The Value of Optimal Design of Experiments (DOE)

“I need you to produce more innovation per unit time.” This is a real quote from an industry CEO to his engineers, and a direct statement of the business problem. We need to go to market faster. Given a fixed quantity of cells that we can build and test, we can learn more with an optimal Design of Experiments than with several smaller A/B tests. We can make discoveries faster and go to market sooner with optimal DOE.

There are a few challenges to implementing optimal DOE in battery R&D. Optimal DOE requires that we test several variables simultaneously. However, it isn’t always obvious how to fit the cell design factors of interest into discrete control variables. It can also be cumbersome to run the statistical analysis needed to understand the results. In A/B tests, the results analysis is simpler, and so this is often the first choice for stressed engineers.

The Micantis platform integrates optimal design of experiments, cell builds specification, test data tracking, and results analysis. The seamless integration makes optimal DOE just as simple and straightforward as A/B testing. It also allows engineers to quickly analyze progress and results while tests are underway, so no time is wasted waiting for a protocol to complete if a test outcome is already clear.