As pharmaceutical and medical-device companies undergo digital transformation, they are turning to artificial intelligence to facilitate one of their top priorities: quality management.
“The life-sciences industry is a growing, innovative space, what I call ‘health care being reimagined,’ ” says Steve McCarthy, vice president of digital innovation at Sparta Systems, a Hamilton, New Jersey-based provider of cloud and quality-management software. “There are new types of life-saving, life-changing products being innovated within this space, including combination products, personalized medicine and gene therapies. In addition, we’re in the midst of Industry 4.0, and that’s having an impact on what’s called the factory of the future.”
With this disruption, supply chains “are increasingly being externalized and outsourced to contract manufacturers and innovative raw-material and component suppliers,” McCarthy says. “The point of manufacturer is shifting more closely to the patient. If you think about gene therapy or a personalized medicine, it’s essentially a treatment made in a batch of one for a patient’s unique genome.”
The same is true of products like 3-D-printed medical devices, he says: “That’s a reality already. More and more, traditional medical-device manufacturers are printing devices, and as time goes on — and I think it will happen rapidly — they will be printed in the surgical suite for your unique anatomy.”
Quality Emerging Importance
The disruption and innovation point to quality management’s increasing importance to companies. Studies have shown there is more investment in quality than ever before, McCarthy says: “It is no longer separated as a topic from things like operational efficiency or cost.”
Quality management spans all aspects of the business, including as a time-to-market factor. “If a company can get a drug to market one month sooner than a competitor, this has an immeasurable impact on its competitive advantage,” McCarthy says. Fixing a quality issue before a product hits the market, rather than after, enormously impacts the revenue stream and productivity — as well as the patients, he adds.
“Quality is integral when it comes to supply chain efficiency and effectiveness,” he says. “And there is a significant and increasingly growing understanding in the C-suite of the importance of quality in all aspects of the business.”
AI can help with quality management in numerous ways, including in data handling and predictive modeling:
Data. Pharmaceutical companies manage volumes of complaint data — but the real issue isn’t quantitative, it’s qualitative, McCarthy says: “A business’s complaint-handling unit could be an army of people who must wade through the (data) and try to decipher if there a real quality or product-safety problem or patient-health issue exists.” The task can be overwhelming. “It can be difficult to detect the signals pointing to a real potential for product-quality or patient harm,” McCarthy says.
By augmenting human intelligence, AI can enhance efficiency and timeliness. Because the technology can learn from past as well as succeeding data, it is not only able to suggest when a potential product quality issue exists but what the severity and likelihood of that problem might be. McCarthy says, “It is able to better decipher (1) the intended sentiment of the complaint, whether positive, negative or neutral; and (2) whether it is high or low risk, and even start to point toward root causes, (thus) making system more efficient and effective.”
Predictive modeling. Deviations can occur in the manufacturing process, during the inspection of incoming raw materials and goods, during testing, on the shop floor or a myriad other situations. “It’s in the nature of manufacturing, but particularly in the manufacture of complex products like drugs and devices,” McCarthy explains. “As you start to point AI at this huge amount of quality data that’s happening during the manufacturing process — maybe even back into the design process of a product — you begin to open up the real opportunities for proactive and predictive quality.”
Thus, AI can help detect a quality problem earlier in the process. “A product quality problem that’s made it all the way to the market is much more expensive and problematic than capturing that product quality problem at the early stages of the value chain. The sooner you can capture the problem, the greater the incremental impact,” says McCarthy.
Supplier/contractor management. With the outsourcing of much of the life sciences supply chain, it’s increasingly important for companies to maintain control of their suppliers and contractors, he says. In 2018, a majority of recalls were caused by a lack of supplier or contract manufacturer control. AI can help health-care organizations manage their supply base, McCarthy says.
As the supply chains of pharmaceutical and medical-device industries become increasing disrupted, companies will look to AI to help with quality management. “As you start to look out into the future, the art of the possible becomes enormous,” McCarthy says.