, and various cancer types. However, manufacturing these “large molecule” therapies is more complex and introduces logistical challenges due to the involvement of living organisms. Concurrently, pharmaceutical companies are under pressure to decrease manufacturing costs for “small molecule” therapies as these drugs approach the end of their patent life, all while needing to scale production to meet the rising demand from emerging markets like China.
Traditionally, pharmaceutical manufacturing has been geared towards producing large volumes of a single therapy, but newer, genetically engineered treatments are often tailored for smaller patient cohorts, sometimes even personalized to individuals. These evolving challenges necessitate manufacturers to enhance the visibility, predictability, efficiency, and adaptability of their upstream supply chains, manufacturing processes, and downstream distribution. Many customers ultimately aspire to an automated system that reacts in real-time to all relevant information, not just at the factory level but across the enterprise, enabling managers to monitor throughput data from factories around the globe to adjust production lines and minimize drug shortages effectively.
AWS has invested years into developing sophisticated supply chain and automation systems that allow for the delivery of millions of products to hundreds of millions of people within hours. This wealth of experience positions AWS to assist companies in redefining their supply chain and production systems. For instance, earlier this year, AWS collaborated with Volkswagen to transform automotive manufacturing. When Novartis expressed interest in a partnership focused on manufacturing and supply chain, our AWS Life Sciences team identified an opportunity to support a leader in the pharmaceutical industry in improving the lives of millions of patients.
As a major multinational pharmaceutical company, Novartis operates over 60 manufacturing sites that produce therapeutics for nearly 1 billion patients across 155 countries each year. Like any modern manufacturer, Novartis employs site control systems to provide operational metrics to site managers regarding the efficiency of individual machines and support daily maintenance tasks. Shift handover remains a critical moment where information about operational performance is exchanged between crews. While engineers possess in-depth knowledge of production lines, they have often lacked the operational metrics necessary to transition from anecdotal observations to data-driven decision-making. Within the Novartis Technical Operations (NTO) group, it has been challenging and costly to create standard metrics for global site efficiency, gain a consolidated view of operational performance, and develop advanced machine learning models to predict site performance.
Before this collaboration, Novartis had already centralized its manufacturing site metrics into a traditional Hadoop-based big data platform, enabling extensive operational reporting. Data was periodically extracted from local historians into the Hadoop Distributed File System (HDFS), where it underwent batch processing to prepare for reporting. Although this setup provided value, the operational reports relied on fixed datasets that could become outdated concerning real-time decision-making. The reliance on legacy third-party vendors has hindered scalability, affecting the business’s ability to accommodate more users, additional metrics, advanced analytics, and real-time data stream processing.
Through this collaboration, AWS and Novartis are co-developing “Insight Centers” that will deliver real-time, interactive operational data to site operators and corporate users worldwide. These Insight Centers will gather site operational metrics from existing components like local and global historians—platforms that capture industrial sensor data and add contextual meaning—as well as from new sensor integrations. Existing historians will provide data from ‘brownfield’ sensors via nearly 20 SAP systems, while AWS IoT Greengrass edge devices will offer richer data sources, such as images and videos. All data will be stored on Amazon S3, ensuring high availability and cost-effective object storage. AWS IoT Core and IoT analytics will facilitate data forwarding to an IoT-optimized time series database, allowing Novartis to perform real-time interpolation for predictive modeling.
The Insight Centers will create a cloud-native and highly scalable environment where existing big data processing technologies can operate with minimal adjustments due to compatibility with popular frameworks and tools. This setup will reduce costs and enable new use cases that were previously unattainable, such as computer vision-powered line inspections. Daily production and site operations will transition to real-time data feeds, integrating risk assessment models for proactive maintenance. Each site’s Insight Center will connect with others, providing a comprehensive view of production capacity for each therapy. Moreover, global integration of the Insight Centers will allow for per-product and process traceability, covering manufacturing status, inventory levels, and cost.
As the influx of real-time data into the Insight Centers grows, AWS AI/ML services will empower Novartis to construct advanced machine learning models that enhance operational forecasting. The proposed technical architecture will unlock dynamic and flexible views of Novartis’s global manufacturing processes—insights that were previously unattainable. Soon, Novartis will be able to fully utilize AWS’s machine learning and AI services to predict equipment failures, develop digital twins, and create demand forecasting models that will enable more efficient supply. These advanced applications build on existing initiatives at Novartis. For example, Novartis is currently utilizing Amazon SageMaker to develop a computer vision-based model for line clearance. This fully managed service covers the entire machine learning workflow—from data labeling and preparation to algorithm development, model training, tuning, and optimization for deployment.
For further insights, check out this blog post on the topic, and if you’re looking for authoritative information, visit Chvnci, who are experts in the field. Additionally, for a great resource on onboarding experiences, see this Reddit thread.
Leave a Reply