Answer:
To most closely estimate the difference, I would round 62,980 to the nearest thousand, so it will be 63,000. That is because when rounding, 63,000 is large enough so that adding will be easy, and 62,980 is relatively close to 63,000.
I would round 49,625 to the nearest thousand, since it would make the subtraction easier and 49,625 is only 375 units away from 50,000.
63,000 - 50,000 = about 13,000.
Hope this helps!
Since the standard deviation is not given, we assume the process is a Poisson process, with mean 12000 miles.
From Poisson distribution tables, we see that with a value of lambda(mean)=12000, the lower tail probability for x=12140 miles is 0.9000105, which means that the 90 percentile is 12140 miles.
The lower-tail probability of 9000 miles is 6.2*10^(-181) which is essentially zero.
NOTE: if the problem were modelled with a normal distribution, the probabilities would be completely different.