Queuing distribution for scenario

Question

May I know what is the best queuing distribution to describe the following scenario? Is it M/M/2, M/G/2, or other distributions?

Ships arrive at a harbour with inter-arrival times that are IID exponential random variables with a mean of 1.25 days. The harbour has a dock with two berths, each equipped with a crane for unloading the ships. Ships arriving when both berths are free will select a berth at random. Ships arriving when both berths are occupied join a FIFO queue. The time for one crane to unload a ship is distributed uniformly between 0.5 and 1.5 days.

Thanks in advance!

Answer 1

In the usual interpretation of 'Kendall's notation', M = Markov or Memoryless stands for exponential, and G = General stands for non-exponential. Here arrivals are exponential and service times are uniform. So M/G/2.

See this Wikipedia page for more.