Based on the previous analysis, your team decides to evalute capacitor performance under the previous

process, so you used the $41$ measurements from the variation study to estimate the worst case

percent defective $($out of tolerance$)$ under that process. As before, the upper and lower specification limits

for the capacitor values are $0\mathrm{.}400$ mF$\text{'}$s and $0\mathrm{.}650$ mF$\text{'}$s$.$ Assume the process is still normally distributed.

Would you recommend making the change back to the previous capacitor process? Assume that nothing

else will be negatively affected.