Advanced pharmaceuticals hold great promise for improving human health. However, complex and costly manufacturing processes and growing regulatory scrutiny combine to create a challenging business environment. Companies that don’t keep costs in line or fail compliance tests can find themselves out of business quickly.
Bigfinite, a startup based in San Francisco, has demonstrated the power of data to solve these challenges and help clients deliver life-changing solutions while improving margins, compliance, and manufacturing efficiency. “The industry has vast amounts of data that can solve efficiency challenges,” says Pep Gubau, chief executive officer at Bigfinite. “Unfortunately, it’s locked up in silos, such as legacy apps that lack interoperability, which can make it challenging to use effectively.” In 2016, Gartner Research reported that 70 percent of manufacturing data from life-sciences companies goes unused. Bigfinite is helping to unlock that value and help the industry move forward.
In 2014, the company began work on Bigengine, a software-as-a-service (SaaS) analytics platform that analyzes disparate data sources to uncover insights that improve productivity and compliance at pharmaceutical manufacturing sites.
The company chose to build Bigengine on Amazon Web Services (AWS) because it enables rapid innovation while meeting the compliance needs of pharmaceutical manufacturers. The company was initially planning to build Bigengine on a different cloud platform. However, after meeting and collaborating with the AWS healthcare and life-sciences team, the company chose AWS because of its demonstrated ability to meet a wide range of compliance needs—especially the set of best practices known as GxP. GxP encompasses a range of compliance-related activities such as Good Laboratory Practices (GLP), Good Clinical Practices (GCP), Good Manufacturing Practices (GMP), and others. AWS has published a white paper on GxP compliance to help customers address these issues.
“Its pioneering focus on compliance makes AWS the most strategic technical partner a life-sciences company could work with,” says Gubau. “By taking advantage of AWS, we can analyze huge amounts of data using the most secure platform and the most advanced technology in a 24/7 environment."
Using AWS, Bigengine can store and analyze vast amounts of pharmaceutical-manufacturing data using advanced analytical techniques such as machine learning, artificial intelligence, and neural networks. The resulting real-time predictions and models solve a wide range of challenges.
For example, Bigengine helps manufacturers predict when equipment needs cleaning or maintenance, or is about to fail, eliminating costly manufacturing disruptions and prolonging service life. It enables them to optimize processes to reduce costs and energy usage, and to detect and address anomalies to increase the quality of the final products. Bigfinite customers can edit and configure these solutions or create new capabilities using the built-in solutions designer.
Multiple AWS services power the Bigengine platform. The platform ingests data from a diverse range of sources, from laboratory information-management systems to manufacturing equipment connected to the cloud through AWS IoT. The data is captured and stored in native formats in a controlled data lake in Amazon Simple Storage Service (Amazon S3). Once captured, data can be interpreted using a range of techniques coordinated by AWS Lambda serverless computing and AWS Step Functions for managing distributed application components. The solution uses Amazon Athena for analyzing data through serverless queries, Amazon Machine Learning for setting up an API-based approach to creating models and predictions, and Amazon Elastic MapReduce (Amazon EMR), for supporting big-data workloads on Apache Spark. Bigfinite utilizes Spark clusters to run its proprietary algorithms that are developed in Python and Scala.
Bigfinite has repeatedly demonstrated the ability of Bigengine to reduce costs, improve compliance, and keep innovation moving forward. For example, in collaboration with a multinational pharmaceutical company based in Europe, Bigfinite deployed a system that puts 50 reactors, which mix compounds in parallel, in a complex drug-development “supersystem.” The bioreactors release volatile organic compounds (VOCs) during the reactions, which are heavily regulated and need to be treated before being released into the atmosphere.
Several times a year, the pharmaceutical companies exceeded the VOC-emission limits, which elicited large fines from regulatory agencies and were damaging to the environment. If this situation continued without a reduction in emissions, the production site could have been shut down. Moreover, the facility’s furnaces needed to be cooled down quickly after treating the VOCs, requiring the use of enormous amounts of electricity for cooling purposes.
Over the initial two-month use of Bigengine, the platform was trained to convert raw production data into recommendations for reorganizing the company’s manufacturing processes at the site. In the month that followed, the company never exceeded the VOC-emission limits and lowered the cooling system’s electrical consumption to 17 percent of what it had been.
“We did all this with the data the customer already had,” says Gubau. “The company saved a massive amount of energy in addition to avoiding the regulatory fines.” By helping organizations reduce risk, improve efficiency, and maximize compliance, Bigfinite is helping the pharmaceutical industry put more resources into innovation—creating big potential for improving health.
Learn how to accelerate scientific discovery and enable operational efficiencies using the AWS Cloud.