A paired design-of-experiment reduces treadwear test cost by reducing the time, vehicles and tires needed to detect statistically significant treadwear changes.
Traditional treadwear test design uses one compound per tire and is expensive due to the number of tires and vehicles needed to improve the signal to noise ratio and make valid comparisons in the face of the high number of uncontrolled test variables. The analysis used for multiple mean comparisons is less powerful than the paired, one-sample mean analysis.