 |
Before simulation can be applied to the Littletown bank problem, more
information needs to be obtained about the anticipated pattern of interarrival
times (the time between consecutive arrivals of customers into the
bank) and of service times (the time required by a teller to serve
a customer). Specifically, an estimate needs to be made of the probability
distribution for each of these two kinds of times. |
Remember that the minimum interarrival time is 0.5 minute (because
the parking validation process takes at least 0.5 minute). The amount by
which the interarrival time exceeds 0.5 minute is estimated to have an
exponential
distribution with a mean of 0.5 minute. Therefore, the total interarrival
time has a translated exponential distribution with a mean of (0.5
+ 0.5) = 1.0 minute. (A translated exponential distribution is just an
exponential distribution with a constant added.) |
Experience in other branch offices indicates that service time has
approximately an Erlang distribution with a mean of 1.5 minutes
and a shape parameter of
(
is required to be a positive integer), so that
|
 |