Julia O’Neill bridges statistics and chemical engineering
Oct 3, 2018 12:12 PM
| Last Modified: Oct 8, 2018 7:42 AM
Julia O’Neill, Principal at Tunnell Consulting, says one of the most common problems that she helps clients overcome is the curse of dimensionality.Currently Principal at Tunnell Consulting with more than 30 years of experience using statistics and chemical engineering, Julia O’Neill uses statistical thinking to solve problems in biotech and pharma. Julia’s career path was influenced by the famous statisticians and authors Bill Hunter and George Box.
She started out in chemical engineering but broadened her scope to include statistics initially through Bill Hunter’s class, Design of Experiments for Engineers, at the University of Wisconsin. In that class, she had a “lightbulb moment” when she thought, “This is always what I wanted to do but didn’t even know what they called it.”
George Box used the term “bridging” when statisticians could help apply statistics in other fields. Julia has made great use of her niche bridging statistics and chemical engineering. I was so pleased to spend some time with her and chat with her on the most recent episode of Analytically Speaking.
One of the most common problems that Julia helps clients overcome now is the curse of dimensionality. She is a fan of the ways JMP helps overcome this. In presenting results to stakeholders, she says the Screening platform in JMP can be an “incredible consensus builder.” Some of the scientists and engineers she consults with have ideas about what causes what, but their hypotheses usually aren’t all right. The Predictor Screening chart in JMP shows an ordered ranking, which Julia says lets them see all of their hypotheses, but in priority order. This helps them determine which factors are important; if their hypotheses are near the bottom of the ranking, it’s time to rethink things.
Julia also shared some of the important work she oversaw investigating complex issues in vaccine manufacturing while at Merck. Her efforts yielded tighter control in manufacturing consistency and improved process robustness. This work led to similar statistical thinking to set up continued process verification for a cancer treatment that Merck was rolling out. It's clear she has a passion for helping pharmaceutical companies get critical new treatments to market quickly, especially for patients with so few options.
Some statisticians at the Food and Drug Administration (FDA) invited Julia to give a course on applications of statistics in chemistry, manufacturing, and controls (CMC). As an advocate for greater use of statistics throughout development and manufacturing, teaching this course has inspired her to write a book to reach more scientists, engineers, and statisticians who want to get more information on applying statistics within CMC. I look forward to spreading the word about it!
Julia talks about the power of design of experiments, quality by design, Six Sigma efforts, and more in our conversation. She has so many fascinating stories to share. Watch my interview in this latest episode, and you'll benefit from her extensive experience and expertise bridging statistics and chemical engineering.