To compute probabilities regarding the sample mean using the normal model, we need to consider the Central Limit Theorem. Since the distribution of the time required to get an oil change is skewed right, the sample size must be greater than or equal to 30 for the normal approximation to be valid.
Thus, the answer is:
\[
\text{The sample size needs to be greater than or equal to } 30.
\]
We want to find the probability that a random sample of \( n = 35 \) oil changes results in a sample mean time less than 20 minutes. The calculated probability is given by:
\[
P(X < 20) = \Phi(Z_{end}) - \Phi(Z_{start}) = \Phi(-1.972) - \Phi(-\infty) = 0.0243.
\]
Thus, the probability that the sample mean is less than 20 minutes is:
\[
\boxed{P \approx 0.0243}.
\]
To find the mean oil-change time such that there is a 10% chance of being at or below this time, we first calculate the mean and standard deviation of the sampling distribution:
Next, we find the z-score corresponding to the 10th percentile:
\[
z = \frac{0.1 - 0}{1} = 0.1.
\]
Finally, we calculate the mean oil-change time for a 10% chance:
\[
X = \mu + z \cdot \sigma = 21.3 + 0.1 \cdot 0.6592 \approx 21.4.
\]
Thus, there is a 10% chance of being at or below a mean oil-change time of:
\[
\boxed{21.4 \text{ minutes}}.
\]
- The sample size needs to be greater than or equal to 30.
- The probability that the sample mean is less than 20 minutes is approximately \( \boxed{0.0243} \).
- There is a 10% chance of being at or below a mean oil-change time of \( \boxed{21.4 \text{ minutes}} \).