“My initial reading on the term ‘Bioprocessing 4.0’ left me with the feeling that it was mostly a group of related buzzwords with no clear structure or demonstrated pathway toward broader impact,” says David Wood, PhD, of the department of chemical & biomolecular engineering at Ohio State University. As an academic, Wood is faced with the stark reality that only novel and clearly impactful ideas get funded, and “publish or perish” is an everyday part of life.
“I find myself wondering if a proposal based on Bioprocessing 4.0, as it is currently described in the literature, would get funded,” Wood argues, “and the answer is almost certainly no.” Put simply, he does not yet see any truly transformative impacts arising from the Bioprocessing 4.0 concept; it does not solve problems that cannot already be solved, and it is not clear that it can deliver enabling advances in scientific understanding.
While working at Amgen in the 1990s, Wood and his team encountered a serious manufacturing issue where process yields were inexplicably dropping over the course of several months. “We did all of the usual things to identify the problem, and eventually made an entirely new Master Cell Bank; none of which helped.” Wood then recounted a complex series of scientific investigations, involving statistical data mining, sample analysis, controlled experiments and some library work, which eventually linked the yield losses to an unexpected relationship between cellular productivity and the time between two key process steps. They ultimately determined that the random decision of the operator on when to have lunch had an enormous impact on the batch yield, even though the operator never deviated from the standard operating procedure (SOP).
“I cannot imagine a computational model identifying this problem,” Wood claims, “simply because the computer would not have known what data to seek and what questions to ask. Yet we were able to solve it using technology and analytical methods that are now nearly 30 years old.”
So is Bioprocessing 4.0 is a true paradigm shift with an inevitable transformative impact, or is it simply a repackaging of existing methods with some increased automation and analytics? Wood suggests that the answer is not yet clear, and that perhaps our faith in vaguely defined technological breakthroughs to solve all of our problems may be overly optimistic.
“Automation is a wonderful thing,” says Wood, “but it cannot take the place of a competent and engaged scientist who sees the bigger picture and truly understands the problem they are trying to solve.”
