Achieve rapid innovation with design of experiments

Watch the Panel and Case Study

Panel Discussion

How companies like Johnson Matthey and F-star Therapeutics take advantage of structured experimentation


How can you help your organization fail fast, remain agile and know when a concept will work – and when it won’t? By actively manipulating factors according to a prespecified plan, you can gain useful, new understanding of relationships among inputs and outputs. This is design of experiments, or DOE.

This panel discussion is about the power of using DOE to speed up innovation; achieve faster, more predictable cycles; and save time. The panelists talk about the successes and challenges they’ve experienced leading DOE initiatives, ideas for integrating DOE throughout an organization and how to overcome objections to the adoption of this methodology.

You’ll hear from:

  • Pilar Gomez Jimenez, Principal Scientist at global science and chemicals company Johnson Matthey, about her 15 years of experience working in research and development of catalysts and materials, and providing DOE training and support across her organization.
  • Jon Armer, Formulation Lead at F-star Therapeutics, on how using DOE has transformed the way he solves problems and why we should advocate for playing with data to understand it in context.
  • Phil Kay, JMP Learning Manager at SAS, with guidance for scientists and engineers whose technical challenges require DOE solutions and lessons from his time as a lab development chemist and data scientist in industry.

Preview the panel discussion

Case Study

The Pain of NOT Using Design of Experiments


JMP Learning Manager Phil Kay contrasts two examples of scientists on a quest to develop innovative processes and products. The first case shows what can go wrong when you don’t use smart experimentation strategies. Kay outlines a year in the life of a development chemist trying to develop a product for an unmet market need. Did he and his team meet the goal? Is theory, experience and intuition enough to guide them to an optimal formulation? Was standard trial and error an effective way to innovate under a tight timeline?

The second case shows how – in a matter of weeks – one small startup developed an efficient commercial process using design of experiments, subsequently selling its technology for $100 million. How did this company:

  • Narrow 35 factors to the 10 most important factors?
  • Test tens of thousands of possibilities and build solid process knowledge?
  • Communicate with investors, explaining how it discovered this viable process and advanced it to the next stage?