A unique feature of the exponential distribution is that the standard deviation of its service time is equal to the mean of its service time.

(a) Run the M/M/1 model with an arrival rate of 20 customers per hour and a service rate of 25 customers per hour. What is the average waiting time in the queue (in minutes)?

(b) Now consider the M/G/1 model with the same parameters as in part (a). Assume when running that model that the service time is not exponentially distributed, and its standard deviation is .03 hours (= 1.8 minutes per customer). Before running the model, do you expect the waiting time to be greater than or less than that in part (a)? Run the model—what is the average waiting time in the queue (in minutes)? Did your answer confirm your intuition?