By Chris Farrow, Vice President, Materials Science Solutions at Enthought
As demand increases and competition becomes tighter for functional materials, such as electrolytes for batteries, consumables for semiconductor manufacturing, and functional plastics, materials and chemical companies are competing to continuously innovate and differentiate themselves in new and existing markets. Despite the urgency, 60-70% of scientists’ time is still spent on non-research activities like administrative tasks and general data work.
The key to accelerating discovery and innovation is to transition from a traditional materials lab to the lab of the future. When successfully implemented, the lab of the future has an infrastructure of purpose-built technology with optimized workflows. Materials scientists and chemists are empowered with digital skills that enable them to make discoveries faster and more efficiently than ever before. To many R&D leaders, however, a digital automated lab remains an out-of-reach, abstract idea. While it’s clear that advanced technologies are essential in scientific discovery today, they often don’t know where to start or struggle with translating the unique challenges of the research lab to company executives and IT stakeholders.
Below are three common barriers preventing materials and chemical companies from fulfilling their lab of the future aspirations, along with what to consider to overcome them and get started.
Rigid, inflexible data management processes
Scientific discovery is a constant iterative cycle of exploration and testing, and as a result, requires tools that should be both structured and flexible as well as easily accessible to scientists. In materials research and formulation, in particular, a strong supporting data infrastructure can remarkably accelerate the materials value chain from discovery to deployment.
But product development teams often work with massive, non-uniform, and dispersed datasets that require special handling and are historically difficult to wrangle and centralize. Most materials labs rely on traditional data management tools and one-size-fits-all software which eventually prove to be woefully insufficient in managing the complexity of scientific data. Many commonly used tools are rigid, inflexible, and difficult for scientists to use for any other purpose than the original task.
To glean the most value from their data, R&D leaders need to look for data tools and solutions that are purpose-built for scientific discovery and the unique challenges materials scientists and chemists face. These tools need to be flexible enough to integrate into existing complex workflows and continuously evolve as needs change. The best R&D tech stack is streamlined and fits the way researchers think and work: domain-centric, agile, and extensible. For example, with a science-driven platform, researchers could access a custom-built experimental recommendation system that uses machine learning and continuously generated data to recommend the next experiment needed to create a material with distinct desired properties. Their valuable time is spent on discovery instead of repetitive tasks and burdensome workflows.
In response to the growing pressure to continuously innovate, many business leaders in materials and chemical companies turn to the newest shiny technology solution to solve all their innovation problems. This technology-first approach usually ends in failing to realize true value or transformation. What we most often see is that the adoption of the new technology has been ill-defined, with a lack of understanding of the organization’s unique opportunities and what its scientists need to advance their work.
This approach is typically unpredictable and can lead to costly errors when implemented without a clearly defined goal or intent. A very common scenario is when business leaders hire entirely new teams to build and implement the newest platforms to improve their R&D processes and shrink timelines, only to find that the process takes years to complete and is insufficient in supporting product development. Or they simply add a new technology solution on top of their existing infrastructure and processes, bandaging, but not eliminating, the bottlenecks and deeper problems.
Successful R&D transformations undoubtedly include investments in technology, such as materials informatics, artificial intelligence, and digital twins, but also in the people and processes around it. Shifting from a technology-first to a more holistic approach sets up the organization for success and changes what’s possible in the lab. When the constraints of the traditional lab are removed — when the purpose-built technology is in place, workflows are optimized and scientists have new capabilities — the purpose of the lab changes. It’s no longer the goal of just getting new products out the door. The purpose of the materials lab of the future is to enable continuous discovery.
Limited line of sight to business value
R&D reflects a significant investment by companies, particularly when factoring in the expensive equipment required to develop, test, and process new products. Additionally, the work environments in materials and chemical research are incredibly complex and specialized, with tasks such as testing and reporting performed manually or in siloed settings. While inefficient workflows and skills gaps are the perfect motivators for building the lab of the future, many stakeholders struggle to see a return on investment.
Lab leadership can increase line of sight to business value to garner support by detailing the valuable business outputs they aim to achieve with the transformed lab, and then identify a couple of smaller, measurable projects that balance short-term impact with long-term value creation. Once those smaller projects come to fruition, leadership will be more receptive to bigger and larger investments down the road. An external partner with R&D and computing expertise can be particularly impactful at this stage, to help teams identify which small-scale projects to tackle first to maximize ROI.
Consider a company that needs to develop new catalysts for carbon utilization. Given the long list of candidate materials, the researchers likely expend a lot of time and energy on the slow trial-and-error process of manually searching for potential candidates, compiling a shortlist, creating an experimental plan, and then executing. A partner with materials informatics expertise can both help automate the process of filtering potential candidates as well as upskill the team to leverage machine learning. As a result, the R&D organization hits performance targets way ahead of schedule while also empowering scientists with new skills. The return is visible and measurable.
Recent global events and market pressures have amplified the importance of transforming traditional materials R&D labs into digital labs of the future. Despite the challenges that may accompany the future-proofing process, organizations must prioritize it to compete today, and tomorrow. For additional motivation to get started, ask yourself: What could be accomplished if our scientists could spend 100% of their time advancing discoveries?
About the Author
Chris Farrow holds a Ph.D. in physics from Michigan State University and degrees in physics and mathematics from the University of Nebraska. Chris has spent 16 years working as a physicist in materials discovery and characterization. Chris currently leads Enthought’s Materials Science Solutions Group, where he oversees Digital Transformation solutions for the Specialty Chemicals and Semiconductor industries, as well as the development of novel technologies for materials data management and discovery. Chris is based at Enthought’s headquarters, in Austin, Texas.