It must be frustrating for executives in pharmaceutical manufacturing to see other industries making such good use of AI to solve their production and value chain challenges.
Alongside widespread understanding of artificial intelligence (AI) and the manifold ways it can transform the industry, research by AspenTech has found almost all executives say their companies struggle to handle their data in ways that open the door to smarter, better ways of working.
The research, among 300 decision-makers in the industry in the UK, US, Germany, France, Spain and Sweden reveals that 50% know AI can help bring new drugs to market more rapidly and securely. Yet almost everyone (96%) says they face challenges with using such leading-edge technology to derive value from their data. And, perhaps even more alarmingly, 43% believe that if companies in their industry fail to learn the lessons of AI and machine learning (ML) adoption from other sectors, they will be in severe financial trouble within two years.
The problems cited by respondents range from the absence of an overarching strategy for AI (49%) to problems with the complexity of data that companies feel ill-equipped to surmount. Nearly a third of respondents (31%) say their company lacks the kind of consistent data structures that make implementation easier, or they have high levels of unstructured data that does not fit into neat rows and tables.
Three-in-ten executives (30%) say their companies struggle with that most familiar of problems – data that is held in separate, siloed systems. Others have difficulties with the skillset required to wrangle or manage the data, or they labor with a risk-averse company culture that does not foster innovation. Even the more advanced, data-driven companies in the research have a problem with risk-averse culture (43%).
What does this tell us? Most obviously it shows pharma companies need to act now to tackle their data challenges so they can implement AI. The competitive, supply chain and regulatory pressures are growing, but most can be resolved by putting AI into practice, as so many executives evidently see. This needs to be built on investment in the ability to break down the barriers between systems and types of data within production processes and supply chains. It also requires organizations to reimagine their digital culture and think more holistically about what data will add across all aspects of drug manufacture.
This has never been truer than at present. More than half of all respondents in the research (54%) believe COVID-19 will continue to disrupt vaccine manufacturing and exactly half think the pandemic will force a rethink of where companies locate their manufacturing plants. These are entirely reasonable beliefs, based on the evidence of the last 15 months and the continuing emergence of new strains of the coronavirus. Yet 42% of senior executives admit their company maintains factory space it might never need. This is not exactly a ringing endorsement of current planning capabilities, indicating that in a time of considerable upheaval, many senior teams operate in a fog of data with lack of insight for future asset-optimization.
It is time for companies that are slow in their approach to AI to grasp that their rivals who are further ahead have the edge in many areas, including asset-optimization and planning. Not only do they benefit from predictive maintenance to reduce downtime, they can also use artificial intelligence and machine learning to reduce unused factory and asset space and be far better prepared for shifts in demand. Many pharma companies still need to develop a culture that embraces artificial intelligence and machine learning to unleash their potential. Solutions are available now and companies are adopting them to get ahead in so many areas, especially in planning the optimal use of their assets in relation to constantly changing parameters, market conditions and global events.
To learn more, register now for our upcoming webinar: How are Pharmaceutical and Biotech Companies Using Data and AI with Confidence?